Browse the corpus

1. Introduction Treatment for childhood cancer has become increasingly successful. Approximately 70% of patients with childhood cancer survive for 5 years, and many long-term survivors are now reaching adulthood [1]. Neuroblastoma is a childhood cancer that originates in the adrenal glands in more than 90% of cases. Pelvic neuroblastoma is thought to have a good prognosis, but women who have previously been treated with curative surgery, chemotherapy, or radiation might have long-term effects on their reproductive organs. There are few reports of pregnancies in women who have previously been treated for neuroblastoma. Here, we report a case of a pregnant patient who had been successfully treated for pelvic neuroblastoma. 2. Case Report The patient was a 30-year-old Japanese woman. This was her first pregnancy. She had a pelvic neuroblastoma in the front of her sacrum at 1 month of age. She had undergone surgical resection, and the pathology revealed rosette-forming neuroblastoma. She received adjuvant chemotherapy with vincristine and cyclophosphamide (the James method) postoperatively up to the age of 2 years. She also received pelvic irradiation with a total dose of 18 Gy. By age of 3 years, there was no evidence of disease, and she was considered cured.

evealed rosette-forming neuroblastoma. She received adjuvant chemotherapy with vincristine and cyclophosphamide (the James method) postoperatively up to the age of 2 years. She also received pelvic irradiation with a total dose of 18 Gy. By age of 3 years, there was no evidence of disease, and she was considered cured. She visited out hospital for prenatal care at 6 weeks of gestation. At 15 weeks and 5 days of gestation, she complained of abdominal pain due to frequent uterine contractions and was diagnosed with threatened abortion or incompetent cervix by transvaginal ultrasound. She was admitted at 16 weeks of gestation and ritodrine tocolytic therapy was initiated to control her frequent uterine contractions. At 20 weeks and 2 days of gestation, she developed acute abdominal pain. Ultrasound and magnetic resonance imaging (MRI) revealed intra-abdominal bleeding of unknown etiology (Figure 1). The patient and her husband chose to continue the pregnancy even after we explained to them the risks involved in doing so and after we suggested termination of pregnancy. She received 4 units of red blood cells but did not require any surgical intervention. At 23 weeks of gestation, she prematurely delivered a 510 g female infant due to sudden onset of labor pain. There were no signs of chorioamnionitis. The infant was admitted to the NICU and was discharged in good condition 110 days after birth.

d 4 units of red blood cells but did not require any surgical intervention. At 23 weeks of gestation, she prematurely delivered a 510 g female infant due to sudden onset of labor pain. There were no signs of chorioamnionitis. The infant was admitted to the NICU and was discharged in good condition 110 days after birth. The patient presented with massive vaginal bleeding 14 days postpartum. Medical and surgical therapy was performed. Uterine curettage revealed a retained placenta. Despite these treatments, the patient experienced continuous slow vaginal bleeding. However, because her bleeding was light, she waited to see if the bleeding would stop naturally. An MRI was conducted at 46 days postpartum, and the scan revealed a placental polyp (Figure 2). Heavy vaginal bleeding occurred at 49 days postpartum. Uterine artery embolization (UAE) was considered, but the patient was noted to shed the placental polyp spontaneously with full recovery. 3. Discussion Long-term female survivors of childhood pelvic neuroblastoma might develop complications during pregnancy due to the negative effects of prior radiation therapy. Our patient received postoperative chemotherapy with vincristine and cyclophosphamide without apparent effects on her pregnancy. Green et al. [2] reported that prior treatment with specified chemotherapeutic agents does not affect subsequent pregnancy; this finding is also supported by other studies.

radiation therapy. Our patient received postoperative chemotherapy with vincristine and cyclophosphamide without apparent effects on her pregnancy. Green et al. [2] reported that prior treatment with specified chemotherapeutic agents does not affect subsequent pregnancy; this finding is also supported by other studies. Critchley and Wallace [3] reported that radiation therapy could have negative effects on the uterus. Specifically, they reported that radiation doses between 14 and 30 Gy resulted in the impairment of uterine musculature and vasculature and that patient age at the time of irradiation determined the final uterine volume. Our patient delivered at 23 weeks of gestation. Most preterm labor is due to chorioamnionitis; but no evidence of infection was found in our patient. It is possible that the prior radiation to her pelvis resulted in an underdeveloped uterus with subsequent preterm labor and delivery [3, 4]. In this case, because we did not measure her uterine volume, we have no evidence to prove that her uterus was underdeveloped. However, Critchley and Wallace [3] reported that a patient treated with radiation prior to puberty had a small uterus. Therefore, we speculated that our patient's uterus was underdeveloped.

4]. In this case, because we did not measure her uterine volume, we have no evidence to prove that her uterus was underdeveloped. However, Critchley and Wallace [3] reported that a patient treated with radiation prior to puberty had a small uterus. Therefore, we speculated that our patient's uterus was underdeveloped. It is possible that the patient's intra-abdominal bleeding might have been a negative consequence of prior radiation. Our patient developed sudden intra-abdominal bleeding at 20 weeks of gestation. The cause is unclear because we did not perform any surgical intervention. The surgery and pelvic irradiation that she underwent in her childhood caused adhesion in her abdomen, and this adhesion may have torn as the uterus extended during pregnancy, resulting in intra-abdominal bleeding at 20 weeks of gestation. In this case, pregnancy termination might have to be considered to avoid the associated risk of continuing the pregnancy. We decided to continue her pregnancy for 2 reasons: (i) the patient was determined to give birth to a living infant, and (ii) this pregnancy might have been her last chance to deliver a baby because this atypical complication—intra-abdominal bleeding during her pregnancy—was an effect of pelvic irradiation in her childhood.

cided to continue her pregnancy for 2 reasons: (i) the patient was determined to give birth to a living infant, and (ii) this pregnancy might have been her last chance to deliver a baby because this atypical complication—intra-abdominal bleeding during her pregnancy—was an effect of pelvic irradiation in her childhood. Uterine irradiation might predispose a woman to abnormal placentation in a subsequent pregnancy. Norwitz et al. [5] reported a patient with a history of abdominal radiation who experienced placenta percreta and uterine rupture at 17 weeks of gestation. Our patient suffered from placental polyps at almost 2 months postpartum. This finding is also consistent with abnormal placentation and is a likely result of her childhood radiation therapy. In summary, pregnancy outcomes may be affected in long-term survivors of childhood cancers who have received pelvic radiation. Obtaining a detailed history is essential to manage the pregnancies of such patients. Patients with a history of prior pelvic radiation might experience decreased uterine volume and increased risks of miscarriage, preterm delivery, or abnormal placentation and should receive appropriate high-risk prenatal care. Figure 1 MRI at 21 weeks of gestation. Arrows show location of intra-abdominal bleeding. Figure 2 MRI at 47 days after delivery. Arrows point to placental polyp.

1. Introduction Humoral hypercalcemia of malignancy (HHM) is a rare but potentially dangerous complication of pediatric cancer. The incidence of pediatric malignancy-associated hypercalcemia is between 0.4 and 1.3% [1]. Ovarian dysgerminomas, which comprise two thirds of malignant ovarian neoplasms in children, are one of the most common ovarian cancers associated with hypercalcemia. This paraneoplastic disorder is usually related to tumor cell synthesis of parathyroid hormone-related peptide (PTHrP) or 1,25-dihydroxyvitamin D (1,25(OH)2D3) and typically resolves within one week of tumor resection [2–4]. We report a case of hypercalcemia associated with a dysgerminoma in which the etiology of hypercalcemia was not determined, in spite of extensive evaluation. This case is remarkable due to the prolonged duration of hypercalcemia following complete surgical resection of the tumor. 2. Case A 13-year-old female presented with a two-week history of anorexia, fatigue, fever, emesis, abdominal pain, and polyuria. There was no family history of endocrine neoplasia or parathyroid disease. Physical exam was notable for a large, firm mass extending from the pelvis to the left costal margin. CT scan and ultrasound revealed a left ovarian mass measuring 13 × 6.6 × 14 cm without evidence of metastases. Initial blood work revealed an elevated total calcium 14.9 mg/dL (9–10.6), normal alkaline phosphatase and albumin, as well as mildly elevated pancreatic enzymes. With the exception of an elevated LDH, tumor markers associated with ovarian cancer were normal.

2. Case A 13-year-old female presented with a two-week history of anorexia, fatigue, fever, emesis, abdominal pain, and polyuria. There was no family history of endocrine neoplasia or parathyroid disease. Physical exam was notable for a large, firm mass extending from the pelvis to the left costal margin. CT scan and ultrasound revealed a left ovarian mass measuring 13 × 6.6 × 14 cm without evidence of metastases. Initial blood work revealed an elevated total calcium 14.9 mg/dL (9–10.6), normal alkaline phosphatase and albumin, as well as mildly elevated pancreatic enzymes. With the exception of an elevated LDH, tumor markers associated with ovarian cancer were normal. To correct the hypercalcemia (and associated pancreatitis), aggressive fluids and furosemide were administered. After several hours, total calcium decreased to 13.4 mg/dL (ionized 1.58 mmol/L). A dysgerminoma was suspected and a left salpingo-oophorectomy was performed. As lymph nodes were clinically suspicious, left pelvic and aortic lymph nodes were also resected. Pathologic evaluation confirmed a left ovarian dysgerminoma with an intact capsule and negative nodes and washings. Postoperatively, IV fluids were continued with appropriate diuresis. Postoperative day (POD) 1, total calcium was 11.1 mg/dL (ionized 1.56 mmol/L). Unexpectedly, the patient's calcium began to rise (Figure 1(a)) and on POD 4, total calcium was 14.3 mg/dL (ionized 1.8 mmol/L). Calcium only stabilized (level <12 mg/dL) after furosemide was initiated.

e continued with appropriate diuresis. Postoperative day (POD) 1, total calcium was 11.1 mg/dL (ionized 1.56 mmol/L). Unexpectedly, the patient's calcium began to rise (Figure 1(a)) and on POD 4, total calcium was 14.3 mg/dL (ionized 1.8 mmol/L). Calcium only stabilized (level <12 mg/dL) after furosemide was initiated. The persistence of hypercalcemia led to further investigation. Bone scan, SPECT, and PET imaging did not show evidence of metastases. Immunolocalization studies indicated that 1 α-hydroxylase was focally expressed in tumor cell cytoplasm (Figure 1(b)). Serum PTH and PTHrP were undetectable, Vitamin 1,25-dihyroxy was 19 pg/mL (15–75), vitamin D 25-hydroxy was 23 ng/mL (30–80), and urine phosphorus and calcium were low and high, respectively. On POD 6, serum calcium had decreased to near the upper range of normal. Fuorsemide and IV fluids were discontinued and the patient was discharged. Outpatient monitoring revealed stable calcium levels over time (Figure 1(a)).

ydroxy was 23 ng/mL (30–80), and urine phosphorus and calcium were low and high, respectively. On POD 6, serum calcium had decreased to near the upper range of normal. Fuorsemide and IV fluids were discontinued and the patient was discharged. Outpatient monitoring revealed stable calcium levels over time (Figure 1(a)). 3. Conclusion Tumor associated hypercalcemia, not resulting from bone marrow invasion, is termed HHM, and is caused by factors secreted by tumors into the circulation. In most described cases, the etiologic factor identified is PTHrP, a compound similar to PTH, which causes increased bone resorption and renal calcium reabsorption. Although clinically PTHrP-induced hypercalcemia is similar to primary hyperparathyroidism, 1,25(OH)2D3 levels are low rather than high [2]. In the present case, low 1,25(OH)2D3 supports a PTHrP-related mechanism of hypercalcemia, however, undetectable PTHrP, low urine phosphorus and high urine calcium make this mechanism unlikely.

nically PTHrP-induced hypercalcemia is similar to primary hyperparathyroidism, 1,25(OH)2D3 levels are low rather than high [2]. In the present case, low 1,25(OH)2D3 supports a PTHrP-related mechanism of hypercalcemia, however, undetectable PTHrP, low urine phosphorus and high urine calcium make this mechanism unlikely. 1,25(OH)2D3 increases intestinal calcium absorption and, along with PTH, increases calcium resorption from bone and has been identified as another cause of HHM. The putative mechanism is increased expression of 1 α-hydroxylase in tumor cells and macrophages, an enzyme that catalyzes the synthesis of 1,25(OH)2D3 [2]. In this case, immunostaining of dysgerminoma tissue for 1 α-hydroxylase was positive. Although tumor expression of 1 α-hydroxylase is a plausible explanation, 1,25(OH)2D3 obtained on POD 1 was in the lower limits of normal making it unlikely that increased 1 α-hydroxylase resulted in persistent hypercalcemia. The half-life of 1,25(OH)2D3 is short, in the range 4–6 hours, but it is theoretically possible that intracellular effects persisted after serum levels of this vitamin D metabolite fell. Besides PTHrP and Vitamin D, production of prostaglandin or cytokines by tumor or immune cells may also play a role in HHM. These factors can increase bone resorption by inducing osteoclasts. We cannot exclude one of these factors or an unknown mediator contributing to the metabolic abnormality in this case.

1,25(OH)2D3 increases intestinal calcium absorption and, along with PTH, increases calcium resorption from bone and has been identified as another cause of HHM. The putative mechanism is increased expression of 1 α-hydroxylase in tumor cells and macrophages, an enzyme that catalyzes the synthesis of 1,25(OH)2D3 [2]. In this case, immunostaining of dysgerminoma tissue for 1 α-hydroxylase was positive. Although tumor expression of 1 α-hydroxylase is a plausible explanation, 1,25(OH)2D3 obtained on POD 1 was in the lower limits of normal making it unlikely that increased 1 α-hydroxylase resulted in persistent hypercalcemia. The half-life of 1,25(OH)2D3 is short, in the range 4–6 hours, but it is theoretically possible that intracellular effects persisted after serum levels of this vitamin D metabolite fell. Besides PTHrP and Vitamin D, production of prostaglandin or cytokines by tumor or immune cells may also play a role in HHM. These factors can increase bone resorption by inducing osteoclasts. We cannot exclude one of these factors or an unknown mediator contributing to the metabolic abnormality in this case. It is also conceivable, but unlikely, hypercalcemia was unrelated to the dysgerminoma. Coexisting conditions excluded were hypervitaminosis A and D, hyperparathyroidism, sarcoidosis, adrenal insufficiency, and hyperthyroidism.

Besides PTHrP and Vitamin D, production of prostaglandin or cytokines by tumor or immune cells may also play a role in HHM. These factors can increase bone resorption by inducing osteoclasts. We cannot exclude one of these factors or an unknown mediator contributing to the metabolic abnormality in this case. It is also conceivable, but unlikely, hypercalcemia was unrelated to the dysgerminoma. Coexisting conditions excluded were hypervitaminosis A and D, hyperparathyroidism, sarcoidosis, adrenal insufficiency, and hyperthyroidism. This case is unique as moderate hypercalcemia persisted 7 days following tumor resection and mild hypercalcemia persisted for an additional 7 days. In most reported cases of dysgerminoma associated hypercalcemia, calcium corrected within 2–7 days of tumor debulking [4]. Physicians should be aware of the association between hypercalcemia and pediatric cancer. Incomplete resolution of hypercalcemia for over a week following tumor resection suggests a previously undescribed mechanism. Elucidation of this mechanism may offer new insights into tumor biology and opportunities for novel therapeutics in this patient population. Acknowledgments The authors would like to thank Martin Hewison for performing the immunohistochemical staining and Beverly Dahms for her assistance with the tissue samples and helpful explanations. Figure 1 (a) Trend of ionized calcium after tumor resection. (b) Immunostaining for 1 α-hydroxylase was focally positive in tumor cell cytoplasm.

1. Introduction Endometrial cancer is the most frequently occurring gynecologic malignancy. It accounts for 6% of all cancers in women, and causes approximately 42,000 deaths annually, which represents 3% of cancer deaths in women in the United States [1]. Most of the cancers are detected at an early stage by common symptom such as postmenopausal bleeding, with the tumor confined to the uterine corpus, so the prognosis is generally favorable and surgery alone may result in a cure. The five-year survival rate for stage IA or IB disease is reported to be over 90% [2, 3].

. Most of the cancers are detected at an early stage by common symptom such as postmenopausal bleeding, with the tumor confined to the uterine corpus, so the prognosis is generally favorable and surgery alone may result in a cure. The five-year survival rate for stage IA or IB disease is reported to be over 90% [2, 3]. The International Federation of Gynecology and Obstetrics (FIGO) recommended in 1988 that adequate surgical staging requires a total abdominal hysterectomy, bilateral salpingo-oophorectomy (TAH-BSO) including pelvic and paraaortic lymphadenectomy [4], and according to this recommendation, some surgeons believe that lymphadenectomy should be performed in all cases to enable the accurate staging and to assess the necessity for postoperative treatment. However, there are some risks to lymphadenectomy such as postoperative deep vein thrombosis or leg lymphedema which may impair the patients' quality of life. Mariani et al. reported that low-risk corpus cancer (endometrioid type, grade 1 or 2 tumor, myometrial invasion ≦50%, and no intraoperative evidence of macroscopic extrauterine spread) could be treated optimally with hysterectomy only [5]. We retrospectively reviewed the cases of low-risk corpus cancer, which were treated in our hospital, and clarified that lymphadenectomy did not provide a significant survival advantage, and increased peri- and postoperative morbidities and complications [6]. According to these results, since 1998, lymphadenectomy have been omitted in low-risk corpus cancer in our hospital. We retrospectively reviewed these cases and evaluated whether omission of lymphadenectomy for low-risk corpus cancer worsen the disease-free survival (DFS), overall survival (OS), and avoid peri- and postoperative morbidities and complications.

ymphadenectomy have been omitted in low-risk corpus cancer in our hospital. We retrospectively reviewed these cases and evaluated whether omission of lymphadenectomy for low-risk corpus cancer worsen the disease-free survival (DFS), overall survival (OS), and avoid peri- and postoperative morbidities and complications. 2. Patients and Methods Eighty-three patients (median age: 55 years, range: 27–80 years) with endometrioid corpus cancer, grade 1 or 2 tumor, myometrial invasion ≦50%, and no intraoperative evidence of macroscopic extrauterine spread, including pelvic and paraaortic lymph node swelling and adnexal metastasis, were treated surgically at the Department of Obstetrics and Gynecology of University of Toyama during the period 1998 to 2007. In all these cases, we preoperatively assessed whether endometrial cancer is considered low-risk (myometrial invasion ≦50%, no lymphadenopathy and grade 1 or 2 endometrioid corpus cancer), using computed tomography, magnetic resonance imaging, glucose analog [18F]-fluoro-2-deoxy-D-glucose positron emission tomography and endometrial biopsy.

preoperatively assessed whether endometrial cancer is considered low-risk (myometrial invasion ≦50%, no lymphadenopathy and grade 1 or 2 endometrioid corpus cancer), using computed tomography, magnetic resonance imaging, glucose analog [18F]-fluoro-2-deoxy-D-glucose positron emission tomography and endometrial biopsy. All patients routinely underwent TAH-BSO without lymphadenectomy. If the depth of myometrial invasion was determined to be >50%, or the tumor was classified grade 3 endometrioid corpus cancer based on intraoperative frozen-section analysis, we performed systemic lymphadenectomy. In cases which lymph nodes were enlarged or suspicious, we performed selective lymph node sampling. Non-endometrioid histologic types such as serous or clear cell type and grade 3 tumor were excluded from this study, since all of these patients underwent TAH-BSO with lymphadenectomy because of their poor prognosis. After operation, patients were seen every month for one year, and every 3 months thereafter for 120 months.

endometrioid histologic types such as serous or clear cell type and grade 3 tumor were excluded from this study, since all of these patients underwent TAH-BSO with lymphadenectomy because of their poor prognosis. After operation, patients were seen every month for one year, and every 3 months thereafter for 120 months. We performed a retrospective review of the medical records, and the disease-free survival (DFS; the interval between the date of operation and the date of recurrence of disease), overall survival (OS; the interval between the date of operation and the date of death), and peri-operative morbidities including operative time, estimated blood loss during operation, and the percentage of patients requiring transfusion were evaluated. We also estimated the incidence of postoperative complications such as leg lymphedema and deep vein thrombosis diagnosed by nuclear venography. The DFS and OS curves were estimated using the Kaplan-Meier method.

estimated blood loss during operation, and the percentage of patients requiring transfusion were evaluated. We also estimated the incidence of postoperative complications such as leg lymphedema and deep vein thrombosis diagnosed by nuclear venography. The DFS and OS curves were estimated using the Kaplan-Meier method. 3. Results Patients' characteristics are shown in Table 1. The median age of the patients was 55 years (range, 27–80 years). The distribution of FIGO surgical stage was Ia, 32; Ib, 43; Ic, 3, IIIa, 5; and 8 cases (9.6%) were upstaged postoperatively. Pelvic lymph node sampling was performed in 7 cases, which were diagnosed with negative nodes. Adjuvant chemotherapy consisting of intravenous paclitaxel (180 mg/m2) and carboplatin (AUC 5) was administered to 9 upstaged, upgraded or lymphvascular space involvement-positive cases. No patients received adjuvant radiotherapy. The median followup period was 72 months (range, 4–120 months). Characteristics of the histopathological prognostic features are shown in Table 2. The distribution of histological grade was grade 1, 72; grade 2, 10; grade 3, 1; and 1 case (1.2%) was upgraded postoperatively. Depth of myometrial invasion was >50% in 3 cases. Lymphvascular space involvement was observed in 5 cases. Positive peritoneal cytology was observed in 5 case. Thirty-nine patients had tumor diameter ≧20 mm. The DFS curve and the OS curve are shown in Figure 1 (left; the DFS curve, right: the OS curve). The 5-year DFS rates and the 5-year OS rates were 97.6% and 98.8%, respectively.

Characteristics of the histopathological prognostic features are shown in Table 2. The distribution of histological grade was grade 1, 72; grade 2, 10; grade 3, 1; and 1 case (1.2%) was upgraded postoperatively. Depth of myometrial invasion was >50% in 3 cases. Lymphvascular space involvement was observed in 5 cases. Positive peritoneal cytology was observed in 5 case. Thirty-nine patients had tumor diameter ≧20 mm. The DFS curve and the OS curve are shown in Figure 1 (left; the DFS curve, right: the OS curve). The 5-year DFS rates and the 5-year OS rates were 97.6% and 98.8%, respectively. Peri- and postoperative morbidities and complications are shown in Table 3. The mean operative time was 129 ± 28 minutes. The mean estimated blood loss during operation was 244 ± 192 mL, and the percentage of transfusion requirement was 2.4%. No patient presented with postoperative leg lymphedema or postoperative deep venous thrombosis.

ative morbidities and complications are shown in Table 3. The mean operative time was 129 ± 28 minutes. The mean estimated blood loss during operation was 244 ± 192 mL, and the percentage of transfusion requirement was 2.4%. No patient presented with postoperative leg lymphedema or postoperative deep venous thrombosis. 4. Discussion In 1988, FIGO recommended a systemic surgical staging system for corpus cancer [4]. According to this recommendation, many gynecologic oncologists believe that systemic surgical staging including pelvic and paraaortic lymphadenectomy is the best surgical treatment to achieve a good prognosis for corpus cancer patients. However, controversy has persisited regarding the need for lymphadenectomy [5–9]. There is no doubt that surgical staging is more accurate than clinical staging, and there are some cases which need upstaging to higher stages after surgery [10–12]. However, Morrow et al. reported that only 18 (2%) out of 895 patients had positive pelvic lymph nodes in the abscence of operative findings by palpation [13]. Creasman et al. demonstrated by multivariate analysis that in clinical stage I patients with grade 1 or 2 and depth of invasion within the middle 1/3, the incidence of pelvic lymph node metastases was only 3.6% [7]. Chi et al. showed that the incidence of pelvic lymph node metastases in low-risk corpus cancer (grade; 1, 2, and depth of myometrial invasion; none or inner half) was 5.3% (14/162) [14]. As for grading, Ben-Shachar et al. reported that only 6 (3.3%) of 181 patients preoperatively diagnosed with grade 1 disease by biopsy were upgraded after surgical staging [15]. In this study, 8 of the 83 were upstaged or upgraded after surgical staging.

myometrial invasion; none or inner half) was 5.3% (14/162) [14]. As for grading, Ben-Shachar et al. reported that only 6 (3.3%) of 181 patients preoperatively diagnosed with grade 1 disease by biopsy were upgraded after surgical staging [15]. In this study, 8 of the 83 were upstaged or upgraded after surgical staging. Regarding the role of lymphadenectomy on prognosis, Mariani et al. reported that patients who had grade 1 or 2 endometrioid corpus cancer with greatest surface dimension ≦2 cm, myometrial invasion ≦50%, and no intraoperative evidence of macroscopic disease could be treated optimally with hysterectomy only with favorable prognosis of up to 97% for 5-year overall cancer-related survival [5]. Furthermore, Trimble et al. demonstrated that the 5-year relative survival for 6,363 women with stage I endometrial cancer who did not undergo lymph node sampling was 98%, compared to 96% for 2,831 women who did undergo lymph node sampling at the time of hysterectomy with a no significant difference, and concluded that lymph node sampling did not appear to convey survival benefit, especially in stage I, grade 1 or 2, endometrial cancer by subgroup analysis [9] Recently, Kitchener et al. compared the standard surgery group (hysterectomy and bilateral salpingo-oophorectomy, peritoneal washings, and palpation of para-aortic nodes; n = 704) and the lymphadenectomy group (standard surgery plus lymphadenectomy; n = 704) for histolgically proven endometrial carcinoma thought preoperatively to be confined to the corpus in a randomised study and showed that the hazard ratio (HR) for overall survival was 1.04 (0.74–1.45; P = .83) and HR for recurrence-free survival was 1.25 (0.93–1.66; P = .14), both in favour of standard surgery, and concluded that there was no evidence of benefit in terms of overall or recurrence-free survival for pelvic lymphadenectomy with early endometrial cancer, and that pelvic lymphadenectomy cannot be recommended as routine procedure for therapeutic purposes outside of clinical trials [16]. Panici et al. also, compared the pelvic systematic lymphadenectomy arm (n = 264) and no lymphadenectomy arm (n = 250) for preoperative FIGO stage I endometrial carcinoma and showed that the 5-year disease-free and overall survival rates were similar between the two arms (81.0% and 85.9% in the lymphadenectomy arm and 81.7% and 90.0% in the no-lymphadenectomy arm) [17].

lymphadenectomy arm (n = 264) and no lymphadenectomy arm (n = 250) for preoperative FIGO stage I endometrial carcinoma and showed that the 5-year disease-free and overall survival rates were similar between the two arms (81.0% and 85.9% in the lymphadenectomy arm and 81.7% and 90.0% in the no-lymphadenectomy arm) [17]. In summary, omission of complete lymphadenectomy is possible in selected cases in which the risk of lymph node spread is low, in other words, low-risk cancer. The definition of low-risk in corpus cancer at the operation is controversial; however, taking many reports into consideration, we regard grade 1 or 2 endometrioid corpus cancer with myometrial invasion ≦50%, and no intraoperative evidence of macroscopic disease as low-risk [5–9, 18, 19].

other words, low-risk cancer. The definition of low-risk in corpus cancer at the operation is controversial; however, taking many reports into consideration, we regard grade 1 or 2 endometrioid corpus cancer with myometrial invasion ≦50%, and no intraoperative evidence of macroscopic disease as low-risk [5–9, 18, 19]. There were 8 upstaged patients on final pathology who were thought low-risk on pre- and intraoperative evaluation. This result means that there is some limitation about the accuracy of pre- and intraoperative evaluation of myometrial invasion and tumor grade. Montalto et al. reported that accuracy of intraoperative frozen section diagnosis for grade of differentiation and depth of myometrial invasion were 84.3% and 94.3%, respectively [20]. For those upgraded, upstaged or lymphvascular space involvement-positive cases, as well as, non-endometrioid adenocarcinoma such as uterine papillary serous carcinoma, which is clinically aggressive, adjuvant treatment should be considered. Recently, paclitaxel has been shown to be effective against advanced and recurrent endometrial carcinoma [21–25]. Therefore, we added systemic chemotherapy (a paclitaxel/carboplatin regimen) as adjuvant treatment for upgraded, upstaged or lymphvascular space involvement-positive cases.

t treatment should be considered. Recently, paclitaxel has been shown to be effective against advanced and recurrent endometrial carcinoma [21–25]. Therefore, we added systemic chemotherapy (a paclitaxel/carboplatin regimen) as adjuvant treatment for upgraded, upstaged or lymphvascular space involvement-positive cases. Several investigators have reported that addition of lymphadenectomy to TAH-BSO increases the risk of complications and morbidities such as more blood transfusions, longer hospital stay, lymphedema, gastrointestinal injury, and the development of lymphocysts [26–28]. Framarino et al. reported that addition of pelvic and paraaortic lymphadenectomy to TAH-BSO significantly increased mean operative time, mean estimated blood loss, and postoperative hospital stay compared to TAH-BSO alone, without improving mortality [28]. Panici et al. showed that postoperative complications occurred statistically significantly more frequently in patients who had received pelvic systematic lymphadenectomy (81/264; 30.7% in the lymphadenectomy arm and 34/250; 13.6% in the no-lymphadenectomy arm, P = .001) [17]. Also in our previous study, mean operative time, mean estimated blood loss during operation, the percentage of cases requiring transfusion, and the incidence of leg lymphedema were significantly (P < .001) increased by addition of lymphadenectomy [6]. In any case, postoperative complications are expected from the surgical procedure itself, and addition of lymphadenctomy may increase the incidence of those complication, especially in corpus cancer patients, many of whom have morbidities such as hypertension, obesity, diabetes mellitus, and older age [29, 30].

my [6]. In any case, postoperative complications are expected from the surgical procedure itself, and addition of lymphadenctomy may increase the incidence of those complication, especially in corpus cancer patients, many of whom have morbidities such as hypertension, obesity, diabetes mellitus, and older age [29, 30]. 5. Conclusion A clinically important goal of surgical treatment including lymphadenectomy for low-risk corpus cancer patients is not only to determine the extent of disease and an accurate prognosis, but also to obtain a favorable prognosis without causing any complications. Our data demonstrate that omission of lymphadenectomy did not worsen the disease-free or overall survival, and as a result, peri- and postoperative morbidities and complications could be avoided. The present findings suggest that systemic lymphadenectomy could be omitted in low-risk endometrial carcinoma. These results should be confirmed in future prospective large-scale randomized clinical trials. Figure 1 Survival curves. Table 1 Patients characteristics (n = 83). Age (years) Mean ± SD 56.2 ± 12.1 Median 55 Range 27–80 WHO* performance status, no. 0 59 1 14 >2 0 FIGO** surgical stage, no. Ia 32 Ib 43 Ic 3 IIIa 5 Adjuvant chemotherapy, no. None 74 Paclitaxel/carboplatin§ 9 Follow up interval (months) Median (range) 72 (4–120) WHO*: World Health Organization, FIGO**: International Federation of Gynecology and Obstetrics Paclitaxel/carboplatin §: Paclitaxel (180 mg/m2) and carboplatin (area under the curve; AUC 5). Table 2 Characteristics of the histopathological prognostic features (n = 83).

Age (years) Mean ± SD 56.2 ± 12.1 Median 55 Range 27–80 WHO* performance status, no. 0 59 1 14 >2 0 FIGO** surgical stage, no. Ia 32 Ib 43 Ic 3 IIIa 5 Adjuvant chemotherapy, no. None 74 Paclitaxel/carboplatin§ 9 Follow up interval (months) Median (range) 72 (4–120) WHO*: World Health Organization, FIGO**: International Federation of Gynecology and Obstetrics Paclitaxel/carboplatin §: Paclitaxel (180 mg/m2) and carboplatin (area under the curve; AUC 5). Table 2 Characteristics of the histopathological prognostic features (n = 83). Histological grade, no. G1 72 G2 10 G3 1 Depth of myometrial invasion >50%, no. None 32 ≦50% 48 >50% 3 Lymphvascular space involvement, no. Yes 5 No 78 Peritoneal cytology, no. Positive 5 Negative 78 Tumor diameter, no. <20 mm 44 ≧20 mm 39 Table 3 Peri- and postoperative morbidities and complications. Peri- and postoperative factors Operative time* (min) 129 ± 28 Estimated blood loss during operation* (mL) 244 ± 192 Transfusion requirement, no. 2 Postoperative leg lymphedema (≧grade 2, NCI-CTC ver. 2.0), no. 0 Postoperative deep vein thrombosis, no. 0 *The values were expressed as the mean ± SD.

ssible to transcription factors [3]. Aberrant recruitment of HDAC activity has been associated with the development of certain human cancers [4]. HDAC inhibitors (HDACIs) can inhibit cancer cell growth in vitro and in vivo, revert oncogene-transformed cell morphology, induce apoptosis, and enhance cell differentiation. The classes of HDACIs that have been identified are: (a) organic hydroxamic acids (e.g., Trichostatin A (TSA) and suberoylanilide bishydroxamine (SAHA)) (b) short-chain fatty acids (e.g., butyrates and valproic acid (VPA)), (c) benzamides (e.g., MS-275), (d) cyclic tetrapeptides (e.g., trapoxin), and (e) sulfonamide anilides [5] (see Table 1). In this review, we discuss the biologic and therapeutic effects of HDACIs in treating endometrial cancer, with a special focus on preclinical studies.

HDACI Cell line ED50 (M) TSA Ishikawa 5.2 × 10−8 HEC-1B 5.1 × 10−8 HEC59 7.0 × 10−8 RL95-2 9.8 ×10−8 KLE 7.2 ×10−8 AN3CA 1.9 ×10−8 Ark2 2.5 ×10−8 SAHA Ishikawa 7.8 × 10−7 HEC-1B 7.8 × 10−7 HEC59 1.2 × 10−6 RL95-2 2.4 ×10−6 KLE 2.5 ×10−6 AN3CA 3.1 ×10−6 CBHA Ishikawa 1.8 × 10−6 HHUA 2.5 × 10−6 HEC-1B 2.2 × 10−6 Scriptaid Ishikawa 9.0 × 10−6 NaB Ishikawa 8.3 × 10−4 HEC-1B 8.4 × 10−4 HEC59 1.8 × 10−3 RL95-2 3.0 ×10−3 KLE 3.9 ×10−3 AN3CA 4.1×10−3 VPA Ishikawa 7.0 × 10−4 HEC-1B 7.5 × 10−4 HEC59 8.2 × 10−4 RL95-2 2.5 ×10−3 KLE 2.3 ×10−3 AN3CA 3.8 ×10−3 MS-275 Ishikawa 9.7 × 10−7 HEC-1B 2.2 × 10−6 RL95-2 1.0 × 10−6 HHUA 7.8 ×10−7 AN3CA 5.0 × 10−7 Ark2 5.0 ×10−7 M344 Ishikawa  2.3 × 10−6 Apicidine Ishikawa 1.0 × 10−6 PsA Ishikawa 7.5 × 10−6 Oxamflatin Ishikawa 2.5 × 10−7 AN3CA 2.5 × 10−7 Ark2 2.5 × 10−7

1. Introduction Endometrial cancer is the seventh most common malignancy among women worldwide. Despite the fact that most cases are diagnosed at an early stage, the death rate has increased steadily over the past 20 years. The lack of an effective, standardized adjuvant treatment for women at a high risk of recurrence has contributed to these disappointing results (reviewed in [1]). The most frequent genetic alteration in type I endometrioid carcinoma is PTEN inactivation by mutation, followed by microsatellite instability and mutations of K-ras and β-catenin. In type II cancers, p53 mutation is the most frequent genetic alteration, followed by inactivation of p16 and e-cadherin and amplification of Her2/neu (reviewed in [1]).

ration in type I endometrioid carcinoma is PTEN inactivation by mutation, followed by microsatellite instability and mutations of K-ras and β-catenin. In type II cancers, p53 mutation is the most frequent genetic alteration, followed by inactivation of p16 and e-cadherin and amplification of Her2/neu (reviewed in [1]). One of the most important mechanisms in chromatin remodeling is the post-translational modification of the N-terminal tails of histones by acetylation, which contributes to a “histone code” determining the activity of target genes [2]. Transcriptionally silent chromatin is composed of nucleosomes in which the histones have low levels of acetylation on the lysine residues of their amino-terminal tails. Acetylation of histone proteins neutralizes the positive charge on lysine residues and disrupts the nucleosome structure, allowing unfolding of the associated DNA with subsequent access by transcription factors, resulting in changes in gene expression. Acetylation of core nucleosomal histones is regulated by the opposing activities of histone acetyltransferases (HATs) and histone deacetylases (HDACs). HDACs catalyze the removal of acetyl groups on the amino-terminal lysine residues of core nucleosomal histones, and this activity is generally associated with transcriptional repression. HDACs remove the acetyl groups which then induce a positive charge on the histones, and this suppresses gene transcription, including tumor suppressor genes silenced in cancer. Moreover, acetylation of histones facilitates destabilization of DNA-nucleosome interaction and renders DNA more accessible to transcription factors [3]. Aberrant recruitment of HDAC activity has been associated with the development of certain human cancers [4]. HDAC inhibitors (HDACIs) can inhibit cancer cell growth in vitro and in vivo, revert oncogene-transformed cell morphology, induce apoptosis, and enhance cell differentiation.

The classes of HDACIs that have been identified are: (a) organic hydroxamic acids (e.g., Trichostatin A (TSA) and suberoylanilide bishydroxamine (SAHA)) (b) short-chain fatty acids (e.g., butyrates and valproic acid (VPA)), (c) benzamides (e.g., MS-275), (d) cyclic tetrapeptides (e.g., trapoxin), and (e) sulfonamide anilides [5] (see Table 1). In this review, we discuss the biologic and therapeutic effects of HDACIs in treating endometrial cancer, with a special focus on preclinical studies. 2. Mechanism of Action Histone deacetylases (HDACs) comprise a family of 18 genes that are subdivided into four classes [6]. Classes I, II, and IV are referred to as ‘‘classical” HDACs and are generally simultaneously targeted by most HDACIs (Table 1). HDACIs were initially discovered on the basis of their ability to reverse the malignant phenotype of transformed cells in culture. It has been shown that HDACIs carry the potential to activate differentiation programs on one hand, while on the other hand they were also shown to inhibit cell proliferation by cell cycle arrest in the G1 and/or G2 phases of the cell cycle and to induce apoptosis in cultured transformed cells. p21WAF1 and p27KIP1 are cyclin-dependent kinase inhibitors (CDKIs) that bind to cyclin-dependent kinase complexes and decrease kinase activity, and may act as key regulators of the G0/G1 accumulation (reviewed in [7]). The p21WAF1 expression in particular is induced by HDACIs in various cell lines. Additionally, this event is associated with both an increase in histone acetylation in the promoter region of the p21WAF1 gene and a selective loss of a specific HDAC enzyme, HDAC1, in the same region [8]. Therefore, the upregulation of p21WAF1 is a direct consequence of HDACIs on p21WAF1 transcription. In the future, testing should be conducted using p21WAF1-negative cell lines to see if p21WAF1 is absolutely required for HDACI-induced growth arrest. Takai et al. examined the effect of HDACIs on the expression of p21WAF1 and p27KIP1 in endometrial cancer cells by Western blot analysis. HDACIs markedly upregulated the level of p21WAF1 and p27KIP1 proteins, which were expressed at negligible levels in the untreated cell lines. Conversely, HDACIs decreased the levels of cyclin D1 and cyclin D2. HDACIs decreased the bcl-2 levels. E-cadherin binds to β-catenin and can act as a tumor suppressor gene; its promoter has CpG islands which are frequently methylated in selected cancers. HDACIs markedly increased the expression level of E-cadherin in endometrial cancer cells and exhibited antiproliferative activity in these cells [9–14].

e bcl-2 levels. E-cadherin binds to β-catenin and can act as a tumor suppressor gene; its promoter has CpG islands which are frequently methylated in selected cancers. HDACIs markedly increased the expression level of E-cadherin in endometrial cancer cells and exhibited antiproliferative activity in these cells [9–14]. HDACIs have also been shown to generate reactive oxygen species (ROS) in solid tumor and leukemia cells [15–17], which may contribute to the mechanism. HDACIs have been shown to inhibit angiogenesis. HDACIs repress the expression of proangiogenic factors such as HIF1α, VEGF, VEGF receptor, endothelial nitric oxide synthase, IL-2 and IL-8 and the induction of antiangiogenic factors, such as p53 and von Hippel-Lindau (reviewed in [18]). HDACIs should not be considered to act solely as enzyme inhibitors of HDACs. A large variety of nonhistone transcription factors and transcriptional co-regulators are known to be modified by acetylation. HDACIs can alter the degree of acetylation nonhistone effector molecules and thereby increase or repress the transcription of genes by this mechanism. Examples include: ACTR, cMyb, E2F1, EKLF, FEN 1, GATA, HNF-4, HSP90, Ku70, NFκB, PCNA, p53, RB, Runx, SF1 Sp3, STAT, TFIIE, TCF, YY1, and so forth, (reviewed in [18]).

. HDACIs can alter the degree of acetylation nonhistone effector molecules and thereby increase or repress the transcription of genes by this mechanism. Examples include: ACTR, cMyb, E2F1, EKLF, FEN 1, GATA, HNF-4, HSP90, Ku70, NFκB, PCNA, p53, RB, Runx, SF1 Sp3, STAT, TFIIE, TCF, YY1, and so forth, (reviewed in [18]). 3. Overview and Preclinical Studies of HDACIs A variety of structurally distinct classes of compounds that inhibit deacetylation of both histone and non-histone proteins have gradually been identified (Table 1). Despite the shared capacity of each class of HDACIs to promote histone acetylation, individual HDACIs exert different actions on signal transduction and the induction of differentiation and/or apoptosis. Table 2 shows data from different reports investigating endometrial cancer cell lines treated with different classes of HDACIs. Many of the in vitro studies use the Ishikawa cell line. Nishida. succeeded in establishing of a well-differentiated endometrial adenocarcinoma cell line, Ishikawa cells, from a 39-year old Japanese patient more than 20 years ago [19]. Because this cell line bears estrogen and progesterone receptors, the cells have been used in numerous basic research areas such as reproductive biology and molecular science.

of a well-differentiated endometrial adenocarcinoma cell line, Ishikawa cells, from a 39-year old Japanese patient more than 20 years ago [19]. Because this cell line bears estrogen and progesterone receptors, the cells have been used in numerous basic research areas such as reproductive biology and molecular science. 3.1. Trichostatin A (TSA) The trichostatins were initially isolated from Streptomyces hygroscopicus as antifungal antibiotics in 1976 [20, 21]. About 10 years later, TSA and its analogues were discovered to induce cell differentiation of murine erythroleukemia cells and to induce hyperacetylation of histone proteins at nanomolar concentrations. TSA has been extensively studied; it has antitumor activity and can induce differentiation of some cancer cell lines, but its clinical utility has been restricted because of toxic side-effects in vivo [22]. TSA causes mitotic arrest through the formation of aberrant mitotic spindles, probably by interfering with chromosome attachment, but does not affect mitotic microtubules [22]. This effect may account for the higher cytotoxicity of TSA in comparison to other HDACIs (i.e., suberoylanilide bishydroxamine). HDAC inhibition is not believed to have a generalized effect on the genome, but only on the transcription of a small subset of the genome. Differential display analysis of transformed lymphoid cell lines revealed that the expression of only 2%–5% of transcribed genes is changed significantly after treatment with TSA [23]. The effective dose of TSA that inhibited 50% clonal growth (ED50) of the endometrial cancer cell lines (Ishikawa, HEC-1B, HEC59, RL95-2, KLE, and AN3CA) was calculated, and ranged between 5.1 × 10−8 M and 1.9 × 10−7 M [9] (Table 2). Dowdy et al. demonstrated that combined treatment with TSA and paclitaxel caused synergistic inhibition of cell growth of Ark2 and KLE endometrial cancer cells [24]. These effects were confirmed in a mouse xenograft model. Treatment with TSA and paclitaxel led to a significant increase in acetylated tubulin and microtubule stabilization. This study provides the evidence of nonhistone protein acetylation as one possible mechanism by which HDACIs reduce cancer growth.

cer cells [24]. These effects were confirmed in a mouse xenograft model. Treatment with TSA and paclitaxel led to a significant increase in acetylated tubulin and microtubule stabilization. This study provides the evidence of nonhistone protein acetylation as one possible mechanism by which HDACIs reduce cancer growth. 3.2. Suberoylanilide Bishydroxamine (SAHA, Vorinostat) Hydroxamic acid type inhibitors make up the largest and broadest group of HDACIs described to date. The inhibition of HDACs by SAHA occurs through a direct interaction with the catalytic site of the enzyme, as shown by X-ray crystallography studies [25]. Among the synthetic HDACIs, SAHA is the most advanced candidate as a cancer therapeutic drug, and is under phase I and II clinical trials [26, 27]. SAHA has significant antitumor activity against many tumor types at dosages that can be tolerated by patients when administered intravenally and orally [28]. Some HDACIs (e.g., TSA and trapoxin) are of limited therapeutic use due to poor bioavailability in vivo and have toxic side effects at high doses. SAHA, however, is relatively safe and non-toxic in vivo. The effective dose of SAHA that inhibited 50% clonal growth (ED50) of the endometrial cancer cell lines (Ishikawa, HEC-1B, HEC59, RL95-2, KLE, and AN3CA) was calculated and ranged between 7.8 × 10−7 M and 3.1 × 10−6 M [9] (Table 2).

ic side effects at high doses. SAHA, however, is relatively safe and non-toxic in vivo. The effective dose of SAHA that inhibited 50% clonal growth (ED50) of the endometrial cancer cell lines (Ishikawa, HEC-1B, HEC59, RL95-2, KLE, and AN3CA) was calculated and ranged between 7.8 × 10−7 M and 3.1 × 10−6 M [9] (Table 2). 3.3. m-carboxycinnamic Acid Bishydroxamide (CBHA) CBHA is a member of a recently synthesized family of hybrid polar compounds that have been shown to be inhibitors of HDAC [29] and potent inducers of transformed cell growth arrest and terminal differentiation at micromolar (LD50 range, 1–4 μM) concentrations [30]. The effective dose of CBHA that inhibited 50% clonal growth (ED50) of the endometrial cancer cell lines (Ishikawa, HEC-1B, HHUA) was calculated and ranged between 1.8 × 10−6 M and 2.5 × 10−6 M for CBHA [10] (Table 2). On the other hand, normal endometrial epithelial cells were viable after treatment with the same doses of CBHA that induced growth inhibition of endometrial cancer cells [10].

endometrial cancer cell lines (Ishikawa, HEC-1B, HHUA) was calculated and ranged between 1.8 × 10−6 M and 2.5 × 10−6 M for CBHA [10] (Table 2). On the other hand, normal endometrial epithelial cells were viable after treatment with the same doses of CBHA that induced growth inhibition of endometrial cancer cells [10]. 3.4. Scriptaid Using a high-throughput system based on a stably integrated transcriptional reporter to screen a library of 16,320 compounds (DIVERset, Chembridge, San Diego, CA), Su et al. identified a novel HDACI, termed Scriptaid [31]. Nullscript, which possesses a shorter side-chain (3C) than Scriptaid (5C) between the tricyclic core and the carbonyl group, was inactive in transcriptional facilitation. This confirms that the linker chain has to be a certain length for HDAC inhibition to occur. Scriptaid has a common structure with TSA and SAHA, that is, a hydroxamic acid zinc-binding group linked via a spacer (5 or 6 CH2) to a hydrophobic group [31]. Using an immunoblotting assay of histone deacetylation, Su et al. demonstrated that Scriptaid is a potent HDACI with a >100-fold increase in histone acetylation, with relatively low toxicity [31]. Scriptaid conferred the greatest effect on augmentation of the signal transduction TGFβ pathway, including a number of human suppressor genes such as SMAD4 [31]. The effect of Scriptaid in human cancers, however, has not been fully examined. A recent study by Keen et al. indicated that Scriptaid had a significant growth-suppressing effect on ER-negative human breast cancer cells [32]. The effective dose of Scriptaid that inhibited 50% clonal growth (ED50) of the Ishikawa endometrial cancer cell line was calculated at 9.0 × 10−6 M [11] (Table 2). On the other hand, normal endometrial epithelial cells were viable after treatment with the same doses of Scriptaid that induced growth inhibition of endometrial cancer cells [11].

riptaid that inhibited 50% clonal growth (ED50) of the Ishikawa endometrial cancer cell line was calculated at 9.0 × 10−6 M [11] (Table 2). On the other hand, normal endometrial epithelial cells were viable after treatment with the same doses of Scriptaid that induced growth inhibition of endometrial cancer cells [11]. 3.5. Sodium Butyrate (NaB) It was first reported in 1978 that millimolar concentrations of sodium butyrate (NaB) inhibited HDACs in vitro [33]. NaB is normally present in the human colon as a product of the metabolic degradation of complex carbohydrates by colonic bacteria and regulates the physiological differentiation of colonocytes, suggesting its possible use in the prevention of colorectal cancer and the treatment of premalignant and neoplastic lesions. Butyrate and its derivatives have a long history of safe clinical use in the treatment of inherited and acquired metabolic disorders. Some studies suggest that many of the cellular activities of phenylbutyrate are more dependent on its butyric acid component than its phenyl group. A recent study by Terao et al. indicated that NaB had a significant growth-suppressing effect on human endometrial and ovarian cancer cells irrespective of their p53 gene status [34]. NaB, a low-potency HDACI, has been extensively studied; it has antitumor activity and can induce differentiation of some cancer cell lines, but its clinical utility has been restricted by its short half-life (5 minutes), limiting the ability to achieve a therapeutic plasma level. NaB and phenylbutyrate are degraded rapidly after i.v. administration and therefore require high doses exceeding 400 mg/kg/day [35]. Furthermore, these compounds are not specific for HDACs as they also inhibit phosphorylation and methylation of proteins as well as DNA methylation [35]. The effective dose of NaB that inhibited 50% clonal growth (ED50) of the endometrial cancer cell lines (Ishikawa, HEC-1B, HEC59, RL95-2, KLE, and AN3CA) has been calculated and ranged between 8.3 × 10−4 M and 4.1 × 10−3 M for NaB [9] (Table 2).

hosphorylation and methylation of proteins as well as DNA methylation [35]. The effective dose of NaB that inhibited 50% clonal growth (ED50) of the endometrial cancer cell lines (Ishikawa, HEC-1B, HEC59, RL95-2, KLE, and AN3CA) has been calculated and ranged between 8.3 × 10−4 M and 4.1 × 10−3 M for NaB [9] (Table 2). 3.6. Valproic Acid (VPA) Valproic acid, a shortchain fatty acid, has been approved for clinical use in the treatment of epilepsy and is frequently used in clinical trials and for in vitro research based on its HDAC inhibitory effect at comparatively high (millimolar) concentrations [36]. Valproic acid has also been identified as an antiproliferative agent and HDACI [37]. Some HDACIs (e.g., TSA and trapoxin) are of limited therapeutic use due to poor bioavailability in vivo as well as toxic side effects at high doses, but VPA is relatively safe and non-toxic in vivo. The effective dose of VPA that inhibited 50% clonal growth (ED50) of the endometrial cancer cell lines (Ishikawa, HEC-1B, HEC59, RL95-2, KLE, and AN3CA) has been calculated and ranged between 7.0 × 10−4 M and 3.8 × 10−3 M [9] (Table 2).

toxic side effects at high doses, but VPA is relatively safe and non-toxic in vivo. The effective dose of VPA that inhibited 50% clonal growth (ED50) of the endometrial cancer cell lines (Ishikawa, HEC-1B, HEC59, RL95-2, KLE, and AN3CA) has been calculated and ranged between 7.0 × 10−4 M and 3.8 × 10−3 M [9] (Table 2). A previous study tested the ability of VPA to inhibit the growth of human HEC-1B endometrial cancer tumors growing in immunodeficient mice [9]. Administration of VPA remarkably suppressed the growth of the tumors (P <.01). No significant differences in either the mean weights, histology of the internal organs, and mean blood chemistries, including liver parameters and hematopoietic values, were found between diluent-treated mice and those that received 5 weeks of therapy. These tumors were sampled for the expression of p21WAF1 using immunohistochemistry on formalin-fixed paraffin-embedded sections. HEC-1B endometrial cancer cells treated with VPA showed strong nuclear staining. Control cancer cells from untreated mice had negative or focal weak staining for p21WAF1. This in vivo study shows that VPA at 0.3–1.5 mM inhibited cell proliferation, induced cell cycle arrest and stimulated apoptosis in endometrial cancer cells. This range of concentrations of VPA can be achieved in a patient's serum when that patient is receiving a daily dose of 20–30 mg/kg for epilepsy. These data are also consistent with the in vitro data. This anticancer activity occurred without any major side-effects, raising hopes that VPA may become a useful therapy for endometrial cancers. Furthermore, VPA has convenient pharmacokinetic properties with a significantly longer biological half-life than the other HDACIs [38].

re also consistent with the in vitro data. This anticancer activity occurred without any major side-effects, raising hopes that VPA may become a useful therapy for endometrial cancers. Furthermore, VPA has convenient pharmacokinetic properties with a significantly longer biological half-life than the other HDACIs [38]. 3.7. MS-275 (Entinostat) MS-275 (MS-27-275) is a synthetic novel benzamide which exerts HDAC inhibitory activity and also induces the expression of the cyclin-dependent kinase inhibitor p21WAF1 and gelsolin, and changes the cell cycle distribution [38, 39]. MS-275 has shown antiproliferative activity in various in vitro and in vivo human tumor models [40, 41], and is currently being tested in clinical trials involving patients with solid tumors or hematological malignancies [42]. The effective dose of MS-275 that inhibited 50% clonal growth (ED50) of the endometrial cancer cell lines (Ishikawa, HHUA, HEC-1B and RL95-2) was calculated and ranged between 7.8 × 10−7 M and 2.2 × 10−6 M [12] (Table 2). On the other hand, normal endometrial epithelial cells were viable after the treatment with the same doses of MS-275 that induced growth inhibition of endometrial cancer cells [12]. Jiang et al. reported that over the course of 4 days, there was a 60% reduction in the serous endometrial cancer cell line Ark2 cell counts by MS-275 (which they called HDAC-I1) (0.5 μM) treatments, as compared to controls treated with DMSO solvent (Table 2). They reported growth inhibition of both endometrioid (Ishikawa and AN3) and serous (Ark2) endometrial carcinomas [43].

e was a 60% reduction in the serous endometrial cancer cell line Ark2 cell counts by MS-275 (which they called HDAC-I1) (0.5 μM) treatments, as compared to controls treated with DMSO solvent (Table 2). They reported growth inhibition of both endometrioid (Ishikawa and AN3) and serous (Ark2) endometrial carcinomas [43]. 3.8. M344 Synthetic amide analogs were discovered to have a common structure with TSA [44]. Using an in vitro enzyme inhibition assay of histone deacetylation, Jung et al. demonstrated that M344 is a potent HDACI and an inducer of terminal cell differentiation [44]. The effective dose of M344 that inhibited 50% clonal growth (ED50) of the Ishikawa endometrial cancer cell line was calculated at 2.3 × 10−6 M [13] (Table 2). On the other hand, normal endometrial epithelial cells were viable after the treatment with the same doses of M344 that induced growth inhibition of endometrial cancer cells [13].

of M344 that inhibited 50% clonal growth (ED50) of the Ishikawa endometrial cancer cell line was calculated at 2.3 × 10−6 M [13] (Table 2). On the other hand, normal endometrial epithelial cells were viable after the treatment with the same doses of M344 that induced growth inhibition of endometrial cancer cells [13]. 3.9. Apicidin Cyclic peptide HDACIs can be further divided into two classes: those with an epoxyketone group such as HC-toxin and trapoxin, and those without such a group (apicidin, depsipeptide or FK228). Apicidin is a novel cyclic tetrapeptide with a potent broad spectrum of antiprotozoal activity against Apicomplexan parasites [45]. Its structure is related to trapoxin, a potent HDACI, and some biological activity, including antiproliferative and toxic effects, have been shown in some cancer cell lines [46]. The effective dose of apicidin that inhibited 50% clonal growth (ED50) of the Ishikawa endometrial cancer cell lines was calculated at 1.0 × 10−6 M [14] (Table 2). On the other hand, normal endometrial epithelial cells were viable after the treatment with the same doses of HDACIs that induced growth inhibition of endometrial cancer cells [14]. Ueda et al. [14] and Ahn et al. [47] independently demonstrated that apicidin has antitumor properties on Ishikawa endometrial cancer cells by selectively inducing the genes related to cell cycle arrest and apoptosis.

eatment with the same doses of HDACIs that induced growth inhibition of endometrial cancer cells [14]. Ueda et al. [14] and Ahn et al. [47] independently demonstrated that apicidin has antitumor properties on Ishikawa endometrial cancer cells by selectively inducing the genes related to cell cycle arrest and apoptosis. 3.10. Psammaplin A Psammaplin A (PsA) is a natural bromotyrosine derivative from a two-sponge association, Poecillastra sp. and Jaspis sp., which was first isolated from the Psammaplysilla sponge [45]. It was reported that PsA has antibacterial and antitumor properties, and also inhibits various enzymes including topoisomerase II, farnesyl protein transferase, leucine amino peptidase, and chitinase (reviewed in [48]). Recently, it was reported that PsA inhibits both HDAC and DNA methyltransferase (DNMT) as epigenetic modifiers of the tumor suppressor gene [49]. PsA caused antiproliferative activity and induced cell cycle arrest or apoptosis in Ishikawa human endometrial cancer cells. PsA inhibited the proliferation of Ishikawa cells in a dose-dependent manner. The 50% inhibitory concentration (IC50) of PsA was found to be 5 μg/mL (7.5 × 10−6 M) after 48 h treatment (Table 2). PsA increased the proportion in the G1 phase and G2/M phases of the cell cycle [48].

shikawa human endometrial cancer cells. PsA inhibited the proliferation of Ishikawa cells in a dose-dependent manner. The 50% inhibitory concentration (IC50) of PsA was found to be 5 μg/mL (7.5 × 10−6 M) after 48 h treatment (Table 2). PsA increased the proportion in the G1 phase and G2/M phases of the cell cycle [48]. 3.11. Oxamflatin Oxamflatin is an aromatic sulfonamide derivative with a hydroxamic acid group that was identified as a compound inducing the morphological reversion of v-K-ras-transformed NIH3T3 cells from a chemical library [50]. In addition, the morphology of NIH3T3 cells transformed by various other oncogenes such as v-sis, v-src, MEKEE or v-fos was also reverted by oxamflatin. Kim et al. analysed the effect of oxamflatin on the proliferation of eight mouse and human tumor cell lines. The 50% inhibitory concentrations of oxamflatin for all the cell lines except CCD-19Lu, a normal human lung cell line, were below 0.72 μg/mL, while that for CCD-19Lu was 1.4 μg/mL [51]. Over the course of 4 days, there was a 78% reduction in the serous endometrial cancer cell line Ark2 cell counts by oxamflatin (0.25 μM) treatments, as compared to controls treated with DMSO solvent. The most striking observation is the 95% reduction in cell counts following the administration of 0.75 μM oxamflatin to Ark2 cells [43]. This report resulted in growth inhibition of both endometrioid (Ishikawa and AN3) and serous (Ark2) endometrial carcinomas.

atments, as compared to controls treated with DMSO solvent. The most striking observation is the 95% reduction in cell counts following the administration of 0.75 μM oxamflatin to Ark2 cells [43]. This report resulted in growth inhibition of both endometrioid (Ishikawa and AN3) and serous (Ark2) endometrial carcinomas. 4. Conclusions In this review, we summarize recent preclinical studies on the use of HDACIs, especially in human endometrial cancer cells. Many questions are currently still unanswered with respect to HDACI specificities for definite tumor subtypes and the molecular mechanisms underlying HDACI-induced differentiation, cell cycle arrest and apoptosis. In addition, the regulation mechanisms of the specific gene expression and recruitment of HDAC complex to the specific promoter sites remain still to be determined. Also, it is still unclear to what extent different HDACs exhibit different and potentially overlapping functions, and it is important to distinguish the HDAC specificity of HDACIs for the development of selective therapy on the molecular level. Further work is needed to improve our understanding of why transformed cells are more susceptible to the effect of HDACIs than normal cells. Also, combinations of HDACIs with differentiation-inducing agents, with cytotoxic agents, and even with gene therapy may represent novel therapeutic strategies and new hope for the treatment of endometrial cancer.

our understanding of why transformed cells are more susceptible to the effect of HDACIs than normal cells. Also, combinations of HDACIs with differentiation-inducing agents, with cytotoxic agents, and even with gene therapy may represent novel therapeutic strategies and new hope for the treatment of endometrial cancer. Acknowledgments The study was supported by a grant (project code FK344 to NT) from the Japan Society of Gynecologic Oncology, a Grant-in-Aid (no. 21592139 to NT) for Scientific Research from the Ministry of Education, Culture, Sports, Science, and Technology, Japan, and a Research Fund at the Discretion of the President, Oita University. Table 1 Overview of frequently used histone deacetylase inhibitors that are available for clinical and research purposes. Substance groups Derivatives Isotype Hydroxamates Trichostatin A (TSA) I, II Suberoylanilide hydroxamic acid (SAHA, vorinostat) I, II, IV LBH589 (panobinostat) I, II, IV PCI24781 (CRA-024781) I, IIb LAQ824 I, II PXD101 (belinostat) I, II, IV ITF2357 I, II SB939 Unknown JNJ-16241199 (R306465) I m-carboxycinnamic acid bishydroxamide (CBHA) Scriptaid Oxamflatin Pyroxamide Cyclic hydroxamic acid containing peptides (CHAPs) Short chain fatty acids Butyrate I, IIa Valproate I, IIa AN-9 OSU-HDAC42 Benzamides MS-275 (entinostat) 1, 2, 3, 9 MGCD0103 1, 2, 3, 11 Pimelic diphenylamide 1, 2, 3 M344 N-acetyldinaline (CI-994) Cyclic tetrapeptides Apicidine I, II Trapoxins HC-toxin Chlamydocin Depsipeptide (FR901228 or FK228) (romidepsin) 1, 2, 4, 6

Substance groups Derivatives Isotype Hydroxamates Trichostatin A (TSA) I, II Suberoylanilide hydroxamic acid (SAHA, vorinostat) I, II, IV LBH589 (panobinostat) I, II, IV PCI24781 (CRA-024781) I, IIb LAQ824 I, II PXD101 (belinostat) I, II, IV ITF2357 I, II SB939 Unknown JNJ-16241199 (R306465) I m-carboxycinnamic acid bishydroxamide (CBHA) Scriptaid Oxamflatin Pyroxamide Cyclic hydroxamic acid containing peptides (CHAPs) Short chain fatty acids Butyrate I, IIa Valproate I, IIa AN-9 OSU-HDAC42 Benzamides MS-275 (entinostat) 1, 2, 3, 9 MGCD0103 1, 2, 3, 11 Pimelic diphenylamide 1, 2, 3 M344 N-acetyldinaline (CI-994) Cyclic tetrapeptides Apicidine I, II Trapoxins HC-toxin Chlamydocin Depsipeptide (FR901228 or FK228) (romidepsin) 1, 2, 4, 6 Sulfonamide anilides N-2-aminophenyl-3-[4-(4-methylbenzenesulfonylamino)-phenyl]-2-propenamide Others Depudecin NDH-51 KD5150 Pan-HDACI Class I: HDAC1, -2, -3 and -8; class IIa: HDAC4, -5, -7, and -9; class IIb: HDAC 6, and -10; class III: SIRT1-7; class IV: HDAC11. Table 2 Data investigating endometrial cancer cell lines treated with different classes of HDACIs. HDACI Cell line ED50 (M) TSA Ishikawa 5.2 × 10−8 HEC-1B 5.1 × 10−8 HEC59 7.0 × 10−8 RL95-2 9.8 ×10−8 KLE 7.2 ×10−8 AN3CA 1.9 ×10−8 Ark2 2.5 ×10−8 SAHA Ishikawa 7.8 × 10−7 HEC-1B 7.8 × 10−7 HEC59 1.2 × 10−6 RL95-2 2.4 ×10−6 KLE 2.5 ×10−6 AN3CA 3.1 ×10−6 CBHA Ishikawa 1.8 × 10−6 HHUA 2.5 × 10−6 HEC-1B 2.2 × 10−6 Scriptaid Ishikawa 9.0 × 10−6 NaB Ishikawa 8.3 × 10−4 HEC-1B 8.4 × 10−4 HEC59 1.8 × 10−3 RL95-2 3.0 ×10−3 KLE 3.9 ×10−3 AN3CA 4.1×10−3

colorectal carcinomas [16]. By multivariate analysis, a significant correlation between MSI-positive tumors and tumor-infiltrating lymphocytes in endometrioid endometrial carcinoma was found: 40 tumor-infiltrating lymphocytes/10 high power fields has a sensitivity of 85% and a specificity of 46% in predicting MSI [16]. 2.4. Molecular Pathology of Nonendometrioid Carcinomas 2.4.1. p53 The p53 tumor suppressor gene locates to chromosome 17p13.1. While p53 mutations occur in 90% of non-endometrioid endometrial carcinoma, they are only present in 10–20% of endometrioid endometrial carcinoma, which are mostly high-grade [7, 18]. The abnormal p53 expression has been found in 11% of grade 1 endometrioid endometrial carcinoma [18]. This finding supports that p53 mutations may influence progression of endometrioid endometrial carcinomas to non-endometrioid endometrial carcinomas [9]. In fact, p53 mutation is the most characteristic genetic alteration of non-endometrioid endometrial carcinomas [9, 10] and may be useful in their distinction from endometrioid endometrial carcinomas [22]. In p53 positive endometrioid endometrial carcinoma, p53 protein accumulation may be secondary to changes in its upstream regulatory proteins rather than the p53 gene itself. Several genes, including MDM2 and p14 AR, that regulate p53 levels have been shown to cause detectable levels of p53 in the absence of p53 mutation. Alternatively, nonspecific DNA damage such as that induced by irradiation is also known to induce accumulation of wild-type p53 [12]. In normal cells, p53 is rapidly degraded and thus cannot be detected by immunostaining. p53 mutations produce a non-functional protein that resists degradation and can be visualized by immunostaining [11, 18]. However, loss of function of p53 resulting from LOH may not correlate with protein overexpression. In addition, frameshift mutations and stop codons lead to a truncated protein, which is not detected by antibodies and leads to negative immunohistochemistry [11, 18] After DNA damage, nuclear p53 accumulates and causes cell cycle arrest by inhibiting cyclin D1 phosphorylation of the Rb gene and thereby promoting apoptosis [9, 13]. Overexpression of p53 is associated with high histological grade and advanced stage as well as unfavorable prognosis [9, 18].

1. Introduction Ovariohysterectomy in the bitch is a surgical procedure consisting of laparotomy with ablation of both ovaries and the uterus. This procedure is indicated for the following [1]. Uterine tumours. Serious uterine lesions, whether traumatic or infectious in origin; the most common cause being dystocia during parturition. Other pathologies that justify an ovariohysterectomy include metorrhagia, pyometria, glandular-cystic uterine hyperplasia with secondary infection leading to chronic metritis; the latter usually occurs postoestrus (“postoestrus metritis”) and is initially treated medically, as with acute postpartum metritis, surgery becomes a necessity once the disease becomes chronic and recurrent [2–5]. These alterations in the uterine mucosa are the result of ovarian hormonal imbalances. Metritic pathologies have become increasingly common since the introduction and growing popularity of synthetic progesterone treatments such as medroxyprogesterone acetate, which are used to prevent or eliminate heats where the onset of metritis is common especially if they are used after the 3rd day of pro-oestrus [6]. This procedure should only be undertaken if the bitch is in a fit state to withstand general anaesthesia. She will reabsorb the toxins produced in the uterus, or lick any pus that accumulates at the lower commissure of the vulva, leading to gastroenteritis and hepatonephritis and subsequently diarrhoea, vomiting, and raised urea (normal value around 0.6 g/L), and creatinine (normal value around 10 mg/L).

neral anaesthesia. She will reabsorb the toxins produced in the uterus, or lick any pus that accumulates at the lower commissure of the vulva, leading to gastroenteritis and hepatonephritis and subsequently diarrhoea, vomiting, and raised urea (normal value around 0.6 g/L), and creatinine (normal value around 10 mg/L). However, if the ureamia is greater than 0.6 g/L, we advise the administration of Lespedeza capitata LESPEDEZIA N.D.v, 0.7–1 mL/kg morning and evening for 2 days via IM or SC injection, without exceeding 20 mL/injection/animal. The latter is a mild diuretic, hypoazotemic agent that acts via renal vasodilatation and stimulates the activity of the renal parenchyma. These injections should be combined with intravenous fluid therapy with isotonic NaCl solution at 0.9% and the urea levels checked 2 days later. Antibiotic prophylaxis with Cefalexin RILEXINE N.D.v, at a dose of 20 mg/kg every 12 hours during 3 days, is also advisable to prevent bacteraemia. Another indication is that of convenience, that is, sterilisation, as many owners complain of the manifestations of heat with vulvular discharge, as well as the problems associated with repeated matings. Finally, and with the owner's consent, ovariohysterectomy can be proposed as a radical alternative to medical abortion following an unwanted pregnancy, as it also involves definitive sterilisation [1].

Another indication is that of convenience, that is, sterilisation, as many owners complain of the manifestations of heat with vulvular discharge, as well as the problems associated with repeated matings. Finally, and with the owner's consent, ovariohysterectomy can be proposed as a radical alternative to medical abortion following an unwanted pregnancy, as it also involves definitive sterilisation [1]. However, there are certain contraindications to the procedure, such as if the bitch presents with a generalised condition with hypothermia, dehydration, and mydriasis. Similarly, animals presenting with hepatorenal insufficiency should not undergo general anaesthesia if the urea levels are greater than 0.6 g/L and the creatinine is greater than 10 mg/L, such animals are associated with poor peri- and postoperative survival. It is therefore essential to perform a complete, detailed preoperative clinical examination, with blood tests for serum biochemistry. 2. Materials and Method 2.1. Anatomy The genital apparatus of the bitch is primarily located in the abdominal cavity, with the exception of the vagina, which lies in the pelvis [6, 7] (Figures 1 and 2).

However, there are certain contraindications to the procedure, such as if the bitch presents with a generalised condition with hypothermia, dehydration, and mydriasis. Similarly, animals presenting with hepatorenal insufficiency should not undergo general anaesthesia if the urea levels are greater than 0.6 g/L and the creatinine is greater than 10 mg/L, such animals are associated with poor peri- and postoperative survival. It is therefore essential to perform a complete, detailed preoperative clinical examination, with blood tests for serum biochemistry. 2. Materials and Method 2.1. Anatomy The genital apparatus of the bitch is primarily located in the abdominal cavity, with the exception of the vagina, which lies in the pelvis [6, 7] (Figures 1 and 2). The neck of the uterus is relatively short, it measures 1-2 cm long, and lies a few centimetres in front of the anterior border of the pubis; it is followed by the body of the uterus, which measures 3–5 cm in length in the intrabdominal position, and which starts from the anterior straight of the pelvis then divides after a few centimetres into two divergent horns, which lie on the floor of the abdomen on either side of the linea alba, then travel back up towards the ovaries; the latter are situated in the costolumbar angle, one or two centimetres from the bisection and buried in a fatty ovarian sac, which opens medially [7, 8].

munohistochemistry [11, 18] After DNA damage, nuclear p53 accumulates and causes cell cycle arrest by inhibiting cyclin D1 phosphorylation of the Rb gene and thereby promoting apoptosis [9, 13]. Overexpression of p53 is associated with high histological grade and advanced stage as well as unfavorable prognosis [9, 18]. Endometrial intraepithelial carcinoma (EIC), the putative precursor lesion to serous carcinomas [4, 13, 18, 22, 23], characterized by replacement of the surface epithelium by malignant cells exhibiting cytological features similar to those of serous carcinoma [9, 23]. EIC has been reported in nearly 90% of uteri containing serous carcinoma that is often extensive and multifocal [23]. Mutations of p53 are also found in 75–80% of EIC. It is postulated that mutation in one allele occurs early during the development of serous carcinoma, and loss of the second normal allele occurs late in the progression to carcinoma [4]. p53 mutations are almost always associated with aneuploidy and do not seem to occur with PTEN mutations in the same tumor [10, 11].

s after a few centimetres into two divergent horns, which lie on the floor of the abdomen on either side of the linea alba, then travel back up towards the ovaries; the latter are situated in the costolumbar angle, one or two centimetres from the bisection and buried in a fatty ovarian sac, which opens medially [7, 8]. The uterus receives its blood supply from the right and left uterine arteries (Figures 1 and 2). The body of the uterus that lies closest to the oviduct is irrigated by the uterine branch of the ovarian artery, whilst the neck and remainder of the body are supplied by the uterine branch of the vaginal artery. The uterine artery provides the majority of the organ's blood supply and serves no other organs; it originates from the internal iliac artery along with the umbilical artery. 2.2. Surgical Approaches Easy access to the genital apparatus is gained via the linea alba; the incision starts at the umbilicus and ends 2 to 3 cm cranial to the anterior border of the pubis. This approach provides direct access to the uterine horns and facilitates prehension of the ovaries. 2.3. Surgical Technique 2.3.1. Surgical Equipment (a) Preparation of the Animal Ensure that the bitch has been fasted since the previous day especially in the context of elective surgery: however, in an emergency situation, after induction of anaesthesia rapid intubation with a cuffed endotracheal tube should prevent aspiration of stomach contents due to gastric reflux.

(a) Preparation of the Animal Ensure that the bitch has been fasted since the previous day especially in the context of elective surgery: however, in an emergency situation, after induction of anaesthesia rapid intubation with a cuffed endotracheal tube should prevent aspiration of stomach contents due to gastric reflux. In the event of pyometria or metritis, the bitch's organism has to eliminate the toxins produced during the infection, it is therefore essential for the success of the procedure to choose anaesthetic agents with minimal toxicity. Various protocols are available, these include: IV premedication with valium and morphine at 0.25 mg/kg and 0.1 mg/kg, respectively, followed by induction of anaesthesia with propofol at a dose of 5 mg/kg, the volume is increased slowly until the animal is sufficiently well anaesthetised to enable intubation. Isoflurane gas is then used to maintain anaesthesia; a flow rate of 2% is normally sufficient to maintain a good level of anaesthesia until the end of the procedure. To control perioperative pain, morphine can be administered at the same dose as for premedication, to a maximum of 10 injections/hour to avoid exceeding the threshold of toxicity. Once anaesthetised, the bitch is positioned in dorsal recumbency with her front legs pulled forward and tied to the table, and the back legs tied back; the bitch is then put on a drip with previously warmed normal saline (0.9%) with glucose.

In the event of pyometria or metritis, the bitch's organism has to eliminate the toxins produced during the infection, it is therefore essential for the success of the procedure to choose anaesthetic agents with minimal toxicity. Various protocols are available, these include: IV premedication with valium and morphine at 0.25 mg/kg and 0.1 mg/kg, respectively, followed by induction of anaesthesia with propofol at a dose of 5 mg/kg, the volume is increased slowly until the animal is sufficiently well anaesthetised to enable intubation. Isoflurane gas is then used to maintain anaesthesia; a flow rate of 2% is normally sufficient to maintain a good level of anaesthesia until the end of the procedure. To control perioperative pain, morphine can be administered at the same dose as for premedication, to a maximum of 10 injections/hour to avoid exceeding the threshold of toxicity. Once anaesthetised, the bitch is positioned in dorsal recumbency with her front legs pulled forward and tied to the table, and the back legs tied back; the bitch is then put on a drip with previously warmed normal saline (0.9%) with glucose. (b) Preparation of the Surgeon The surgeon should wear a clean and sterile gown, scrub their hands thoroughly using surgical scrub solution, and wear sterile gloves. (c) Preparation of the Material In addition to the standard laparotomy kit, the surgeon requires the following instruments:2 babcock forceps, 4 artery forceps, 4 doyen bowel clamps, resorbable multifilament suture material, VICRYL, dec. 3.

(b) Preparation of the Surgeon The surgeon should wear a clean and sterile gown, scrub their hands thoroughly using surgical scrub solution, and wear sterile gloves. (c) Preparation of the Material In addition to the standard laparotomy kit, the surgeon requires the following instruments:2 babcock forceps, 4 artery forceps, 4 doyen bowel clamps, resorbable multifilament suture material, VICRYL, dec. 3. And finally material for the septic phase of the surgery: scalpel, mayo scissors, and resorbable VICRYL Dec 3.5 or 4 for closure of the abdominal wall. 2.3.2. Surgery The surgical zone should be carefully scrubbed using the same type of surgical scrub solution as used by the surgeon, and disinfected using alcohol and surgical antiseptic solution several times over. (a) Principal Phases Laparotomy The skin is incised along the linea alba, that is, the sheath of the rectus abdominus, starting from the umbilicus and ending a few centimetres in front of the pubis. Using a pair of scissors, the subcutaneous connective tissue, which may contain a substantial amount of fatty tissue, is bluntly dissected to visualise the linea alba. Haemostasis is performed before opening the abdominal cavity. If simple swabbing proves insufficient, any bleeders should be ligated or twisted to obtain a very clean surgical field.

Using a pair of scissors, the subcutaneous connective tissue, which may contain a substantial amount of fatty tissue, is bluntly dissected to visualise the linea alba. Haemostasis is performed before opening the abdominal cavity. If simple swabbing proves insufficient, any bleeders should be ligated or twisted to obtain a very clean surgical field. Using rat-tooth forceps, the linea alba is grasped in the middle and tented up before being incised with a pair of scissors. The peritoneum is then punctured using a cannula that is slid towards the umbilicus to enable incision of the linea alba without damaging the abdominal contents, with the cutting edge of the blade turned uppermost. The same procedure is then performed in the opposite direction towards the pubis. If the uterine horns are voluminous they will be seen in the bottom of the surgical field following incision of the peritoneum; normal-sized horns will not be visible, for example, following recovery from postoestral metritis or during routine spaying. To find the uterine horns easily, the operating table is tilted so that the animal's head is below its feet, to move the abdominal organs towards the diaphragm; this is known as the TRENDELENBURG position.

If the uterine horns are voluminous they will be seen in the bottom of the surgical field following incision of the peritoneum; normal-sized horns will not be visible, for example, following recovery from postoestral metritis or during routine spaying. To find the uterine horns easily, the operating table is tilted so that the animal's head is below its feet, to move the abdominal organs towards the diaphragm; this is known as the TRENDELENBURG position. To locate the genital apparatus with ease, the bladder is retracted laterally; cranial to the bladder, the body of the uterus and bifurcation of the horns are easily locatable. One of the horns is then followed cranially up to the ovary, which is hidden in the fat-filled ovarian bursa. The ovary is not visible but can be felt through this ovarian bursa. It is a 1-2 cm long mass, which is exposed after incision of the bursa. Sectioning the Ovarian Pedicle and Broad Ligament – Ovarian Pedicle. The ovary is grasped and babcock forceps placed. The latter are handed to an assistant who holds the ovarian pedicle taught out of the abdomen to facilitate placement of a ligature as close as possible to the root of the pedicle to ensure haemostasis of the ovarian artery. The broad ligament is then punctured with a clamp to grasp the suture material and a ligature is placed in the ovarian pedicle as close as possible to the lumbar wall. Once this ligature has been placed, the ends of the threads are kept long so that the ovarian pedicle can be found with ease in the event of haemorrhage.

The broad ligament is then punctured with a clamp to grasp the suture material and a ligature is placed in the ovarian pedicle as close as possible to the lumbar wall. Once this ligature has been placed, the ends of the threads are kept long so that the ovarian pedicle can be found with ease in the event of haemorrhage. A clamp is then placed between this ligature and the ovary, and the pedicle is sectioned between the two. The ovarian pedicle is held throughout this procedure with a clamp. The quality of the haemostasis is checked; the long ends of the suture material on the ovarian pedicle are then cut. In some cases, such as in the event of hypertrophy of the vascular bundle, it may be advisable to place two ligatures, one around the artery and one around the ovarian vein. Never hold the ligature itself with the clamp, as it might slip off the pedicle when being released back into the abdomen. – Broad Ligament. If the broad ligament is seen to contain large vessels, they should be ligated prior to be being cut. However, if the vessels are invisible and buried under fat, the ligament can simply be torn in the middle above the uterine artery by exerting traction between two swabs with the fingers to tear it from front to back to the level of the cervix, and as close as possible to the lumbar wall. A point of resistance will be encountered within the round ligament; this corresponds to the vaginal process (which corresponds to the scrotum in the male) which explains the risk of inguinal herniation of the uterus in bitches following relaxation of the latter.

the cervix, and as close as possible to the lumbar wall. A point of resistance will be encountered within the round ligament; this corresponds to the vaginal process (which corresponds to the scrotum in the male) which explains the risk of inguinal herniation of the uterus in bitches following relaxation of the latter. Another technique for sectioning the broad ligament involves the placement of a row of overlapping mattress sutures along the length of the ligament before making the section with a scalpel or a pair of scissors. Once the ovarian pedicle has been sectioned, the second horn is located and the corresponding ovarian bursa grasped with Babcock forceps. The ovarian pedicle and broad ligament are sectioned as described previously. Finally, the two uterine horns are replaced back onto pelvis. Suturing the Anterior Portion of the Laparotomy Incision The prolapse of intestinal loops through the incision can cause significant heat and fluid loss, which can have very serious consequences, especially if the bitch is already suffering from deterioration in general status due to severe pyometria, for example. It is therefore advisable to suture the anterior portion of the laparotomy wound before continuing the surgery. However, if the haemostasis of the ovarian pedicles or broad ligaments is a source of concern, the placement of a few forceps should suffice to provide temporary closure of the anterior portion of the laparotomy wound. Sectioning the Cervix [9]

Suturing the Anterior Portion of the Laparotomy Incision The prolapse of intestinal loops through the incision can cause significant heat and fluid loss, which can have very serious consequences, especially if the bitch is already suffering from deterioration in general status due to severe pyometria, for example. It is therefore advisable to suture the anterior portion of the laparotomy wound before continuing the surgery. However, if the haemostasis of the ovarian pedicles or broad ligaments is a source of concern, the placement of a few forceps should suffice to provide temporary closure of the anterior portion of the laparotomy wound. Sectioning the Cervix [9] – Ligating the Uterine Arteries and Veins (Figure 3). Once both uterine horns have been flipped back onto the pelvis, the uterine cervix is sectioned, following ligation of the uterine arteries and veins. The veins can be visualised passing on either side of the cervix. The arteries run under the veins in the musculosa of the cervix, which is why the haemostatic sutures should transfix the lateral walls of the cervix. However, if the uterine artery is perforated during ligation, a wider transfixion is needed, more caudal to the previous attempt. – Forcep Placement (Figure 3). Once both of the ligatures have been placed, the cervix is crushed at their level with an intestinal clamp. Another clamp is then placed just above the first and the contents of the uterus are pushed back towards the horns; two other clamps are placed in the same way above the 2nd clamp. The 2nd and 3rd clamps are removed, thus leaving a secretion-free zone.

ced, the cervix is crushed at their level with an intestinal clamp. Another clamp is then placed just above the first and the contents of the uterus are pushed back towards the horns; two other clamps are placed in the same way above the 2nd clamp. The 2nd and 3rd clamps are removed, thus leaving a secretion-free zone. – Sectioning the Cervix (Figure 3). Once both intestinal clamps have been placed, the anterior section of the cervix is performed; the cervix may be normal or pathological. Normal Cervix. The cervix is simply sectioned with a scalpel between the two clamps. Pathological Cervix. For pathological cervixes, the serosa is dissected just caudal to the clamp that is placed on the uterus; the serosa is then retracted caudally. The musculosa is then sectioned cranial to the intestinal clamp placed on the cervix; if the clamps have been placed correctly, no fluid should leak from the cut ends. Dealing with the Stump (Figure 3) – Small, Normal Cervix. The stump is simply replaced in the abdominal cavity. It is however advisable to suture it or bury it in a fold of omentum. – Pathological Cervix. The cut section of the musculosa, mucosa, is cauterised with an iodine-based solution, and then sutured in two phases: Septic Phase. For the septic phase, a simple continuous suture is made in the musculosa with VICRYL N.D dec. 3.

– Small, Normal Cervix. The stump is simply replaced in the abdominal cavity. It is however advisable to suture it or bury it in a fold of omentum. – Pathological Cervix. The cut section of the musculosa, mucosa, is cauterised with an iodine-based solution, and then sutured in two phases: Septic Phase. For the septic phase, a simple continuous suture is made in the musculosa with VICRYL N.D dec. 3. Aseptic Phase. The needle is changed and either a buried simple continuous suture is made with the serosa (sero-serous continuous suture), or the stump is enfolded in one of the broad ligaments, which is fixated with a suture in the bursa. The ligament will weld itself to the stump. Finally, the stump can be invaginated by burying it in the vagina, then placing a ligature a few centimetres behind the original section. However, invagination is practically impossible to perform in small dogs due to the small size of their genital tract. These suture procedures eliminate the risk of peritoneal infection, since the pathological secretions drain into the vagina. Suturing the Abdominal Wall The sutured stump is returned to the abdominal cavity and the abdominal wall is closed using “X”-shaped interrupted sutures with VICRYL N.D. Dec.4. If the subcutaneous connective tissue is very abundant, a simple continuous subcutaneous suture is performed using VICRYL N.D. Dec.3.

These suture procedures eliminate the risk of peritoneal infection, since the pathological secretions drain into the vagina. Suturing the Abdominal Wall The sutured stump is returned to the abdominal cavity and the abdominal wall is closed using “X”-shaped interrupted sutures with VICRYL N.D. Dec.4. If the subcutaneous connective tissue is very abundant, a simple continuous subcutaneous suture is performed using VICRYL N.D. Dec.3. Finally, the skin is sutured using simple interrupted sutures or mattress sutures with non-resorbable filament such as MONOSIN N.D. Dec.3. The wound is then disinfected with antiseptic solution and protected with a few swabs and an adhesive dressing. 3. Results 3.1. Surgical Variation Hysterectomy in the bitch via the linea alba is not very difficult. Nevertheless, it is sometimes necessary to perform the surgery via a vaginal approach rather than via the linea alba. 3.1.1. Hysterectomy via the Vaginal Approach Per-vaginal hysterectomy is performed in the event of uterine prolapse, if the latter cannot be reduced or if has been traumatised to such an extent that it cannot be replaced safely. The elective site for amputation is between the cervix and urinary meatus, in which case there are two different possible techniques, either with an elastic ligature, or by suturing. (a) Elastic Ligature An elastic band is placed between the cervix and the urinary meatus, the exeresis is then performed and the stump sutured by joining the internal and external segments with a perforating simple continuous suture.

The elective site for amputation is between the cervix and urinary meatus, in which case there are two different possible techniques, either with an elastic ligature, or by suturing. (a) Elastic Ligature An elastic band is placed between the cervix and the urinary meatus, the exeresis is then performed and the stump sutured by joining the internal and external segments with a perforating simple continuous suture. (b) Suture Firstly, an intestinal clamp is placed between the cervix and the urinary meatus to crush the pedicle, then, either a transfixing suture or overlapping mattress suture is placed. Once the sutures have been placed, the vagina is excised. The stump will be expelled within 15 days. 3.2. Postoperative Care Firstly, advise perioperative oxygenation if the surgical shock is very great. The animal is warmed, especially if the female was in poor condition prior to the procedure, she must be rolled in a blanket and placed in a heated kennel. Intravenous fluid therapy is administered with isotonic saline along with an injection of Vitamin C and corticosteroids. The bitch is then placed under antibiotic therapy for at least 5 days. The sutures are removed after 10 days. Any stagnant uterine secretions in the cervix and vagina will be eliminated in the days following and then cease completely. 3.3. Complications These can be classified as general or specific. 3.3.1. General Evisceration. Abdominal herniation. Suppuration from the cutaneous wound. Peritonitis.

The sutures are removed after 10 days. Any stagnant uterine secretions in the cervix and vagina will be eliminated in the days following and then cease completely. 3.3. Complications These can be classified as general or specific. 3.3.1. General Evisceration. Abdominal herniation. Suppuration from the cutaneous wound. Peritonitis. 3.3.2. Specific Haemorrhage: occurring during the intervention and continuing in the hours following. The latter represents one of the most common causes of the death of the animal. They can be situated at the level of the following. Ovarian pedicle: this is why it is not advisable to operate during oestrus, where the uterine arteries and veins are hypertrophied. Broad ligament: always check that there is no significant haemorrhage after rupture. Uterine cervix: treatment involves a blood transfusion and repeat surgery to ligate the bleeding vessel. Abscess: These form especially at the level of the anterior straight of the pelvis and lumbar, when the cervix is not correctly treated, resulting in pain during defecation, vomiting, and finally an occlusive syndrome. The diagnosis is established via an exploratory laparotomy. Treatment is surgical, and may even necessitate nephrectomy or enterectomy. Abdominal adherences: these result from a localised peritonitis and cause an occlusive syndrome. Urinary incontinence: around 20% of spayed bitches are affected, especially large breeds.

Abscess: These form especially at the level of the anterior straight of the pelvis and lumbar, when the cervix is not correctly treated, resulting in pain during defecation, vomiting, and finally an occlusive syndrome. The diagnosis is established via an exploratory laparotomy. Treatment is surgical, and may even necessitate nephrectomy or enterectomy. Abdominal adherences: these result from a localised peritonitis and cause an occlusive syndrome. Urinary incontinence: around 20% of spayed bitches are affected, especially large breeds. Recurrence of metritis: occurs in the month or years following the intervention, and a serous, purulent, or haemorrhagic vulvular discharge may occur. This complication occurs in bitches who have undergone hysterectomy alone, and which did not resolve the problem given that part of the genital apparatus remains in place: vagina, cervix, and occasionally a short section of uterus in front of the cervix, and when the oestrus cycle is abnormal, the uterine mucosa is not the only organ to suffer the consequences, the cervix and to a lesser extent the vagina also react. There are two possible solutions.

e genital apparatus remains in place: vagina, cervix, and occasionally a short section of uterus in front of the cervix, and when the oestrus cycle is abnormal, the uterine mucosa is not the only organ to suffer the consequences, the cervix and to a lesser extent the vagina also react. There are two possible solutions. Ablation of the cervix: repeat surgery with placement of a clove hitch as far as possible from the cervix on the vagina. Then make a cut a few millimetres from the ligature. The stump is then disinfected with an iodine solution and sutured with a perforating simple continuous suture with VICRYL Dec. 3, and returned to the pelvic cavity. The disadvantage of this method is that the ovaries remain in place and the posterior portion of the vaginal mucosa may therefore continue to secrete abnormally for a few months, or even for up to a year later, which can result in the emission of drops of pus at the lower commissure of the vulva. This is why an ovariectomy is preferable and necessary (Figure 4).

ovaries remain in place and the posterior portion of the vaginal mucosa may therefore continue to secrete abnormally for a few months, or even for up to a year later, which can result in the emission of drops of pus at the lower commissure of the vulva. This is why an ovariectomy is preferable and necessary (Figure 4). Return to oestrus: the differential diagnosis should include cystitis, vaginitis, cervicitis, and inflammation or infection of the anal glands, which may be mistaken for signs of oestrus by the owners, which can be detected using a vaginal swab, and/or serum progesterone assay. If the bitch is confirmed as being in oestrus, there are two possible explanations: a residual fragment of the ovary, which was overlooked at the time of the ovariohysterectomy, or the secretion of oestrogens from the corticoadrenal glands due to hypophyseal dysfunction. Repeat surgery is therefore advisable, if the latter fails or if the surgeon is confident that no ovarian fragments have been overlooked then hormonal treatment can be started [6].

looked at the time of the ovariohysterectomy, or the secretion of oestrogens from the corticoadrenal glands due to hypophyseal dysfunction. Repeat surgery is therefore advisable, if the latter fails or if the surgeon is confident that no ovarian fragments have been overlooked then hormonal treatment can be started [6]. Bitches who have undergone surgery for pyometria may present with thrombocytopenia: in the days following the procedure, the bitch presents with a marked tendency to haemorrhage. The mucosae are pale, the pulse weak, the bitch is hypothermic (36°C), and petechiae appear on the gingival and labial mucosae. Subcutaneous oozing is seen at the abdominal incision, and an angry red patch covering the entire caudo-ventral abdominal zone. Occasionally, a large quantity of blood clots is discharged from the vulva. Haematology reveals a significant reduction in circulating platelets, to the order of 3,500 rather than 300,000–450,000/mm3. The red blood cell count is also markedly reduced to around 120,000 instead of 6,000,000/mm3. The tendency to haemorrhage increases over time and the edges of the wound disunite within 3-4 days, the animal is very pale, comatose, and death follows. 4. Discussion The current success rate is close to 95%, whilst several decades ago failures were to the order of 50%.

Bitches who have undergone surgery for pyometria may present with thrombocytopenia: in the days following the procedure, the bitch presents with a marked tendency to haemorrhage. The mucosae are pale, the pulse weak, the bitch is hypothermic (36°C), and petechiae appear on the gingival and labial mucosae. Subcutaneous oozing is seen at the abdominal incision, and an angry red patch covering the entire caudo-ventral abdominal zone. Occasionally, a large quantity of blood clots is discharged from the vulva. Haematology reveals a significant reduction in circulating platelets, to the order of 3,500 rather than 300,000–450,000/mm3. The red blood cell count is also markedly reduced to around 120,000 instead of 6,000,000/mm3. The tendency to haemorrhage increases over time and the edges of the wound disunite within 3-4 days, the animal is very pale, comatose, and death follows. 4. Discussion The current success rate is close to 95%, whilst several decades ago failures were to the order of 50%. 5. Conclusion Ovariohysterectomy is the only effective treatment for pyometria and it is a radical treatment for postoestrus metritis when it recurs following failure of medical treatment. For the procedure to have the best chances of success, it is important to perform it on a bitch in good general condition. It is also important to remember that after an ovariectomy bitches tend to put on weight, 1 kg for every 10 kg after 90 days; the metabolism of the bitch falls from 37 to 33 Kcal/day, it is therefore important to reduce the daily ration by 10%.

5. Conclusion Ovariohysterectomy is the only effective treatment for pyometria and it is a radical treatment for postoestrus metritis when it recurs following failure of medical treatment. For the procedure to have the best chances of success, it is important to perform it on a bitch in good general condition. It is also important to remember that after an ovariectomy bitches tend to put on weight, 1 kg for every 10 kg after 90 days; the metabolism of the bitch falls from 37 to 33 Kcal/day, it is therefore important to reduce the daily ration by 10%. Figure 1 Arteries and veins of the genital apparatus of the bitch (Seen side left). Figure 2 Uterine irrigation. Figure 3 Section of the pathological cervix. Figure 4 Complication: Pathological cervix amputation.

1. Introduction Hysterectomy is the ultimate treatment for women suffering from symptomatic fibroids, abnormal uterine bleeding and uterine malignancy and is one of the most frequent performed surgical procedures [1, 2]. There is no universal agreement about the optimal method of hysterectomy—abdominal, laparoscopic, or vaginal—and there is a question whether the cervix should be removed as a routine part of the hysterectomy. The world's first successful supracervical abdominal hysterectomy (SAH) was performed in 1853 by Gilman Kimball in USA. Since then, the advantages and disadvantages of supracervical versus total hysterectomy technique have been discussed, with variable enthusiasm in different time periods and between countries. More recently, there has been a swing back to supracervical, with marked geographic variations [3–6]. In Scandinavia, the ratio of supracervical to total hysterectomy is traditionally high. At our department in Oslo, Norway, supracervical hysterectomy is the recommended procedure for women with benign conditions requiring hysterectomy and with no previous history of cervical dysplasia. Although laparoscopic supracervical hysterectomy (LSH) has gradually replaced abdominal hysterectomy, SAH is still performed in women where laparoscopic or vaginal approach is not feasible, mainly due to significant enlarged uterus [7].

nditions requiring hysterectomy and with no previous history of cervical dysplasia. Although laparoscopic supracervical hysterectomy (LSH) has gradually replaced abdominal hysterectomy, SAH is still performed in women where laparoscopic or vaginal approach is not feasible, mainly due to significant enlarged uterus [7]. Opponents of supracervical hysterectomy, either it is performed open or by a laparoscopic approach, often seem to be concerned with the risk of cervical stump symptoms such as vaginal bleeding and pelvic pain following the hysterectomy, causing patient distress and eventually repeated surgery. In a previous publication reporting long-term outcomes after LSH, we found that although vaginal bleeding and pelvic pain are frequently observed following LSH, the patient satisfaction following the procedure was high [8]. In this study, we wanted to evaluate whether the occurrence of vaginal bleeding and persistent pelvic pain are consequences of the cervix-sparing technique, or related to the surgical approach. Here we present the long-term outcomes in terms of cervical stump symptoms, women acceptability of such symptoms, and overall patient satisfaction after both abdominal and laparoscopic supracervical hysterectomies performed during the same time period.

the cervix-sparing technique, or related to the surgical approach. Here we present the long-term outcomes in terms of cervical stump symptoms, women acceptability of such symptoms, and overall patient satisfaction after both abdominal and laparoscopic supracervical hysterectomies performed during the same time period. 2. Material and Methods Following ethical approval, all women who were treated by LSH and SAH on the basis of a benign condition during 2004 and 2005 at the Department of Gynaecology, Oslo University Hospital Ullevål, Oslo, Norway, were sent a questionnaire. Nonresponders were sent a reminder letter four weeks following the original mail out. Firstly, the LSH-treated group were contacted between 12 and 36 months after surgery, while the SAH-treated group were contacted somewhat later, between 17 and 41 months after surgery. The questionnaire was divided into two sections. The first section contained questions about reasons for having the hysterectomy, as well as menstrual pain and bleeding prior to the hysterectomy. In the second section, questions about experiences of menstrual bleeding and pain following the hysterectomy, any further treatments for such symptoms, any new symptoms related to the hysterectomy and overall satisfaction with the surgery were included. Standard 10-point visual analogue scales (VAS) were used to measure pain intensity and the extent to which bleeding was bothersome. The remaining questions were either dichotomous yes or no responses, or they provided women with either four or five categories of responses to choose from.

with the surgery were included. Standard 10-point visual analogue scales (VAS) were used to measure pain intensity and the extent to which bleeding was bothersome. The remaining questions were either dichotomous yes or no responses, or they provided women with either four or five categories of responses to choose from. Data were analysed using SPSS for Windows (SPSS 14.0, SPSS, Inc, Chicago, IL, USA). Normally distributed continuous data from two groups of women were analysed using a two-sided Independent Samples Student t-test and when paired, the Paired Samples t-test. Categorical data were analysed using Pearson Chi-Square. Forward stepwise logistic regression analysis was used to calculate the adjusted odds ratios for continued menstrual/cyclical pain, continued vaginal bleeding, and patient satisfaction. 3. Results Altogether, 134 women were identified as having had an SAH and 315 women an LSH during 2004 and 2005 and were therefore sent a questionnaire. Twelve women could not be contacted (five women in SAH group and seven women in LSH group); nine had moved to unknown addresses, and three had died from nongynecological conditions. The response rate in the two groups of women was 82% (SAH group : 106/129) and 78% (LSH group : 240/308), respectively.

t a questionnaire. Twelve women could not be contacted (five women in SAH group and seven women in LSH group); nine had moved to unknown addresses, and three had died from nongynecological conditions. The response rate in the two groups of women was 82% (SAH group : 106/129) and 78% (LSH group : 240/308), respectively. Out of all 449 procedures, 228 (51%) were performed in 2004 and 221 (49%) in 2005. The total response rates for 2004 and 2005 were 75% and 81%, respectively. Mean age of the responders was 48 years (SD 7) in the SAH group and 45 years (SD 6) in the LSH group. There were no significant differences between responders and nonresponders in terms of age and incidence of repeated surgery in either of the two groups of women. 3.1. Reasons for Having the Hysterectomy Most women in both treatment groups (59% in SAH group, 70% in LSH group) stated two or more reasons for having the hysterectomy, the dominating reasons being fibroids (86% versus 68%) and/or heavy bleeding (50% versus 67%). Among women having SAH, 22% stated pain and 4% endometriosis as a reason for the hysterectomy. Respectively, in the LSH group, 46% of women stated pain and 16% endometriosis as a reason for the hysterectomy.

the hysterectomy, the dominating reasons being fibroids (86% versus 68%) and/or heavy bleeding (50% versus 67%). Among women having SAH, 22% stated pain and 4% endometriosis as a reason for the hysterectomy. Respectively, in the LSH group, 46% of women stated pain and 16% endometriosis as a reason for the hysterectomy. 3.2. Menstrual Bleeding Nineteen women (5%) had reached the menopause before the hysterectomy, and three women had medically induced amenorrhoea. Self-reported preoperative menstrual data were available for 96 women in the SAH group and 220 women in the LSH group. The majority in both treatment groups reported their preoperative periodic bleedings to be very heavy (38% in SAH group versus 43% in LSH group) or heavy (18% versus 25%). The remaining women reported their preoperative periodic bleedings as moderately heavy (21% versus 18%), normal (19% versus 11%), or minimal (5% versus 3%).

rity in both treatment groups reported their preoperative periodic bleedings to be very heavy (38% in SAH group versus 43% in LSH group) or heavy (18% versus 25%). The remaining women reported their preoperative periodic bleedings as moderately heavy (21% versus 18%), normal (19% versus 11%), or minimal (5% versus 3%). Out of the responders, 19 women (17%) in the SAH group and 57 women (24%) in the LSH group reported experiencing vaginal bleedings after the hysterectomy. When comparing the occurrence of vaginal bleeding in the two treatment groups, the difference was not statistically significant (P > .05). Among women who underwent SAH, ten women (9%) reported to experience regular periodic bleedings, and nine (8%) experienced irregular vaginal bleedings. In the LSH group, 16 women (7%) experienced regular periodic bleedings, 25 women (10%) irregular bleeding patterns, seven women (3%) bleedings in relation to sexual activity, and the remaining nine women (4%) a combination of regular and irregular bleedings which also were related to sexual activity. In the SAH group, no significant age-related differences were found (OR 0.95, 95% CI: 0.87, 1.05), and reason for hysterectomy appeared not to influence the occurrence of persistent vaginal bleeding (Table 1). Adjusted odds ratios (OR) revealed that older women who had been treated by LSH were less likely to experience persistent vaginal bleeding (OR 0.89, 95% CI: 0.83, 0.95). Furthermore, vaginal bleedings after the procedure were less likely to be reported by women who had LSH because of pain (OR 0.41, 95% CI: 0.20, 0.85) or fibroids (OR 0.47, 95% CI: 0.23, 0.93), but more likely to be reported in women with heavy periods prior to surgery (OR 4.07, 95% CI: 1.32, 12.57) (Table 1).

3, 0.95). Furthermore, vaginal bleedings after the procedure were less likely to be reported by women who had LSH because of pain (OR 0.41, 95% CI: 0.20, 0.85) or fibroids (OR 0.47, 95% CI: 0.23, 0.93), but more likely to be reported in women with heavy periods prior to surgery (OR 4.07, 95% CI: 1.32, 12.57) (Table 1). All women who experienced persistent vaginal bleedings reported the amount of bleeding to be minimal (88% in SAH group, 90% in LSH group), or less than their normal preoperative periodic bleeding (12% versus 10%). The mean degree of bother caused by vaginal bleedings after the hysterectomy, scored on a 10-point VAS, was 1.7 (SD 2.7) in the SAH group and 1.1 (SD 2.0) in the LSH group, respectively. 3.3. Pain Out of the responders, 60 women (55%) in the SAH group and 178 women (74%) in the LSH group suffered from menstrual pain before the hysterectomy. The preoperative mean pain score measured by a 10-point VAS was 5.3 (SD 2.5) in the SAH group and 6.8 (SD 2.1) in the LSH group, respectively. Twenty-three women (21%) in the SAH group and 89 women (37%) in the LSH group reported continued menstrual/periodic pain following their hysterectomy. The mean pain score after hysterectomy was significantly less than before surgery after both procedures (SAH: mean pain score 2.3, SD 1.9, mean pain reduction 2.5, 95% CI: 1.3, 3.7, P < .01; LSH: mean pain score 3.5, SD 2.2, mean pain reduction 3.3, 95% CI: 2.7, 3.9, P < .01).

eriodic pain following their hysterectomy. The mean pain score after hysterectomy was significantly less than before surgery after both procedures (SAH: mean pain score 2.3, SD 1.9, mean pain reduction 2.5, 95% CI: 1.3, 3.7, P < .01; LSH: mean pain score 3.5, SD 2.2, mean pain reduction 3.3, 95% CI: 2.7, 3.9, P < .01). Whilst all women reported a significant decrease of pain intensity experienced after the hysterectomy, women having a hysterectomy because of pain and/or endometriosis reported significant higher levels of remaining menstrual/cyclical pain after both procedures (SAH: mean pain score 3.4, SD 2.1; LSH: mean pain score 3.5, SD 2.9), compared with women who did not report endometriosis and/or pain as a reason for the hysterectomy (SAH: mean pain score 1.9, SD 1.7; LSH: mean pain score = 0.9, SD 1.7), P = .05 and P < .001, respectively (Table 2). The adjusted odds ratios (OR) revealed that increased intensity of preoperative pain resulted in a greater chance of experiencing pain after both procedures (SAH: OR 3.9, 95% CI: 2.0, 5.7; LSH: OR 1.1, 95% CI: 1.0, 1.3).

ore 1.9, SD 1.7; LSH: mean pain score = 0.9, SD 1.7), P = .05 and P < .001, respectively (Table 2). The adjusted odds ratios (OR) revealed that increased intensity of preoperative pain resulted in a greater chance of experiencing pain after both procedures (SAH: OR 3.9, 95% CI: 2.0, 5.7; LSH: OR 1.1, 95% CI: 1.0, 1.3). 3.4. Repeated Surgery In total, 30 out of 449 women (7%) had a further related surgery after the hysterectomy (6% after SAH, 7% after LSH), the most common procedures being performed because of postoperative bleeding, hematomas with secondary infection and adhesions (Table 3). Three of the seven women who went on to have their cervix removed had their original hysterectomy because of endometriosis. Out of the women who reported experiencing continued menstrual bleeding after the hysterectomy, 11% and 7% underwent repeated surgery after SAH and LSH, respectively. 3.5. New Symptoms Following Hysterectomy Out of the responders, 81 women (26% in SAH group, 22% in LSH group) reported experiencing new symptoms following their hysterectomy. Although some women reported to suffer from different forms of pain (pelvic pain, dyspareunia, pain related to the scar) after the hysterectomy, the majority of symptoms reported by the women appeared to be related to the menopause (vasomotoric symptoms, vaginal dryness, reduced libido, gained weight) (Table 4).

ctomy. Although some women reported to suffer from different forms of pain (pelvic pain, dyspareunia, pain related to the scar) after the hysterectomy, the majority of symptoms reported by the women appeared to be related to the menopause (vasomotoric symptoms, vaginal dryness, reduced libido, gained weight) (Table 4). 3.5.1. Satisfaction with Surgery Almost all women reported being satisfied (SAH: 28%, LSH: 20%) or very satisfied (SAH: 58%, LSH: 70%) with the hysterectomy. After both procedures, women who reported being satisfied with the preoperative information were more likely to report being very satisfied with the hysterectomy (SAH: OR 5.5, 95% CI: 2.4, 13.0; LSH: OR 3.3, 95% CI: 1.8, 6.2). No significant difference regarding degree of total satisfaction was found comparing women who experienced persistent vaginal bleeding and women who had no vaginal bleeding after the hysterectomy. In both groups, women who reported having a new symptom following their hysterectomy, were less likely to report being very satisfied (SAH: OR 0.2, 95% CI: 0.1, 0.6; LSH: OR 0.2, 95% CI: 0.1, 0.4). No significant difference regarding degree of total satisfaction was found comparing women who had repeated surgery following SAH or not, whereas women who had repeated surgery following LSH were less likely to report being very satisfied (OR 0.1, 95% CI: 0.1, 0.5).

.2, 95% CI: 0.1, 0.6; LSH: OR 0.2, 95% CI: 0.1, 0.4). No significant difference regarding degree of total satisfaction was found comparing women who had repeated surgery following SAH or not, whereas women who had repeated surgery following LSH were less likely to report being very satisfied (OR 0.1, 95% CI: 0.1, 0.5). 4. Discussion This study reports the occurrence of long-term outcomes following both LSH and SAH in women with benign conditions, as well as the impact these outcomes have on the women's experiences. The relatively large sample size and an excellent response rate represent strengths of the study. Although all procedures were performed at the same department during the same time period, comparisons between the outcomes after the two supracervical procedures should be interpreted with care, as the women were selected and not randomised to the treatment groups. As the benefits of a laparoscopic compared with an abdominal approach are well documented, it would be unethical to randomise women to LSH or SAH [9]. Given these limitations, our study enables a comparison of long-term outcomes after cervix-sparing surgery with both laparoscopic and abdominal techniques. Other limitations of the study are that the data were collected retrospectively and that the questionnaires were sent to the SAH group after the results of the LSH group were known. Furthermore, we were unable to compare the results of supracervical hysterectomy to those of total hysterectomy irrespective of surgical approach.

mitations of the study are that the data were collected retrospectively and that the questionnaires were sent to the SAH group after the results of the LSH group were known. Furthermore, we were unable to compare the results of supracervical hysterectomy to those of total hysterectomy irrespective of surgical approach. The results of our study demonstrate that although cervical stump symptoms are relatively common following the two surgical procedures, the overall patient satisfaction is high. The occurrence of vaginal bleedings after SAH is in previous studies reported to be 7%–20% [10–12]. Similarly, the occurrence of persistent vaginal bleedings following LSH is reported in the wide range of 0%–25% [13–17]. The occurrence of vaginal bleedings after both procedures in our study (18% after SAH and 24% after LSH) were relatively high compared to these previous reports. This may partly be explained by different definition of vaginal bleeding in the different studies. We included both regular and irregular bleedings as well as bleeding related to sexual activity, whereas some previous reports have only reported the occurrence of vaginal bleeding on a regular monthly basis. When we included only regular bleedings in our analyses, the proportion decreased to 8% in both study groups. In spite of the high number of women with continued vaginal bleedings, the women rarely found the bleeding to be bothersome, and it did not affect their overall satisfaction.

leeding on a regular monthly basis. When we included only regular bleedings in our analyses, the proportion decreased to 8% in both study groups. In spite of the high number of women with continued vaginal bleedings, the women rarely found the bleeding to be bothersome, and it did not affect their overall satisfaction. Insufficient surgical experience and skill, resulting in amputation above the level of the internal cervical ostium, have been suggested as possible causes of the high occurrence of vaginal bleedings following supracervical hysterectomy [13]. Whether other mechanisms, like more meticulous destruction of any remaining endometrium in the spared cervix, could reduce the occurrence of persistent vaginal bleedings after surgery remains to be demonstrated.

ed as possible causes of the high occurrence of vaginal bleedings following supracervical hysterectomy [13]. Whether other mechanisms, like more meticulous destruction of any remaining endometrium in the spared cervix, could reduce the occurrence of persistent vaginal bleedings after surgery remains to be demonstrated. The proportion of women who reported suffering from preoperative menstrual pain in our study was relatively high, 74% in the LSH group and 57% in the SAH group. The pain score after hysterectomy was significantly less than before surgery after both procedures, with less pain in the LSH group, possibly related to less adhesion formation after surgery. Although the pelvic pain was reduced after the procedure, a relatively large proportion of women reported continued pain. Some may argue that women with endometriosis and/or pelvic pain would have a more favourable outcome after total hysterectomy compared with a supracervical procedure. However, to our knowledge, no evidence from randomised control trials suggests that total hysterectomy in women with preoperative pain results in greater pain reduction. Furthermore, an eventual effect of a subsequent removal of the cervix following a supracervical procedure due to pain does not appear to have been reported.

r, to our knowledge, no evidence from randomised control trials suggests that total hysterectomy in women with preoperative pain results in greater pain reduction. Furthermore, an eventual effect of a subsequent removal of the cervix following a supracervical procedure due to pain does not appear to have been reported. The results indicate that women are overall satisfied regarding the outcome following both supracervical procedures. However, as cervical stump symptoms appear to be relatively common, it is important to inform women preoperatively regarding the risk of persistent menstrual bleedings and/or pain. In our study, women who were not satisfied regarding information, reported a significantly lower degree of total patient satisfaction, which illustrates the importance of proper and adequate preoperative information. 5. Conclusion Persistent vaginal bleedings are relatively common following supracervical hysterectomy. The bleedings are, however, reported by women as minimal and rarely bothersome, and the patient satisfaction following both supracervical procedures is high. All women should have proper preoperative information. Women with pelvic pain and/or endometriosis should in addition be informed of the possibility of persistent pelvic pain and the increased risk of repeated surgery following supracervical hysterectomy. Table 1 Adjusted risk estimates for experiencing persistent vaginal bleeding following abdominal and laparoscopic supracervical hysterectomy.

5. Conclusion Persistent vaginal bleedings are relatively common following supracervical hysterectomy. The bleedings are, however, reported by women as minimal and rarely bothersome, and the patient satisfaction following both supracervical procedures is high. All women should have proper preoperative information. Women with pelvic pain and/or endometriosis should in addition be informed of the possibility of persistent pelvic pain and the increased risk of repeated surgery following supracervical hysterectomy. Table 1 Adjusted risk estimates for experiencing persistent vaginal bleeding following abdominal and laparoscopic supracervical hysterectomy. Reason for hysterectomy Number of women with vaginal bleeding Odds ratio (95% CI) Supracervical abdominal hysterectomy Fibroids Yes 16 0.66 (0.16, 2.71)* No 3 Heavy bleeding Yes 11 1.28 (0.49, 3.35)* No 8 Pain and/or endometriosis Yes 5 1.37 (0.43, 4.32)* No 14 Supracervical laparoscopic hysterectomy Fibroids Yes 33 0.47 (0.23, 0.93)* No 24 Heavy bleeding Yes 41 4.07 (1.32, 12.57)* No 16 Pain and/or endometriosis Yes 23 0.41 (0.20, 0.85)* No 34 *Forward stepwise logistic regression analysis. Table 2 Pain intensity prior to and after surgery by reasons for hysterectomy. Reason for hysterectomy N Mean pain score prior to surgery* (SD) Mean pain score after surgery (SD) Mean diff. in pain scores (SD) [95% CI] P-value Supracervical abdominal hysterectomy Fibroids 91 5.4 (2.5) 2.3 (1.9) 2.6 (2.8) [1.4, 3.8] <.01** Heavy bleeding 52 5.6 (2.6) 2.3 (2.0) 2.5 (3.2) [0.7, 4.3] <.01** Endometriosis/pain 27 6.7 (1.9) 3.4 (2.1) 3.0 (3.1) [0.1, 5.9] .04**

(SD) Mean pain score after surgery (SD) Mean diff. in pain scores (SD) [95% CI] P-value Supracervical abdominal hysterectomy Fibroids 91 5.4 (2.5) 2.3 (1.9) 2.6 (2.8) [1.4, 3.8] <.01** Heavy bleeding 52 5.6 (2.6) 2.3 (2.0) 2.5 (3.2) [0.7, 4.3] <.01** Endometriosis/pain 27 6.7 (1.9) 3.4 (2.1) 3.0 (3.1) [0.1, 5.9] .04** Supracervical laparoscopic hysterectomy Fibroids 151 4.8 (3.4) 1.1 (1.9) 3.7 (3.3) [3.2, 4.2] <.001** Heavy bleeding 151 5.6 (3.2) 1.3 (2.0) 4.4 (3.4) [3.2, 4.9] <.001** Endometriosis 37 7.5 (2.5) 3.5 (2.9) 4.0 (4.0) [2.7, 5.3] <.001** Pain 104 7.1 (2.6) 1.9 (2.5) 5.3 (3.5) [4.6, 6.0] <.001* *10-point VAS,**Paired Samples t-test. Table 3 Surgical procedures performed after the hysterectomy. Surgical procedures after hysterectomy Supracervical abdominal hysterectomy N* Supracervical laparoscopic hysterectomy N** Resurgery within 24 hours due to bleeding 3 — Laparoscopic adhesiolysis — 6 Laparoscopic extirpation of cervix uteri — 7 Drainage of postoperative hematoma/abscess 2 1 Bowel resection due to postoperative peritonitis 1 1 BSOE and removal of cervix uteri (sarcoma uteri) — 1 Scar correction 1 3 Umbilical hernia repair — 1 Tension-free vaginal tape procedure (TVT) — 2 Cystoscopy 1 — Total 8 (6%) 22 (7%) *Total number of women having supracervical abdominal hysterectomy: 134. **Total number of women having supracervical laparoscopic hysterectomy: 315. Table 4 New symptoms reported following the hysterectomy.

Surgical procedures after hysterectomy Supracervical abdominal hysterectomy N* Supracervical laparoscopic hysterectomy N** Resurgery within 24 hours due to bleeding 3 — Laparoscopic adhesiolysis — 6 Laparoscopic extirpation of cervix uteri — 7 Drainage of postoperative hematoma/abscess 2 1 Bowel resection due to postoperative peritonitis 1 1 BSOE and removal of cervix uteri (sarcoma uteri) — 1 Scar correction 1 3 Umbilical hernia repair — 1 Tension-free vaginal tape procedure (TVT) — 2 Cystoscopy 1 — Total 8 (6%) 22 (7%) *Total number of women having supracervical abdominal hysterectomy: 134. **Total number of women having supracervical laparoscopic hysterectomy: 315. Table 4 New symptoms reported following the hysterectomy. New symptoms Supracervical abdominal hysterectomy N (%) Supracervical laparoscopic hysterectomy N (%) Changes related to urination and/or defecation 4 (3.8) 9 (3.8) Vasomotor symptoms — 5 (2.1) Vaginal discharge — 5 (2.1) Vaginal dryness and/or dyspareunia 3 (2.8) 5 (2.1) Pelvic pain 3 (2.8) 5 (2.1) Problems related to the scar (pain and/or cosmetics) 6 (5.7) — Other (depression, gained weight, reduced libido, 12 (9.0) 24 (10.0) cystitis, candida infection, fear of cervical cancer)

1. Introduction Prematurity is the leading cause of neonatal morbidity and mortality [1, 2]. Another important cause of perinatal morbidity and mortality is intrauterine growth restriction (IUGR), a condition in which the fetus is undernourished for gestational age [3]. Normally, IUGR is present in only a small percentage of deliveries, but an increased frequency has been observed among women who go into preterm labor followed by premature delivery [4]. Late-preterm birth is defined as birth between 34 weeks and 36 6/7 weeks of gestation [5]. During the past decade, the proportion of all U.S. births defined as late-preterm births has increased by 16% [6]. The overall rate of preterm births in the United States increased from 10.9% in 1990 to 12.8% in 2007, an increase of 16.6% [7]. This increase was mainly due to an increase in late-preterm births. In Brazil, approximately 88% of the 188,223 pre-term births recorded in 2005 occurred at an gestational age above 32 weeks [8]. The prevalence of IUGR is high in high-risk pregnancies. As a consequence, this condition is common among elective preterm deliveries and is therefore associated with late prematurity, with the observation of a recent increase in the incidence of these electively delivered late-preterm infants [4]. There are conflicting findings in the literature regarding outcomes of preterm infants with IUGR and appropriate-for-gestational-age (AGA) infants. Neonatal morbidity and mortality have been reported to be decreased [9–11], unchanged [12, 13] and increased [14–18] in IUGR infants compared to AGA infants.

The prevalence of IUGR is high in high-risk pregnancies. As a consequence, this condition is common among elective preterm deliveries and is therefore associated with late prematurity, with the observation of a recent increase in the incidence of these electively delivered late-preterm infants [4]. There are conflicting findings in the literature regarding outcomes of preterm infants with IUGR and appropriate-for-gestational-age (AGA) infants. Neonatal morbidity and mortality have been reported to be decreased [9–11], unchanged [12, 13] and increased [14–18] in IUGR infants compared to AGA infants. The aim of this study was to compare neonatal morbidity and mortality between late-preterm IUGR and AGA infants of the same gestational age. 2. Methods We retrospectively analyzed 86 singleton pregnancies, including 50 pregnancies involving infants with a birth weight of or below the 10th percentile (IUGR) delivered between 34 weeks and 36 6/7 weeks of gestation due to maternal and/or fetal indications. The control group consisted of 36 singleton pregnancies with spontaneous preterm delivery at the same gestational age, in which the birth weight ranged from the 11th to 89th percentile (AGA). The study was performed between 2005 and 2007. Birth weight percentiles were based on the standard growth curve of Alexander et al. [19]. Pregnancies complicated by diabetes (preexisting or gestational) and premature membrane rupture, pregnancies with fetal anomalies and pregnancies with unknown or conflicting dating criteria were excluded.

2. Methods We retrospectively analyzed 86 singleton pregnancies, including 50 pregnancies involving infants with a birth weight of or below the 10th percentile (IUGR) delivered between 34 weeks and 36 6/7 weeks of gestation due to maternal and/or fetal indications. The control group consisted of 36 singleton pregnancies with spontaneous preterm delivery at the same gestational age, in which the birth weight ranged from the 11th to 89th percentile (AGA). The study was performed between 2005 and 2007. Birth weight percentiles were based on the standard growth curve of Alexander et al. [19]. Pregnancies complicated by diabetes (preexisting or gestational) and premature membrane rupture, pregnancies with fetal anomalies and pregnancies with unknown or conflicting dating criteria were excluded. Maternal characteristics included age, preexisting medical problems and pregnancy complications. Delivery characteristics included gestational age at delivery, birth weight, route of delivery, indication of elective birth, and Apgar scores. Neonatal data included death, transient tachypnea of the newborn (TTN), neonatal sepsis, intraventricular hemorrhage (IVH), hypoglycemia, jaundice, total number of days the infant was in the neonatal intensive care unit (NICU), and length of hospital stay. Gestational age at delivery was defined based on the mother's last menstrual period and was confirmed by early ultrasound examination.

neonatal sepsis, intraventricular hemorrhage (IVH), hypoglycemia, jaundice, total number of days the infant was in the neonatal intensive care unit (NICU), and length of hospital stay. Gestational age at delivery was defined based on the mother's last menstrual period and was confirmed by early ultrasound examination. Preexisting medical problems included hypertensive disorders (chronic hypertension and pre-eclampsia), heart diseases, systemic lupus erythematosus and others (pulmonary disease, hepatitis, thrombophilia, anemia, etc.). Antepartum complications included oligohydramnios and fetal distress. Possible signs of fetal distress were a constant decrease in fetal heart rate variability, the occurrence of late or variable decelerations upon cardiotocography, or a high systolic/diastolic ratio in the umbilical artery. Amniotic fluid volume was estimated during the evaluation of the fetal biophysical profile. Neonatal acidosis was defined as an arterial umbilical cord pH less than 7.2 [20].

variability, the occurrence of late or variable decelerations upon cardiotocography, or a high systolic/diastolic ratio in the umbilical artery. Amniotic fluid volume was estimated during the evaluation of the fetal biophysical profile. Neonatal acidosis was defined as an arterial umbilical cord pH less than 7.2 [20]. Diagnostic criteria for each neonatal problem are applied concurrently by neonatologists as follows: (1) TTN: clinical and radiographic features identified during the first hours of life, followed by characteristic resolution during the subsequent 24–48 hours; (2) neonatal sepsis: positive blood culture and clinical manifestations, or clinical manifestations, radiologic findings and laboratory indicators; (3) IVH: identified by serial cranial ultrasonography (all infants have head US); (4) hypoglycemia: blood glucose level below 40 mg/dL; (5) hyperbilirubinemia; (6) neonatal thrombocytopenia: platelet count less than 150.000/μL (150 × 109/L); (7) apnea of prematurity: prolonged respiratory pause (20 s or longer) accompanied by cyanosis, pallor or bradycardia. The discharge criteria for preterm infants included weight > 2 kg and good suction upon breast-feeding accompanied by adequate weight gain. Categorical data were compared between IUGR and AGA pregnancies by X2 analysis and Fisher's exact test. Ordinal measures were compared using the Kruskal-Wallis test. IUGR was considered to be significantly related to outcome when P < .005.

The discharge criteria for preterm infants included weight > 2 kg and good suction upon breast-feeding accompanied by adequate weight gain. Categorical data were compared between IUGR and AGA pregnancies by X2 analysis and Fisher's exact test. Ordinal measures were compared using the Kruskal-Wallis test. IUGR was considered to be significantly related to outcome when P < .005. 3. Results Of the 86 neonates included in the study, 50 belonged to the IUGR group and 36 to the AGA group. There was no significant difference in maternal age which ranged from 16 to 45 years (mean ± standard deviation: 25.1 ± 5.5 years) (P > .05). There was a predominance of white women (66.3%, n = 57), with no significant difference between groups. Among mothers of the IUGR group, 39 (78%) presented some underlying disease or obstetric complication in addition to IUGR, whereas 11 (22%) did not. Hypertensive syndromes were the most frequent condition and were observed in 24 (48%) women of the IUGR group. Heart disease was observed in 5 (10%) mothers of this group, systemic lupus erythematosus in 4 (8%), and other underlying diseases in 6 (12%) (pulmonary disease, hepatitis, thrombophilia, anemia, etc.) (Table 1).

ertensive syndromes were the most frequent condition and were observed in 24 (48%) women of the IUGR group. Heart disease was observed in 5 (10%) mothers of this group, systemic lupus erythematosus in 4 (8%), and other underlying diseases in 6 (12%) (pulmonary disease, hepatitis, thrombophilia, anemia, etc.) (Table 1). Gestational age at delivery ranged from 34 to 36.9 weeks (mean ± standard deviation: 35.5 ± 0.7, median: 35.6 weeks) and did not differ between groups. Preterm induction and preterm cesarean delivery were observed in 39 (78%) women of the IUGR group, whereas 11 (22%) patients went into spontaneous preterm labor. The indications for elective preterm delivery in the 39 patients of the IUGR group included oligohydramnios in 20 cases (51.3%), severe maternal disease in 8 (20.5%), presence of fetal maturity in 2 (5.1%), and abnormalities upon cardiotocography, fetal biophysical profile or umbilical artery Doppler in 9 (23.1%) (Table 1). There was a significant difference in mean birth weight between the two groups (IUGR: 1810 g and AGA: 2695 g, P = .0001). The frequency of cesarean sections was 92% in the IUGR group and 25% in the AGA group (P < .0001). No difference in mean umbilical cord pH or the presence of neonatal acidosis was observed between groups. Only one newborn of the IUGR group and none of the AGA infants presented Apgar scores <7 at 5 minutes. The length of stay of the newborn in the nursery, as well as the need for and duration of hospitalization in the NICU, differed significantly between the two groups (Table 2).

There was a significant difference in mean birth weight between the two groups (IUGR: 1810 g and AGA: 2695 g, P = .0001). The frequency of cesarean sections was 92% in the IUGR group and 25% in the AGA group (P < .0001). No difference in mean umbilical cord pH or the presence of neonatal acidosis was observed between groups. Only one newborn of the IUGR group and none of the AGA infants presented Apgar scores <7 at 5 minutes. The length of stay of the newborn in the nursery, as well as the need for and duration of hospitalization in the NICU, differed significantly between the two groups (Table 2). TTN or apnea rates did not differ significantly between IUGR and AGA infants. Late-preterm IUGR infants were found to be at a higher risk for IVH. There were only Grade 1 IVH in this sample. No respiratory distress syndrome, pulmonary hemorrhage or bronchopulmonary dysplasia was observed in either group. The frequency of sepsis or thrombocytopenia did not differ between groups. Hypoglycemia was more frequent in the IUGR group. The presence of hyperbilirubinemia was similar in the two groups (98% in the IUGR group versus 100% in the AGA group) (P = .52, Fisher's exact test) (Table 3). However, there was a difference in the number of days the newborn required phototherapy between the IUGR and AGA groups, which was higher in the former (Table 2). None of the newborns died or developed retinopathy of prematurity or necrotizing enterocolitis during their stay in the nursery.

TTN or apnea rates did not differ significantly between IUGR and AGA infants. Late-preterm IUGR infants were found to be at a higher risk for IVH. There were only Grade 1 IVH in this sample. No respiratory distress syndrome, pulmonary hemorrhage or bronchopulmonary dysplasia was observed in either group. The frequency of sepsis or thrombocytopenia did not differ between groups. Hypoglycemia was more frequent in the IUGR group. The presence of hyperbilirubinemia was similar in the two groups (98% in the IUGR group versus 100% in the AGA group) (P = .52, Fisher's exact test) (Table 3). However, there was a difference in the number of days the newborn required phototherapy between the IUGR and AGA groups, which was higher in the former (Table 2). None of the newborns died or developed retinopathy of prematurity or necrotizing enterocolitis during their stay in the nursery. 4. Discussion Late-preterm births account for 70% of all preterm births [21]. Since the number of late-preterm births is increasing, it is important to understand the unique problems that this growing population of infants may experience. In this respect, a higher incidence of neonatal mortality has been reported for births that occur between 34 weeks and 36 weeks and 6 days [22] when compared to term newborns (>37 weeks). In addition, an increase in various neonatal complications such as hypoglycemia, respiratory distress syndrome (RDS), thermal instability, apnea of prematurity, hyperbilirubinemia, neonatal sepsis, prolonged stay in the nursery, and need for NICU treatment has been observed [22–28].

mpared to term newborns (>37 weeks). In addition, an increase in various neonatal complications such as hypoglycemia, respiratory distress syndrome (RDS), thermal instability, apnea of prematurity, hyperbilirubinemia, neonatal sepsis, prolonged stay in the nursery, and need for NICU treatment has been observed [22–28]. IUGR is a frequent complication in preterm infants and is the cause of most elective late-preterm deliveries. IUGR may also contribute to the increased morbidity and mortality observed among late-preterm infants. These infants are at risk for hypoglycemia, IVH, prolonged hospital stay and increased need for NICU treatment when compared to AGA infants, thus demonstrating the greater severity of these cases. There is a lack of studies regarding the evolution of late-preterm infants with and without IUGR. Gilbert and Danielsen [29] reported a higher incidence of IVH and higher hospital costs for late-preterm IUGR infants when compared to AGA infants. In contrast to the present results, Sharma et al. [8] observed a lower incidence of RDS in late-preterm infants with IUGR; however, as observed in the present study, these infants required prolonged hospitalization in the nursery. Tyson et al. [15] and Piper et al. [30] compared the incidence of RDS between IUGR and AGA infants but we could not compare our results to theirs because these authors used a different definition of RDS.

th IUGR; however, as observed in the present study, these infants required prolonged hospitalization in the nursery. Tyson et al. [15] and Piper et al. [30] compared the incidence of RDS between IUGR and AGA infants but we could not compare our results to theirs because these authors used a different definition of RDS. Although we did not observe more severe neonatal complications in the present cases (e.g., respiratory distress syndrome, bronchopulmonary dysplasia, pulmonary hemorrhage, necrotizing enterocolitis and neonatal death), hypoglycemia, neonatal sepsis, IVH, thrombocytopenia and hyperbilirubinemia were present in the late-preterm infants studied and were the cause of NICU treatment and prolonged stay in the nursery. The lack of observation of these more severe complications might be explained by the small sample studied since the frequency of these complications is rare in this gestational age group. Mean gestational age was 35.5 weeks in the two groups, a gestational age at which the incidence of respiratory distress syndrome is very low [31]. Thus, the number of newborns necessary to detect cases of this disease would have to be high. In the present series no cases of severe pulmonary complications such as bronchopulmonary dysplasia or pulmonary hemorrhage were observed, a finding that might be explained by the small sample size since these complications are also rare in this gestational age group [25]. The absence of neonatal death in the present sample is probably due to the low mortality of these newborns, which is approximately 7.7 per 1000 live births [21].

morrhage were observed, a finding that might be explained by the small sample size since these complications are also rare in this gestational age group [25]. The absence of neonatal death in the present sample is probably due to the low mortality of these newborns, which is approximately 7.7 per 1000 live births [21]. Despite the low rates of severe neonatal complications, these newborns are of marked importance for public health since they account for a large percentage of newborns hospitalized in NICUs and require large amounts of public resources during and after their stay in the nursery [32–34]. In addition, our study only analyzed neonatal morbidity immediately after delivery during the stay of the newborn in the nursery, but not the long-term consequences of late-preterm birth. The Institute of Medicine analyzed the late consequences of preterm birth in the United States and demonstrated marked human and economic impacts during childhood resulting from preterm birth [35].

ly after delivery during the stay of the newborn in the nursery, but not the long-term consequences of late-preterm birth. The Institute of Medicine analyzed the late consequences of preterm birth in the United States and demonstrated marked human and economic impacts during childhood resulting from preterm birth [35]. As also reported in the study of Gilbert and Danielsen [29], the frequency of IVH differed significantly between the two groups and was more common in IUGR infants, a finding suggesting that IUGR is indeed a risk factor for IVH in late-preterm infants. Laptook and Jackson [36] have demonstrated an elevated incidence of hypoglycemia in late-preterm infants as a result of deficient neoglycogenesis, hepatic glycogenolysis and lipolysis and of hormonal irregularities. Neonatal sepsis was rare in the present sample and was similar in the two groups (4% versus 0). These findings agree with Arnon et al. [24] who observed an incidence of 5% of neonatal sepsis in neonates born at 34 weeks of gestation and no case among those born at 36 weeks. The exclusion of cases with premature membrane rupture may have contributed to this result [24]. Neonatal jaundice was very common in both groups (98% versus 100%). Furthermore, IUGR infants required phototherapy for a longer period of time than AGA infants, that is, jaundice was more severe in this group.

at 36 weeks. The exclusion of cases with premature membrane rupture may have contributed to this result [24]. Neonatal jaundice was very common in both groups (98% versus 100%). Furthermore, IUGR infants required phototherapy for a longer period of time than AGA infants, that is, jaundice was more severe in this group. Late-preterm birth poses various risks to the newborn and the obstetrician should always weigh the risks and benefits in each case to decide whether to interrupt pregnancy between 34 weeks and 36 weeks and 6 days of gestation. We believe that the technological advances in obstetrics that have occurred over the last few years permit a better control of high-risk pregnancies. Thus, the priority of the obstetrician is to strive for delivery as close to term as possible. 5. Conclusion In conclusion, our study showed that late-preterm IUGR infants present a significantly higher risk of neonatal complications and a significantly longer NCIU and hospital stay when compared to late-preterm AGA infants. Thus, evaluation of the birth weight percentile for gestational age may provide a more realistic expectation of outcome among late-preterm infants. Table 1 Maternal characteristics and indications for elective preterm delivery in the IUGR group (n = 50). Underlying disease/   Indications for obstetric complications   elective resolution Systemic 4 (8%) Oligohydramnios 20 (51.3%) Lupus Erythematosus Heart diseases 5 (10%) Severe maternal 8 (20.5%) Disease Hypertensive 24 (48%) Cardiotocographic 6 (15.4%) Disorders Abnormalities

Table 1 Maternal characteristics and indications for elective preterm delivery in the IUGR group (n = 50). Underlying disease/   Indications for obstetric complications   elective resolution Systemic 4 (8%) Oligohydramnios 20 (51.3%) Lupus Erythematosus Heart diseases 5 (10%) Severe maternal 8 (20.5%) Disease Hypertensive 24 (48%) Cardiotocographic 6 (15.4%) Disorders Abnormalities Others 6 (12%) FBP or 3 (7.7%) Doppler alterations Fetal maturity 2 (5.1) IUGR: intrauterine growth restriction; FBP: fetal biophysical profile. Data are reported as number of cases and percentage. Table 2 Comparison of neonatal outcomes between the IUGR and AGA groups. Outcome IUGR group AGA group P Length of stay Mean 16.36 4.58 (days) Median 16.5 3 .0001 SD 10.77 2.18 Length of Mean 5.92 1.28 NICU stay Median 2.5 0 <.0001 (days) SD 7.71 2.28 Phototherapy Mean 5.78 3.19 (days) Median 5 3 .005 SD 3.71 2.11 IUGR: intrauterine growth restriction; AGA: appropriate-for-gestational age; NICU: neonatal intensive care unit; SD: standard deviation. Table 3 Comparison of neonatal complications between the IUGR and AGA groups. Neonatal IUGR group AGA group P complications TTN 27 (54%) 16 (44.4%) >.05 Apnea of 3 (6%) 0 >.05 prematurity Intraventricular 6 (12%) 0 .037 hemorrhage Neonatal 5 (10%) 0 >.05 thrombocytopenia Neonatal sepsis 2 (4%) 0 >.05 Hypoglycemia 12 (24%) 2 (6%) .047 Hyperbilirubinemia 49 (98%) 36 (100%) .52 Data are reported as number of cases and percentage. IUGR: intrauterine growth restriction; AGA: appropriate-for-gestational age; TTN: transient tachypnea of the newborn.

1. Introduction Endometrial malignancies can be categorized into two main groups based on the cell of origin: (i) endometrial carcinoma including carcinosarcoma and (ii) endometrial stromal sarcoma. Endometrial carcinomas show a broad spectrum of phenotypes which show various histologic appearances for example, endometrioid, serous, mucinous, squamous, urothelial, or clear cell, reflecting the differentiation potential of the müllerian epithelium and the difference in the tumorigenetic pathways of each tumor type. Women with an inherited predisposition for endometrial neoplasm have been reported, associated with autosomal dominant disorders such as hereditary nonpolyposis colorectal carcinoma (HNPCC) and Cowden syndrome. Some endometrial carcinomas undergo mesenchymal differentiation and are termed carcinosarcomas (formerly termed malignant mixed müllerian tumors). Pathogenetically and clinically, two distinct forms of endometrial adenocarcinoma, type I and type II, have been described. The molecular alterations driving endometrial carcinogenesis may follow a sequence similar to Vogelstein's model for the progression of colorectal adenoma to carcinoma. This process is accompanied by stepwise genetic changes of oncogenes and tumor suppressor genes. Endometrial stroma may give rise to neoplasms that resemble normal endometrial stromal cells. The spectrum of endometrial stromal tumors ranges from the benign stromal nodule to the malignant endometrial stromal sarcoma. An oncogenic fusion gene, JAZF1/JJAZ1 plays a significant role in tumor development of endometrial stromal sarcomas [1].

y give rise to neoplasms that resemble normal endometrial stromal cells. The spectrum of endometrial stromal tumors ranges from the benign stromal nodule to the malignant endometrial stromal sarcoma. An oncogenic fusion gene, JAZF1/JJAZ1 plays a significant role in tumor development of endometrial stromal sarcomas [1]. 2. Molecular Profiling of Endometrial Carcinoma 2.1. Dualistic Model of Endometrial Tumorigenesis Endometrial carcinoma is the most common malignant neoplasm of female genital tract in developed countries [2] with an estimated 42,160 new cases diagnosed in the United States for 2009 [3]. Approximately 90% of cases of endometrial carcinoma are sporadic, whereas the remaining 10% of cases are hereditary [4]. Clinically, the patients with endometrial carcinomas most often present with abnormal uterine bleeding. In advanced stages, patients may complain of pelvic pain, reflecting spread of the carcinoma. Bokhman [5] first described the pathogenetic classification of 2 different types of endometrial carcinoma, designated as type I and type II carcinomas, according to the determination of biological properties of the tumor, its clinical course, and the prognosis of the disease.

c pain, reflecting spread of the carcinoma. Bokhman [5] first described the pathogenetic classification of 2 different types of endometrial carcinoma, designated as type I and type II carcinomas, according to the determination of biological properties of the tumor, its clinical course, and the prognosis of the disease. 2.1.1. Type I (Endometrioid Endometrial Carcinoma) [1, 2, 4–13] Type I carcinomas represent the majority of sporadic cases of endometrial carcinoma, accounting for 70–80% of new cases [4, 9–12] which occur predominantly in pre- and perimenopausal women. These cancers are typically of endometrioid type (Figure 1(a)). Risk factors include obesity, hyperlipidemia, and hyperestrogenism for example, anovulation, nulliparity/infertility, late onset of menopause, and endometrial hyperplasia. The tumors in this category are generally low-grade, low-stage, and indolent. They commonly express estrogen and progesterone receptors [2, 4–11]. The rare mucinous carcinomas are also considered type I carcinomas because they usually express estrogen and/or progesterone receptors and are of low histologic grade [10, 11].

tumors in this category are generally low-grade, low-stage, and indolent. They commonly express estrogen and progesterone receptors [2, 4–11]. The rare mucinous carcinomas are also considered type I carcinomas because they usually express estrogen and/or progesterone receptors and are of low histologic grade [10, 11]. 2.1.2. Type II (Nonendometrioid Endometrial Carcinoma) [1, 2, 4–13] Type II carcinomas are less common, accounting for 10–20% of endometrial carcinoma [4, 6]. They are nonendometrioid in differentiation, most frequently papillary serous (Figure 1(b)) and less frequently clear cell, have high-grade histology, typically arise in an atrophic endometrial background, and often have deep myometrial penetration. They usually occur at an older age, approximately 5–10 years later than type I tumors. There is no relationship to estrogen stimulation. Clinically, type II cancers have an aggressive behavior, with a high frequency of distant spread to pelvic lymph nodes. Small cell, undifferentiated and squamous cell carcinomas may also be encountered among type II carcinomas, but little is known about their tumorigenesis [11]. The clinical and pathological features of the two types of endometrial carcinomas are summarized in Table 1.

frequency of distant spread to pelvic lymph nodes. Small cell, undifferentiated and squamous cell carcinomas may also be encountered among type II carcinomas, but little is known about their tumorigenesis [11]. The clinical and pathological features of the two types of endometrial carcinomas are summarized in Table 1. 2.2. Common Molecular Genetic Alterations in Dualistic Model Evidence for divergent molecular alterations supporting the dualistic model of endometrial tumorigenesis became available approximately 10 years after Bokhman's description of the clinical and pathologic features. The two distinct histological types of carcinomas are associated with genetic alterations of independent sets of genes. These genetic changes may occur singly or in various combinations which differ between individual cases [9].

approximately 10 years after Bokhman's description of the clinical and pathologic features. The two distinct histological types of carcinomas are associated with genetic alterations of independent sets of genes. These genetic changes may occur singly or in various combinations which differ between individual cases [9]. Westin and colleague [14] described that expression of estrogen-induced genes, RALDH2, EIG121, SFRP1, SFRP4, IGF-1, and IGF-IR, tend to be highest in the well-to-moderately differentiated endometrioid carcinoma. This finding supports the partitioning of endometrial carcinoma into two distinct groups by traditional estrogen-related classification. According to this model, normal endometrial cells would transform into endometrioid endometrial carcinoma through 5 different molecular changes, including, mutations of PTEN, PIK3CA, KRAS, and CTNNB1 (β-catenin) genes and microsatellite instability (MSI) while non-endometrioid endometrial carcinoma is frequently related to alterations of p53 and chromosomal instability [7, 8, 15, 16]. Non-endometrioid endometrial carcinoma frequently demonstrates high-ordered aneuploidy and has an intact mismatch repair (MMR) mechanism [12]. Furthermore, none of the five main alterations of endometrioid endometrial carcinoma (mutations of PTEN, PIK3CA, KRAS, and CTNNB1 genes and MSI) plays a major role in non-endometrioid endometrial carcinoma. However, in many endometrial carcinomas exhibit overlapping clinical, morphologic, immunohistochemical, and molecular features of the both types of carcinoma for example, a subset of endometrioid endometrial carcinoma is found with a background of atrophic endometrium or papillary serous carcinoma may occasionally develop from a pre-existing endometrioid endometrial carcinoma and may share histological and genetic features [8–10]. Matias-guiu et al. [8] described the development of non-endometrioid endometrial carcinoma through these possible pathways: (i) de novo, through p53 mutations, loss of heterozygosity (LOH) at several loci, and some other still unknown gene alterations; or (ii) through dedifferentiation of a pre-existing endometrioid carcinoma. These dedifferentiated non-endometrioid endometrial carcinomas exhibit overlapping features with type I endometrioid endometrial carcinoma [8].

ations, loss of heterozygosity (LOH) at several loci, and some other still unknown gene alterations; or (ii) through dedifferentiation of a pre-existing endometrioid carcinoma. These dedifferentiated non-endometrioid endometrial carcinomas exhibit overlapping features with type I endometrioid endometrial carcinoma [8]. Comparison of the major genetic alterations between type I and type II endometrial carcinomas is shown in Table 2. Molecular genetic alterations have been extensively investigated in endometrioid and papillary serous adenocarcinomas of the endometrium. These two tumor types are characterized by distinctive molecular alterations, and their tumorigenesis follow separate pathways. 2.3. Molecular Pathology of Endometrioid Carcinomas 2.3.1. PTEN The most frequently altered gene in endometrioid endometrial carcinoma is PTEN (phosphatase and tensin homologue deleted from chromosome 10), also called MMAC1 (mutated in multiple advanced cancers 1). PTEN behaves as a tumor suppressor gene, is located on chromosome 10q23.3 and encodes a lipid phosphatase that antagonizes the PI3K/AKT pathway by dephosphorylating PIP3, the product of PI3K. This lipid molecule is an important second messenger that regulates the phosphorylation of a protein termed AKT, also known as protein kinase B. Decreased PTEN activity causes increased cell proliferation and survival through modulation of signal transduction pathways.

ay by dephosphorylating PIP3, the product of PI3K. This lipid molecule is an important second messenger that regulates the phosphorylation of a protein termed AKT, also known as protein kinase B. Decreased PTEN activity causes increased cell proliferation and survival through modulation of signal transduction pathways. PTEN may be inactivated by several mechanisms such as mutation, LOH, and promoter hypermethylation. Somatic PTEN mutations are common in endometrial carcinoma, and they are almost exclusively restricted to endometrioid endometrial carcinomas, occurring up to 83% of them [1, 4, 11, 12]. Germline mutations of PTEN are responsible for Cowden syndrome [9, 12]. PTEN may be also inactivated by deletion, as shown by LOH in 40% of endometrial carcinomas [7–9, 17]. Promoter hypermethylation leading to PTEN inactivation, is found in about 20% of tumors, most of which are high-stage [10].

f them [1, 4, 11, 12]. Germline mutations of PTEN are responsible for Cowden syndrome [9, 12]. PTEN may be also inactivated by deletion, as shown by LOH in 40% of endometrial carcinomas [7–9, 17]. Promoter hypermethylation leading to PTEN inactivation, is found in about 20% of tumors, most of which are high-stage [10]. PTEN mutations have been detected in 15–55% of endometrial hyperplasias with and without atypia [9, 13]. Interestingly, concordance between MSI status and PTEN mutations has been found; the mutations occur in 60–86% of MSI-positive endometrioid endometrial carcinoma but in only 24–35% of the MSI-negative cases [7–9, 13, 17]. This suggests that PTEN could be a target for mutations in the context of DNA repair deficiency [13]. In addition, identical PTEN mutations have been also identified in hyperplasias coexisting with MSI-positive endometrioid endometrial carcinoma, which suggests that PTEN mutations are early events in their development [8]. On the other hand, identical PTEN mutations have been detected in MSI-negative endometrial hyperplasia with coexisting MSI-positive endometrioid endometrial carcinomas. Thus, some PTEN mutations may precede MSI, and coexistence of both alterations does not necessarily mean a cause-effect relationship [9]. Evaluation of PTEN inactivation in endometrial carcinoma precursor lesions by PTEN immunostaining has been proposed. However, commercially available antibodies (e.g., clone 10P03, 28H6, polyclonal, 6H2.1) do not have statistically significant associations with the molecular genetic alterations [7, 9, 19]. Some data suggest that PTEN is associated with younger age, low stage, endometrioid histology, low histologic grade, and favorable prognosis (78% 5-year survival for patients without mutations, compared with 95% and 93% for patients with one or more mutations, resp.) [7, 9]. In addition, recent data suggest that only PTEN mutations outside exons 5–7 may predict favorable survival, independent of the clinical and pathological features of the tumors [9].

osis (78% 5-year survival for patients without mutations, compared with 95% and 93% for patients with one or more mutations, resp.) [7, 9]. In addition, recent data suggest that only PTEN mutations outside exons 5–7 may predict favorable survival, independent of the clinical and pathological features of the tumors [9]. 2.3.2. PIK3CA The PIK3CA (p110α catalytic subunit of PI3K) gene locates on chromosome 3q26.32. Phosphatidylinositol-3-kinase (PI3K) is heterodimeric lipid kinase consisting of a catalytic subunit (p110) and a regulatory subunit (p85) in PI3K/AKT signaling pathway. This pathway is frequently activated in endometrial carcinoma through various genetic alterations and their combinations. Activation of PI3K produces the second messenger PIP3 which subsequently activates various down-stream pathways such as AKT. This regulation involves suppression of apoptosis and enhancement of cell proliferation [9]. PIK3CA activation is reported in 26–36% of endometrial carcinoma and may coexist with PTEN (15–27%) [7, 9, 12, 15, 20] and KRAS mutations [9, 15, 20] suggesting that the PIK3CA mutations cooperate with these alterations in malignant transformation [16]. Mutations in AKT family members and their correlation with other gene alterations are found in endometrial carcinoma, including AKT2 (D399N), AKT2 (D32H) and AKT3 (E438D) mutations. Mutations of AKT3 (E438D) also have amplification of and a mutation in PIK3CA [21]. AKT1 E17K mutation is not associated with either PTEN or PIK3CA genomic alteration [21]. In vitro studies showed that activating mutations of PIK3CA in combination with PTEN mutations led to an additional increase in phosphorylated AKT when compared with cells with only inactivated PTEN [6]. Some investigators have claimed that PIK3CA mutations are mutually exclusive of PTEN mutations, suggesting that tumorigenic signaling through this pathway can occur either through activation of PIK3CA or inactivation of PTEN [9]. Recently, interactions between the PI3K/AKT and p53 signaling pathways have been described in which activation of the PI3K/AKT pathway through PTEN or PIK3CA mutations, together with p53 inactivation, results in malignant transformation [15]. Moreover, patients with dysregulation of PI3K/AKT signaling pathway and p53 alterations had shorter survival than patients with only p53 alterations [15].

een described in which activation of the PI3K/AKT pathway through PTEN or PIK3CA mutations, together with p53 inactivation, results in malignant transformation [15]. Moreover, patients with dysregulation of PI3K/AKT signaling pathway and p53 alterations had shorter survival than patients with only p53 alterations [15]. Mutations were more common in mixed endometrioid-nonendometrioid adenocarcinomas (44%) than in pure endometrioid adenocarcinomas (28%) or pure nonendometrioid adenocarcinomas (21%) [15]. In fact, PIK3CA mutations are usually missense and cluster in exons 9 (helical domain) and 20 (kinase domain). The tumors carrying exon 9 PIK3CA mutations are more likely to be low-grade carcinomas; in contrast, carcinomas with exon 20 mutations or PIK3CA mRNA overexpression are often high-grade carcinomas associated with myometrial invasion and tended to have lymphovascular invasion [15]. Furthermore, in high-grade endometrioid adenocarcinomas and mixed carcinomas, PIK3CA mutations in exon 20 coexist with p53 alterations more frequently than in nonendometrioid adenocarcinomas. However, PIK3CA mRNA overexpression occurs in concert with p53 alterations only in nonendometrioid endometrial carcinomas [15]. PIK3CA mutations did not correlate with MSI or β-catenin/CTNNB1 mutations [9, 18]. PIK3CA mutations, particularly exon 20 mutations or PIK3CA mRNA overexpression, are frequent in endometrioid endometrial carcinoma in association with invasion and adverse prognostic factors such as blood vessel invasion [7, 15]. 2.3.3. KRAS

Mutations were more common in mixed endometrioid-nonendometrioid adenocarcinomas (44%) than in pure endometrioid adenocarcinomas (28%) or pure nonendometrioid adenocarcinomas (21%) [15]. In fact, PIK3CA mutations are usually missense and cluster in exons 9 (helical domain) and 20 (kinase domain). The tumors carrying exon 9 PIK3CA mutations are more likely to be low-grade carcinomas; in contrast, carcinomas with exon 20 mutations or PIK3CA mRNA overexpression are often high-grade carcinomas associated with myometrial invasion and tended to have lymphovascular invasion [15]. Furthermore, in high-grade endometrioid adenocarcinomas and mixed carcinomas, PIK3CA mutations in exon 20 coexist with p53 alterations more frequently than in nonendometrioid adenocarcinomas. However, PIK3CA mRNA overexpression occurs in concert with p53 alterations only in nonendometrioid endometrial carcinomas [15]. PIK3CA mutations did not correlate with MSI or β-catenin/CTNNB1 mutations [9, 18]. PIK3CA mutations, particularly exon 20 mutations or PIK3CA mRNA overexpression, are frequent in endometrioid endometrial carcinoma in association with invasion and adverse prognostic factors such as blood vessel invasion [7, 15]. 2.3.3. KRAS KRAS encodes a member protein of the small GTPase superfamily and is involved in signal transduction pathways between cell surface receptors and the nucleus. KRAS mutations have been identified in 10–30% of endometrioid endometrial carcinomas [1, 2, 4, 7–12, 17] while some investigators have reported an almost complete absence of KRAS mutations in serous and clear cell carcinomas of endometrium [8]. Some studies found a higher frequency of KRAS mutations in MSI-positive carcinomas than in MSI-negative tumors [8–10, 16] suggesting that both events may occur simultaneously before clonal expansion [10, 13]. KRAS mutations were detected in endometrial hyperplasias at a similar rate to that observed in endometrioid endometrial carcinomas, suggesting that KRAS mutations are early events in endometrial carcinogensis [9, 13]. No relationship has been found between KRAS mutations and tumor stage, histologic grade, depth of myometrial invasion, age, or clinical outcome in endometrioid endometrial carcinomas [9].

served in endometrioid endometrial carcinomas, suggesting that KRAS mutations are early events in endometrial carcinogensis [9, 13]. No relationship has been found between KRAS mutations and tumor stage, histologic grade, depth of myometrial invasion, age, or clinical outcome in endometrioid endometrial carcinomas [9]. 2.3.4. β-Catenin (CTNNB1) The β-catenin gene (CTNNB1) maps to 3p21. It appears to be important in the functional activities of both APC (adenomatous polyposis coli) and E-cadherin. It is a component of the E-cadherin-catenin unit, essential for cell differentiation and maintenance of normal tissue architecture and also plays an important role in Wnt signal transduction pathway. Mutations in exon 3 of CTNNB1 result in stabilization of a protein that resists degradation, leading to nuclear accumulation of β-catenin, have been described in endometrioid endometrial carcinoma. The accumulation of β-catenin can be demonstrated by immunostaining. Several studies have analyzed endometrial carcinomas showing that nuclear accumulation of β-catenin is significantly more common in endometrioid lesions (31–47%) compared with nonendometrioid histology (0–3%) [4]. By comparison in colonic adenocarcinomas, elevated β-catenin levels caused by mutations in CTNNB1 or APC result in activation of the Wnt/β-catenin/LEF1 pathway through a LEF1 binding site in the cyclin D1 promotor, triggering cyclin D1 gene expression, and subsequently, uncontrolled progression of tumor cells into the cell cycle [8, 12]. Furthermore, β-catenin might regulate the expression of the matrix metalloproteinase-7 that would have a role in the establishment of the microenvironment necessary for the initiation and maintenance of growth of the primary tumor and metastasis [8, 12]. The reported frequency of CTNNB1 mutations in endometrioid endometrial carcinoma ranges from 14–44% [7, 8]. They seem to be independent from the presence of MSI and the mutations of PTEN and KRAS, suggesting that the Wnt pathway may play an independent role in endometrial cancer [10, 13]. In all cases, the mutations were homogeneously distributed in different areas of the tumors suggesting that they play a role in early steps of endometrial tumorigenesis. Alterations in β-catenin have been reported in endometrial hyperplasias with squamous metaplasia [7, 9].

ependent role in endometrial cancer [10, 13]. In all cases, the mutations were homogeneously distributed in different areas of the tumors suggesting that they play a role in early steps of endometrial tumorigenesis. Alterations in β-catenin have been reported in endometrial hyperplasias with squamous metaplasia [7, 9]. Although there was a good correlation between CTNNB1 mutations and β-catenin nuclear immunostaining, the presence of cytoplasmic and nuclear β-catenin immunoreactivity in some endometrial carcinomas without CTNNB mutation suggests that the changes of other genes in the Wnt/β-catenin/LEF-1 pathway may be responsible for the stabilization and putative transcription activator role of β-catenin [7, 8]. Endometrioid endometrial carcinomas with CTNNB1 mutations are characteristically early stage tumors associated with favorable prognosis [7, 9]. Two members of the secreted frizzled-related protein (SFRP) family, SFRP1 and SFRP4, were more frequently down-regulated in MSI-positive carcinomas compared with MSI-negative carcinomas. This down-regulation was associated with frequent promoter methylation of SFRP1 and led to an activation of the β-catenin pathway. In addition, the Wnt-target fibroblast growth factor 18 was up-regulated in endometrioid carcinomas with MSI compared with normal endometrium [1].

mas compared with MSI-negative carcinomas. This down-regulation was associated with frequent promoter methylation of SFRP1 and led to an activation of the β-catenin pathway. In addition, the Wnt-target fibroblast growth factor 18 was up-regulated in endometrioid carcinomas with MSI compared with normal endometrium [1]. 2.3.5. Microsatellite Instability Microsatellite DNA sequences are polymorphic, short-tandem repeats distributed throughout the genome. The most common microsatellite in human is a dinucleotide repeat of CA, (CA)n, and there are 50,000 to 100,000 (CA)n repeats scattered in the human genome [8, 9]. Microsatellite instability (MSI) is a condition manifested by damaged DNA because of defects in normal DNA repair process. Mammalian mismatch repair (MMR) genes encode for nine proteins (MLH1, MLH3, PMS1, PMS2, MSH2, MSH3, MSH4, MSH5, and MSH6) that interact with each other to form complexes and heterodimers that mediate distinct functions in MMR-related system. This repair process plays a central role in promoting genetic stability by repairing DNA replication errors, inhibiting recombination between non-identical DNA sequences and participating in responses to DNA damage. MSI is a common genetic abnormality that has been detected in 20–45% of sporadic endometrioid endometrial carcinoma [7–10]. In addition, MSI in nonendometrioid endometrial carcinomas has been reported (0–11%) [8, 9], particularly in mixed endometrioid and serous carcinomas, but not in pure serous carcinomas [10]. In sporadic endometrial carcinoma, epigenetic cause of MSI is more common involving MLH1 promotor hypermethylation which is the main cause of MMR deficiency [7–9, 13, 15]. This epigenetic inactivation usually occurs in atypical hyperplasia, most of which coexists with carcinomas. Thus, MLH1 hypermethylation is an early event in the pathogenesis of endometrioid endometrial carcinoma, which precedes the development of MSI [7–9, 15]. The remaining unmethylated MLH1 cases reveal MSH2 mutations (15%) and MSH6 mutations (60%), of which almost half are germline mutations. Thus, MSH6 mutations seem to be a frequent cause of MSI [11, 12]. Tumors with MSI of CpG island methylation in the promoter region have been identified in some other genes, for example, p16, PTEN, and E-cadherin (CDH1), suggesting altered methylation may be a coexisting independent early change [9].

germline mutations. Thus, MSH6 mutations seem to be a frequent cause of MSI [11, 12]. Tumors with MSI of CpG island methylation in the promoter region have been identified in some other genes, for example, p16, PTEN, and E-cadherin (CDH1), suggesting altered methylation may be a coexisting independent early change [9]. The presentation of some small short-tandem repeats such as mononucleotide repeats located within the coding sequence of important genes for example, transforming growth factor β receptor type II (TGF-βRII), BAX, insulin-like growth factor II receptor (IGFIIR), MSH3, MSH6, caspase-5, and PTEN may promote MSI-positive endometrial carcinoma [8, 9]. Secondary mutations at one or more mononucleotide tracts found in 72.7% of tumors with MSI, are responsible for tumor progression [7–9]. International Federation of Gynecology and Obstetrics (FIGO) grade has been found to be higher in endometrioid endometrial carcinomas with MSI in some, but not all studies, similar to the well-established association between MSI and high-grade colorectal carcinomas [16]. By multivariate analysis, a significant correlation between MSI-positive tumors and tumor-infiltrating lymphocytes in endometrioid endometrial carcinoma was found: 40 tumor-infiltrating lymphocytes/10 high power fields has a sensitivity of 85% and a specificity of 46% in predicting MSI [16].

It is postulated that mutation in one allele occurs early during the development of serous carcinoma, and loss of the second normal allele occurs late in the progression to carcinoma [4]. p53 mutations are almost always associated with aneuploidy and do not seem to occur with PTEN mutations in the same tumor [10, 11]. 2.4.2. HER2/neu Epidermal growth factor receptor II or HER2/neu is an oncogene that codes for a transmembrane receptor tyrosine kinase involved in cell signaling and located at the long arm of human chromosome 17q12. HER2/neu overexpression or amplification is more frequently found in non-endometrioid endometrial carcinoma (18–80%) [13] than in grade 2 and 3 endometrioid carcinoma (10–30%) [7, 9, 10] and has been associated with adverse prognostic parameters including advanced stage, high histologic grade, and low overall survival [9, 13]. 2.4.3. p16 p16 plays an important role in regulating the cell cycle. It is a tumor suppressor gene located on chromosome 9p21 [10]. p16 inactivation can lead to uncontrolled cell growth. Inactivation of p16 is more frequent in non-endometrioid endometrial carcinoma (40–45%) than in endometrioid endometrial carcinoma (10%) [4, 7, 10]. The underlying mechanism is unclear [7, 11], because neither promoter hypermethylation nor deletion or mutation is frequently found [11]. Loss of p16 expression is correlated with KRAS and p53 mutations and is associated with high stage, high grade, and poor survival [10].

trioid endometrial carcinoma (10%) [4, 7, 10]. The underlying mechanism is unclear [7, 11], because neither promoter hypermethylation nor deletion or mutation is frequently found [11]. Loss of p16 expression is correlated with KRAS and p53 mutations and is associated with high stage, high grade, and poor survival [10]. 2.4.4. E-Cadherin Cadherins are a family of adhesion molecules essential for tight connection between cells. E-cadherin is encoded by CDH1 gene and locates on chromosome 16q22.1. It is thought to be a tumor suppressor gene, the loss of which has been demonstrated to promote tumor invasion and metastasis. Decreased expression of E-cadherin is frequent in endometrial carcinoma and may be caused by LOH or promotor hypermethylation. LOH at 16q22.1 is seen in almost 60% of non-endometrioid endometrial carcinoma, but in only 22% of endometrioid endometrial carcinoma [7]. In endometrial carcinoma, partial or complete loss of E-cadherin expression correlates with aggressive behavior [9].

carcinoma and may be caused by LOH or promotor hypermethylation. LOH at 16q22.1 is seen in almost 60% of non-endometrioid endometrial carcinoma, but in only 22% of endometrioid endometrial carcinoma [7]. In endometrial carcinoma, partial or complete loss of E-cadherin expression correlates with aggressive behavior [9]. Among type II carcinomas, clear cell carcinomas seem to follow a separate pathway that shows some overlap with serous and endometrioid carcinomas. p53 mutations are only present in about 30–40% of clear cell carcinomas compared to 90% of serous carcinomas. However, the frequency of MSI and PTEN alterations in clear cell carcinoma is higher than in serous carcinoma (15% versus <5 for MSI and 30% versus 10% for PTEN) but lower compared with endometrioid carcinoma (20–40% and 35–50%, resp.) [24]. A recent molecular study demonstrated that the majority of pure clear cell carcinomas do not show mutations in either PTEN or p53, the most commonly altered genes in type I and type II tumors, respectively. These findings suggest that clear cell carcinoma may arise through a distinct pathologic pathway [6].

esp.) [24]. A recent molecular study demonstrated that the majority of pure clear cell carcinomas do not show mutations in either PTEN or p53, the most commonly altered genes in type I and type II tumors, respectively. These findings suggest that clear cell carcinoma may arise through a distinct pathologic pathway [6]. 2.4.5. Apoptosis Resistance in Endometrial Carcinoma Several of the molecular abnormalities that have been detected in EC may be associated with apoptosis deregulation. Apoptosis can be initiated by two main mechanisms: (i) the “intrinsic pathway” activated by released mitochondrial proteins, such as cytochrome-c; and (ii) the “extrinsic pathway” activated by ligand-bound death receptors such as tumor necrosis factor (TNF), Fas or TNF-related apoptosis including ligand (TRAIL) receptors. Some studies have shown that cellular apoptosis susceptibility (CAS) gene, BCL2, BAX, and caspase-3 are apparently involved in the progressive deregulation of proliferation and apoptosis, leading from simple and complex endometrial hyperplasia to adenocarcinoma. As described above, PTEN antagonizes the PI3K/AKT pathway by dephosphorylating PIP3, resulting in decreased translocation of AKT activation. Thus, loss of PTEN function leads to increased levels of phospho-AKT, activation of anti-apoptotic protein, and cell cycle progression [9]. NF-κB, frequently activated in endometrioid endometrial carcinomas, may inhibit apoptosis by activation of target genes such as FLIP and Bcl-XL. Furthermore, there are reports that apoptosis-related protein survivin is frequently overexpressed in endometrial carcinomas [7, 9] and correlates inversely with PTEN expression [9]. Where widespread genetic abnormalities exist that cannot be corrected, MMR proteins initiate apoptosis as a more energy efficient option of universal genomic preservation [16]. MMR deficiency lowers the apoptotic rate, leading a survival advantage to the mutated cells [16].

orrelates inversely with PTEN expression [9]. Where widespread genetic abnormalities exist that cannot be corrected, MMR proteins initiate apoptosis as a more energy efficient option of universal genomic preservation [16]. MMR deficiency lowers the apoptotic rate, leading a survival advantage to the mutated cells [16]. 2.4.6. cDNA Array Studies cDNA analyses have demonstrated that the expression profiling of endometrioid endometrial carcinoma is different from that of non-endometrioid endometrial carcinoma. These studies have identified gene signatures specific for non-endometrioid endometrial carcinomas as well as genes specifically up- or down-regulated in endometrioid endometrial carcinomas when compared with normal endometrium. Intestinal trefoil protein, TFF3, AGR2 developmental gene, estrogen-regulated genes (MGB2, LTF, END1, MMP11), FOXA2, and MSX2 were significantly up-regulated in endometrioid endometrial carcinomas, while increased expression of FOLR, genes involved in the regulation of mitotic spindle checkpoint (STK15, BUB1, CCNB2), IGF2, PTGS1 and p16 were seen in non-endometrioid endometrial carcinomas. STK-15 also known as BTAK, Aurora-A, is a serine/threonine kinase which is essential for chromosome segregation and centrosome functions [7, 9]. Overexpression of STK15 induces increased numbers of centrosomes, aneuploidy, and malignant transformation. One study found STK15 amplification in 9 of 15 (60%) non-endometrioid endometrial carcinomas but in none of endometrioid endometrial carcinomas [9]. Furthermore, a different expression profile was also found between endometrial carcinoma associated with MSI and stable endometrial carcinoma. SFRP1 and SFRP4 were more frequently down-regulated in endometrial carcinoma with MSI. One study compared the expression profiles of similar histological subtypes of ovarian and endometrial carcinomas, and showed that clear cell carcinomas had a very similar profile, regardless of the organ of origin. In contrast, differences were seen when comparing endometrioid and serous carcinomas of ovarian and endometrial origin [7].

the expression profiles of similar histological subtypes of ovarian and endometrial carcinomas, and showed that clear cell carcinomas had a very similar profile, regardless of the organ of origin. In contrast, differences were seen when comparing endometrioid and serous carcinomas of ovarian and endometrial origin [7]. 3. Genetic Changes in Endometrial Carcinogenesis (Progression Models) of Endometrioid and Serous Carcinomas, Including Molecular Changes of Premalignant Disease (Hyperplasia/EIC) By epidemiological and molecular evidence, endometrial hyperplasia represents a true precursor lesion for endometrioid endometrial carcinomas, whereas non-endometrioid endometrial carcinomas are frequently associated with endometrial intraepithelial carcinoma (EIC) [13]. 3.1. Progression Model for Endometrioid (Type I) Carcinomas A progression model of endometrioid carcinoma resembling the Vogelstein progression model for colorectal carcinoma has been proposed. This hypothesis is supported by the evidence that (i) some of the genetic alterations found in endometrioid endometrial carcinomas are already present in atypical hyperplasia, (ii) increased genetic alterations are found in well-differentiated endometrioid carcinoma compared with atypical hyperplasia, (iii) the number of genetic alterations increase according to higher histologic grade, and (iv) more chromosomal imbalances are identified in endometrial carcinoma compared with atypical hyperplasia, using comparative genomic hybridization (CGH) [11].

rentiated endometrioid carcinoma compared with atypical hyperplasia, (iii) the number of genetic alterations increase according to higher histologic grade, and (iv) more chromosomal imbalances are identified in endometrial carcinoma compared with atypical hyperplasia, using comparative genomic hybridization (CGH) [11]. Most simple hyperplasias and a subset of complex hyperplasias are polyclonal and considered reactive processes due to hyperestrogenism, which may regress through progesterone therapy [11, 23]. In contrast, most atypical hyperplasias are monoclonal. A subset of complex hyperplasia without atypia has been reported to be monoclonal. In addition, the number of chromosomal aberrations in complex hyperplasia is significantly higher than simple hyperplasia and close to the number found in atypical hyperplasia. Most of the genetic alterations identified in endometrioid endometrial carcinoma seem to occur very early in the development of endometrioid carcinoma, although it is not clear which alterations are associated with the earliest changes of malignant transformation and progression to neoplasia [10, 11]. In atypical hyperplasia, alterations of PTEN, β-catenin, KRAS, and MSI are present, with PTEN inactivation occuring in about 50% of the cases. However, PTEN and KRAS mutations seem to occur earlier, since they were found in simple hyperplasia, partially associated with monoclonality. PTEN inactivation has been reported in normal endometrial glands but its significance is yet unknown [11]. The inactivation of E-cadherin gene by methylation seems to play a role during progression of endometrioid carcinoma, since it is most frequently found in grade 3 and least frequently in grade 1 tumors [10]. Furthermore, p53 mutations, HER2/neu overexpression or amplification, and p16 inactivation are considered in late events during carcinogenesis of endometrioid carcinoma, since they are predominantly identified in grade 3 tumors, but rarely in grade 1 tumors, and are absent in atypical hyperplastic lesions. Hypothetically, p53 mutations and HER2/neu amplification might also be early events in de novo poorly differentiated endometrioid carcinomas [10, 11] (Figure 2). Endometrial pre-cancers (e.g., EIC) have been postulated to share common genetic alterations with endometrioid endometrial carcinoma, including PTEN mutations and MSI [13].

, p53 mutations and HER2/neu amplification might also be early events in de novo poorly differentiated endometrioid carcinomas [10, 11] (Figure 2). Endometrial pre-cancers (e.g., EIC) have been postulated to share common genetic alterations with endometrioid endometrial carcinoma, including PTEN mutations and MSI [13]. 3.2. Progression Model for Nonendometrioid (Type II) Carcinomas Mutations of p53 were found in approximately 80% of EIC, but in contrast to most serous carcinomas, there is no LOH at the locus TP53. Thus, it is hypothesized that p53 mutation of one allele occurs early, whereas loss of the normal second allele accompanies progression into serous carcinoma [10, 11]. The alterations of E-cadherin, p16, and HER2/neu seem to affect the progression from EIC to serous carcinoma [10]. Another group hypothesized that serous carcinoma may develop from endometrioid carcinoma through p53 mutation based on findings in mixed endometrioid and serous carcinomas. Early genetic alterations during carcinogenesis are not clear, as these authors presented no data for EIC [10, 11] (Figure 3).

rous carcinoma [10]. Another group hypothesized that serous carcinoma may develop from endometrioid carcinoma through p53 mutation based on findings in mixed endometrioid and serous carcinomas. Early genetic alterations during carcinogenesis are not clear, as these authors presented no data for EIC [10, 11] (Figure 3). 4. Hereditary Endometrial Carcinoma Hereditary endometrial carcinoma has been found in 2–5% of endometrial cancer [24]. Hereditary nonpolyposis colon cancer (HNPCC), also known as Lynch syndrome or cancer family syndrome, accounts for the majority of inherited cases [24]. It is an autosomal dominant syndrome that predisposes its carriers to multiple malignancies particularly colorectal, and endometrial carcinomas [25], caused by a germline mutation in one of the DNA MMR genes occurring in 30–60% of cases [8, 9]. Endometrial carcinoma is the most common extracolonic malignancy in patients with HNPCC. In women with HNPCC, the incidence of endometrial carcinoma equals or exceeds that of colorectal carcinoma, compared with 1% in the general population [26], and in more than 50% of HNPCC cases, these women present with a gynecological cancer as their first or “sentinel” malignancy [25]. The frequency of germline DNA MMR gene mutations among unselected patients with endometrial carcinoma has been found to be 1.8–2.1%, which is similar to the frequency of HNPCC in colorectal carcinoma [25]. Patients with endometrial carcinoma in the HNPCC population have an inherited germline mutation in MLH1, MSH2, MSH6, or PMS2 (first hit) but endometrial carcinoma develops only after the initiation of a deletion or mutation in the contralateral MLH1, MSH2, MSH6, or PMS2 allele (second hit) in endometrial cells [7, 8]. Once the 2 hits have occurred, the deficient MMR function of MLH1, MSH2, MSH6, or PMS2 causes the acquisition of MSI and subsequent tumor development [7–9]. Unlike HNPCC associated colorectal carcinoma, which appears to frequently have MLH1 and MSH2 mutations, endometrial carcinomas have a higher probability of MSH2 and MSH6 mutations [13, 25]. Women with HNPCC who carry MSH2 and MSH6 mutations have a higher chance to present initially with endometrial rather than colorectal cancer [16]. MSI has been detected in 75% of endometrial carcinoma associated with HNPCC [8, 9]. Many studies have shown that MSI is associated with endometrioid histologic type.

13, 25]. Women with HNPCC who carry MSH2 and MSH6 mutations have a higher chance to present initially with endometrial rather than colorectal cancer [16]. MSI has been detected in 75% of endometrial carcinoma associated with HNPCC [8, 9]. Many studies have shown that MSI is associated with endometrioid histologic type. However, 14–21% of HNPCC-associated endometrial carcinomas are non-endometrioid, but only 3.3–4.5% of sporadic MSI tumor [16]. Women with an inherited predisposition for endometrial neoplasia tend to develop the disease 10 years earlier than the general population [9]. There is 18–23% incidence of HNPCC syndrome in endometrial carcinoma patients younger than 50 years old [16]. In addition to endometrial carcinoma arising from HNPCC, occasional families show clustering of endometrial cancer alone, without colon or other cancers. This group was termed as “familial site-specific endometrial cancer” [10]. Loss of protein expression seems to occur frequently for both MLH1 and MSH2 in endometrial hyperplasia and is considered an early event during tumor development [11]. PTEN inactivation by mutation seems to also be involved in tumorigenesis, since it occurs in about 90% of type I carcinomas [11]. Currently, there are no data to suggest that the prognosis for women with HNPCC-associated endometrial cancers is either better or worse than for women with sporadic cancers [16, 24]. In one study, endometrial carcinoma in HNPCC kindreds was a cause of death in 12% of cases; in 61% of cases these patients had a second primary malignancy; and 15% of cases had more than 2 additional primary cancers. Nieminen et al. [26] studied serial endometrial biopsy samples taken during a 10-year followup of HPNCC mutation carriers and found abnormal MMR protein expression, MSI, or tumor suppressor promotor hypermethylation in various endometrial histologies, including normal and hyperplastic endometria. The most frequently methylated genes were CDH13, RASSF1A, and GSTP1. These defects in MMR and methylation appeared up to 12 years before endometrial carcinoma [26].

MR protein expression, MSI, or tumor suppressor promotor hypermethylation in various endometrial histologies, including normal and hyperplastic endometria. The most frequently methylated genes were CDH13, RASSF1A, and GSTP1. These defects in MMR and methylation appeared up to 12 years before endometrial carcinoma [26]. PTEN hamartoma tumor syndrome, caused by a germline mutation in PTEN gene on chromosome 10q, comprises a group of disorders including Cowden syndrome, Bannayan-Riley-Ruvalcaba syndrome, Proteus syndrome, Proteus-like syndrome, and autism spectrum disorder with macrocephaly [27, 28]. Cowden syndrome, also known as multiple hamartoma syndrome, is an autosomal dominant disorder with high risk of breast, thyroid, and endometrium cancer. The incidence of Cowden syndrome remains unclear due to underdiagnosis from variable penetration and subtle clinical findings [29]. Cowden syndrome is characterized by the development of intestinal hamartomas, facial trichilemmomas and mucocutaneous papillomatosis [29, 30], and is rarely identified before adulthood [28]. PTEN mutations in exon 5, coding for the active site and flanking amino acids, is a common site for mutations in patients with Cowden syndrome, and missense mutations are only found in this active area [30]. However, germline PTEN mutation has been detected in approximately 80% of Cowden syndrome patients [27]. The lifetime risk for endometrial carcinoma in Cowden syndrome is approximately 5–10%, compared with a 2.5% lifetime risk of women in the general population [27, 28]. Cowden syndrome and Bannayan-Riley-Ruvalcaba syndrome have overlapping phenotypic features. Bannayan-Riley-Ruvalcaba syndrome is a congenital, autosomal dominant condition manifested by macrocephaly, hamartomatous intestinal polyposis, lipomas, developmental delay or autism or both, and pigmented penile macules [29, 31]. Unlike Cowden syndrome, Bannayan-Riley-Ruvalcaba syndrome tends to be diagnosed at early life [28]. Approximately 60% of Bannayan-Riley-Ruvalcaba syndrome patients have an identifiable germline mutation in PTEN gene [27]. This syndrome also has the same increased risk of cancer as Cowden syndrome [29].

acules [29, 31]. Unlike Cowden syndrome, Bannayan-Riley-Ruvalcaba syndrome tends to be diagnosed at early life [28]. Approximately 60% of Bannayan-Riley-Ruvalcaba syndrome patients have an identifiable germline mutation in PTEN gene [27]. This syndrome also has the same increased risk of cancer as Cowden syndrome [29]. 5. Carcinosarcoma Carcinosarcomas (formerly known as malignant mixed mesodermal or mullerian tumors) are highly aggressive, biphasic neoplasms composed of carcinomatous and sarcomatous components (Figure 4). Carcinosarcomas account for 1-2% of all malignancies of uterine corpus [1] and usually present in postmenopausal women. Uterine bleeding is the most frequent presenting symptom. These tumors have traditionally been regarded as a subtype of uterine sarcomas or as a mixture of true carcinoma and sarcoma, but they are now regarded as metaplastic carcinomas or carcinomas with sarcomatous metaplasia [1, 18, 32–34]. Carcinosarcomas can be classified as type II endometrial carcinomas and their epithelial component can resemble high grade endometrioid, serous or clear cell carcinoma [35]. Etiologically, a few cases may be secondary to prior pelvic irradiation. In addition, an association between long term tamoxifen therapy and development of uterine carcinosarcoma has been suggested [32].

carcinomas and their epithelial component can resemble high grade endometrioid, serous or clear cell carcinoma [35]. Etiologically, a few cases may be secondary to prior pelvic irradiation. In addition, an association between long term tamoxifen therapy and development of uterine carcinosarcoma has been suggested [32]. Schipf et al. [36] analyzed a series of 30 paraffin-embedded carcinosarcomas, including 24 ovarian and 6 uterine, using fluorescence in situ hybridization (FISH) and CGH. Many carcinosarcomas contained aberrations on chromosome 8 and 20 detected by FISH. FISH showed C-MYC (8q24.12) and ZNF217 (20q13.2) amplification in 78% and 87%, respectively. The results demonstrate a uniform pattern of chromosomal gains and losses in CGH analysis. Gains or amplifications of 8q are the most common genetic aberration in carcinosarcomas [35]. One of the genes located within 8q is C-MYC (8q24) found to be amplified in 18 of 23 uterine and ovarian carcinosarcomas through FISH and overexpression in 9 of 9 uterine carcinosarcomas through immunostaining. C-MYC amplification is often present in carcinomas but was also present in 6 of 12 uterine leiomyomas and 11 of 23 uterine leiomyosarcomas [35].

is C-MYC (8q24) found to be amplified in 18 of 23 uterine and ovarian carcinosarcomas through FISH and overexpression in 9 of 9 uterine carcinosarcomas through immunostaining. C-MYC amplification is often present in carcinomas but was also present in 6 of 12 uterine leiomyomas and 11 of 23 uterine leiomyosarcomas [35]. LOH was seen in 5 of 6 uterine carcinosarcomas, and identical alleles were lost in the epithelial and mesenchymal components. p53 mutations and LOH for TP53 occur frequently in both carcinosarcoma components which are associated with frequent protein overexpression. Sherman et al. [18] reported immunoreactivity of p53 in 7 (70%) of 10 carcinosarcomas and noted that the similar staining pattern presented in both carinomatous and sarcomatous areas. In about 20% of carcinosarcomas, MSI-high was described with an 83% concordance between the carcinomatous and sarcomatous components [1]. One study found identical mutations of p53 and KRAS in the two components [33]. Fujii et al. [37] analyzed allelic status with polymorphous microsatellite markers on 172 carcinomatous/sarcomatous foci after microdissection of 17 carcinosarcomas. A close relationship between the carcinomatous and the sarcomatous component was found. No difference was seen in CGH patterns of carcinosarcomas [36]. Moreover, there is evidence that in most carcinosarcomas, the carcinomatous and the sarcomatous components are genetically the same, as shown for 21 of 25 carcinosarcomas (84%) using the human androgen receptor (HUMARA) for detection of X-chromosome inactivation. These results support a monoclonal origin of uterine carcinosarcomas and one can hypothesize that either the sarcomatous component develops from the carcinomatous component (conversion theory) or both are derived from a stem cell that undergoes divergent differentiation (combination theory) [33]. In the process of epithelial-mesenchymal transition, cells of epithelial origin lose epithelial characteristics and polarity acquiring a mesenchymal phenotype with increased migratory behavior. By molecular mechanisms, down-regulation of epithelial markers and up-regulation of mesenchymal markers result in acquisition of a fibroblast-like morphology with cytoskeleton reorganization and increase in motility, invasiveness, and metastatic capacity. A hallmark of epithelial-mesenchymal transition is loss of E-cadherin expression.

anisms, down-regulation of epithelial markers and up-regulation of mesenchymal markers result in acquisition of a fibroblast-like morphology with cytoskeleton reorganization and increase in motility, invasiveness, and metastatic capacity. A hallmark of epithelial-mesenchymal transition is loss of E-cadherin expression. A number of specific transcription factors, including Snail, Slug, SIP-1, and Twist, contribute to induction of epithelial mesenchymal transition, acting as transcriptional repressors of the E-cadherin gene. The oncogenic serine/threonine kinase AKT also promotes the process, modulating several signaling and transcriptional networks linking Wnt/β-catenin, NF-κB/p65, and Rb [38]. However, some investigators also found that a subset of carcinosarcoma was biclonal tumor, consisting of independent unrelated carcinomas and sarcomas, according to X chromosome inactivation and clinicopathological criteria [35]. In two collision reported tumors, the carcinomatous and sarcomatous components were histologically separate, with no intermingling, and there was a nodal metastasis that consisted purely of the sarcomatous element from one of these tumors [33]. One study examined 26 carcinosarcomas and found adenosarcoma-like components in 4 cases, suggesting that many of the true collision lesions may arise from malignant transformation of either benign epithelium within an adenosarcoma or adjacent benign endometrium [35]. The prognosis of the collision tumor depends on the most aggressive component, and may be better than for a carcinosarcoma of similar stage [33]. Overall, the carcinomatous component has been shown to have a more aggressive behavior and be a better predictor of clinical outcome in carcinosarcomas [35].

ndometrium [35]. The prognosis of the collision tumor depends on the most aggressive component, and may be better than for a carcinosarcoma of similar stage [33]. Overall, the carcinomatous component has been shown to have a more aggressive behavior and be a better predictor of clinical outcome in carcinosarcomas [35]. 6. Endometrial Stromal Sarcoma and Undifferentiated Endometrial Sarcoma Endometrial stromal sarcoma and undifferentiated endometrial sarcoma are in the same neoplastic spectrum. Diagnosis of endometrial stromal tumors has been based on histologic criteria. Low grade endometrial stromal sarcoma is composed of uniform, oval to spindle-shape cells of endometrial stromal-type with numerous small arterioles that resemble the spiral arterioles of late secretory endometrium. Mitotic rate is not a consideration in the distinction between low and high grade stromal sarcoma. In addition, characteristic tongue-like growth of the stromal cells into the myometrium and/or myometrial vasculature is noted [39, 40] (Figure 5). Endometrial stromal sarcoma usually occurs in middle aged women [41], and most present with uterine bleeding. Undifferentiated endometrial sarcoma, on the other hand, is defined as a high-grade neoplasm that lacks specific differentiation and bears no histological resemblance to endometrial stroma. Also, undifferentiated endometrial sarcomas have marked nuclear pleomorphism with high mitotic rate and display destructive myometrial invasion [40, 42]. Undifferentiated endometrial sarcomas should be diagnosed only after extensive sampling to exclude smooth or skeletal muscle differentiation, to exclude high grade leiomyosarcoma or rhabdomyosarcoma. Carcinosarcoma or adenosarcoma with sarcomatous overgrowth should also be excluded before making the diagnosis of undifferentiated endometrial sarcomas [39, 40].

uld be diagnosed only after extensive sampling to exclude smooth or skeletal muscle differentiation, to exclude high grade leiomyosarcoma or rhabdomyosarcoma. Carcinosarcoma or adenosarcoma with sarcomatous overgrowth should also be excluded before making the diagnosis of undifferentiated endometrial sarcomas [39, 40]. In endometrial stromal sarcomas, the tumor cells are typically immunoreactive for estrogen and progesterone receptors, CD10, vimentin, and sometimes focally with actin, while they are generally negative for desmin, and h-caldesmon. Expression of androgen receptor is observed in 41% of examined sarcoma cases [41]. Approximately 70% of low grade endometrial stromal sarcomas also expresses epidermal growth factor receptor (EGFR; HER1). Undifferentiated endometrial sarcomas are estrogen and progesterone receptor negative, but a high proportion is EGFR positive. Endometrial stromal sarocmas are typically diploid with a low S-phase fraction whereas S-phase fraction exceeds 10% in undifferentiated endometrial sarcomas [41]. No c-kit (CD117) expression has been demonstrated in endometrial stromal sarcomas [43]. Liegl et al. [43] found 22 of 37 endometrial stromal sarcomas showed platelet-derive growth factor (PDGF)-α (CD140α) and 8 of 37 endometrial stromal sarcomas showed PDGF-β expression.

% in undifferentiated endometrial sarcomas [41]. No c-kit (CD117) expression has been demonstrated in endometrial stromal sarcomas [43]. Liegl et al. [43] found 22 of 37 endometrial stromal sarcomas showed platelet-derive growth factor (PDGF)-α (CD140α) and 8 of 37 endometrial stromal sarcomas showed PDGF-β expression. In contrast to epithelial endometrial carcinoma, endometrial stromal tumors are characterized by distinct cytogenetic abnormalities, particularly translocations leading to gene fusion. Cytogenetic studies reported to-date are primarily for low grade endometrial stromal sarcomas, mostly showing rearrangement of chromosomes 6, 7, and 17 [44, 45]. Loss of chromosome arm 7p (55.6% of the cases) is the most frequent aberration and may play a role in tumor development and progression [41]. Reverse transcription polymerase chain reaction (RT-PCR) and FISH studies on large series showed the presence of t(7;17)(p15;q21), leading to the fusion of two zinc finger genes, JAZF1 (juxtaposed with another zinc finger gene 1) and JJAZ1 (joined to JAZF1; also named SUZ12, suppressor of zeste-12 protein). JAZF1 is expressed in normal endometrial stroma, but the specific functions of the JAZF1 and the JJAZ1 genes as well as the JAZF1/JJAZ1 fusion gene are still unknown [1, 42]. Based on the evidence of loss of expression for normal versions of JAZF1 in multiple tumors suggests a possible role of this gene as a tumor suppressor [42]. This gene fusion is a distinctive molecular genetic alteration for endometrial stromal sarcoma and benign endometrial stromal nodules [1, 41]. The JAZF1/JJAZ1 fusion gene is frequently present in classical endometrial stromal sarcomas and less often in cases with variant histology [46]. However, of seven high-grade endometrial stromal sarcomas/undifferentiated endometrial sarcomas studied, only three cases showed evidence of the fusion [42]. In contrast, many studies reported the fusion gene to be absent in undifferentiated endometrial sarcomas [41]. The fusion gene is not present in normal endometrial stroma [41]. The presence of the JAZF1/JJAZ1 fusion gene within the spectrum of endometrial stromal tumors from benign to malignant raises possibility that the endometrial stromal nodule may transform into malignant endometrial stromal sarcoma [41, 47]. The frequencies of this gene fusion in low grade endometrial stromal sarcoma have been reported in many studies showing a wide range of positivity, 23–80% [44, 46, 48, 49].

from benign to malignant raises possibility that the endometrial stromal nodule may transform into malignant endometrial stromal sarcoma [41, 47]. The frequencies of this gene fusion in low grade endometrial stromal sarcoma have been reported in many studies showing a wide range of positivity, 23–80% [44, 46, 48, 49]. The studies with RT-PCR only can give false-positive results due to PCR contamination. FISH may be useful as a complementary technique to exclude the possibility of false positive contamination of cases by RT-PCR [44]. Although the JAZF1/JJAZ1 fusion gene seems to be the major molecular alterations in endometrial stromal sarcomas, there is some evidence for alternative pathways in the development of endometrial stromal sarcomas. A major subgroup of endometrial stromal sarcomas has been found to have translocations involving short arm of chromosome 6, particularly band 6p21 [41, 44, 45]. Micci et al. [50] showed that the PHF1 (PHD finger protein 1) gene in 6p21 was recombined with two different partners, (i) with JAZF1 gene showing a 6p;7p rearrangement, which results in the formation of a JAZF1/PHF1 fusion gene and (ii) with EPC1 (enhancer of polycomb) gene in 10p11 that had a 6;10;10 translocation. Panagopoulos et al. [45] introduced that a low-grade endometrial stromal sarcoma cell line carrying a der(7)t(6;7)(p21;p22) also harbors a JAZF1/PHF1 fusion. Both t(7;17) and t(6;7) comprise 62% of the reported endometrial stromal sarcomas [44]. Additionally, few endometrial stromal sarcoma cases were reported with a t(X;17)(p11.2;q23) and a t(10;17)(q22;p13) [51–53]. Although JAZF1/JJAZ1 fusion may not be universally present in all low grade endometrial stromal sarcoma, this aberration may still be diagnostically useful [44]. The JAZF1/JJAZ1 fusion has been identified in areas of smooth muscle differentiation in endometrial stromal neoplasms (50% of the cases). This finding supports that the endometrial stromal and smooth muscle components of these tumors have the same origin, either from a common precursor cell with pluripotential differentiation or from endometrial stromal cells that have undergone smooth muscle metaplasia [44, 54]. Halbwedl et al. [55] described 9 cases of low grade endometrial stromal sarcoma and 3 cases of undifferentiated endometrial sarcoma in aCGH study revealing a variety of gains and losses that apparently did not correlate with morphology.

om endometrial stromal cells that have undergone smooth muscle metaplasia [44, 54]. Halbwedl et al. [55] described 9 cases of low grade endometrial stromal sarcoma and 3 cases of undifferentiated endometrial sarcoma in aCGH study revealing a variety of gains and losses that apparently did not correlate with morphology. There is no accumulation of aberrations in undifferentiated endometrial sarcoma compared to endometrial stromal sarcoma, indicating these two types of uterine sarcomas are probably not related to each other.

om endometrial stromal cells that have undergone smooth muscle metaplasia [44, 54]. Halbwedl et al. [55] described 9 cases of low grade endometrial stromal sarcoma and 3 cases of undifferentiated endometrial sarcoma in aCGH study revealing a variety of gains and losses that apparently did not correlate with morphology. There is no accumulation of aberrations in undifferentiated endometrial sarcoma compared to endometrial stromal sarcoma, indicating these two types of uterine sarcomas are probably not related to each other. LOH and MSI have been evaluated in both low grade endometrial stromal sarcomas (20 cases) and undifferentiated endometrial sarcomas (3 cases). LOH with at least one polymorphic DNA marker was identified in all 3 cases (100%) of undifferentiated endometrial sarcomas, 10 (50%) low-grade endometrial stromal sarcomas and 2 (50%) benign endometrial stromal nodules. Moreover, concurrent and independent LOH were noted in adjacent normal appearing myometrium or endometrium, either close to or at a distance from the tumors [41]. LOH was mostly identified at PTEN, a tumor suppressor gene located on chromosome 10q [56]. No tumor was associated with MSI [41, 55]. Loss of functions of certain tumor suppressor genes such as PTEN in surrounding nontumor uterine tissues could influence and facilitate tumor proliferation, cellular spread, and invasion by malignant endometrial stromal cells [41]. However, one should keep in mind the false positive scoring of LOH in normal tissues may occur both from the imperfect methodology and from contamination by tumor samples/cells. The use of repeated experiments and several polymorphic markers has been advised to overcome these methodology problems [56]. Other frequently altered loci by LOH were at 14q32 (D14S267) and 3p (D3S1300). The former locus is frequently altered in uterine leiomyosarcoma but, in addition, in a variety of epithelial neoplasms such as ovarian, colorectal and esophageal carcinoma. Locus D3S1300 harbors the FHIT gene which is frequently mutated in cervical carcinoma of the uterus. LOH for TP53 and p53 overexpression are rarely present in endometrial stromal sarcomas (5% and 15%, resp.). The importance of p53 mutations for the development of undifferentiated endometrial sarcomas is not evident, but p53 overexpression was detected in three of four high-grade stromal sarcomas [1]. Furthermore, Kurihara et al. [49] have recently found frequent nuclear accumulation of p53 and TP53 gene missense mutations in undifferentiated endometrial sarcoma with nuclear pleomorphism, 3 (50%) of 6 cases.

arcomas is not evident, but p53 overexpression was detected in three of four high-grade stromal sarcomas [1]. Furthermore, Kurihara et al. [49] have recently found frequent nuclear accumulation of p53 and TP53 gene missense mutations in undifferentiated endometrial sarcoma with nuclear pleomorphism, 3 (50%) of 6 cases. There is no evidence of p53 aberration in 18 low grade endometrial stromal sarcomas and 7 cases undifferentiated endometrial sarcoma with nuclear uniformity. p53 alteration may be one different pathway that contributes the tumorigenesis of undifferentiated endometrial sarcoma. Expression of SFRP4 and β-catenin is also detected. SFRP4 acts in Wnt-signaling pathway, which is a complex cascade of heterogeneous molecules playing an important role in organ development, via β-catenin. SFRP4 is expressed in normal endometrial stromal cells but not in glandular epithelium. Compared with normal endometrium, the expression of SFRP4 was decreased in both low grade endometrial stromal sarcomas and undifferentiated endometrial sarcomas. Through its involvement in the Wnt signaling pathway, SFRP4 may act as a tumor suppressor by regulating the cytosolic β-catenin pool in the cell. Beta-catenin regulates in the opposite manner to SFRP4, being particularly increased in undifferentiated sarcoma [57]. Dysregulation of these pathways allows β-catenin to accumulate and translocate to the nucleus, where it forms complexes with T-cell factor/lymphoid enhancing factor (TCF/LEF) leading to uncontrolled cell growth and carcinogenesis [57]. High level nuclear staining for β-catenin was seen in 40% of endometrial stromal sarcomas and may be used as a diagnostic tool [42].

to accumulate and translocate to the nucleus, where it forms complexes with T-cell factor/lymphoid enhancing factor (TCF/LEF) leading to uncontrolled cell growth and carcinogenesis [57]. High level nuclear staining for β-catenin was seen in 40% of endometrial stromal sarcomas and may be used as a diagnostic tool [42]. 7. Diagnostic Utility Based on the Molecular Knowledge 7.1. Endometrioid Carcinoma versus Serous/Clear Cell Carcinoma At times, the histological type of endometrial carcinoma is not clearly defined, especially in poorly differentiated tumors, and knowledge of the dualistic model, with the common molecular changes in each type, can help clarify the diagnosis. If there is non-carcinomatous endometrium present, the presence of hyperplasia is supportive evidence of an endometrioid carcinoma, whereas atrophic endometrium is supportive of non-endometrioid carcinoma. Molecular studies on endometrium are not often performed in most hospital surgical pathology laboratories today; however, immunohistochemical studies can detect the abnormal protein products of the gene mutations. Therefore, we can exploit our knowledge of the dualistic model and their typical gene mutations and use the immunoprofile as a diagnostic tool, in concert with the histomorphologic features to specify the tumor type, particularly in difficult cases such as in the differentiation between high-grade endometrioid carcinoma and serous carcinoma (Table 3).

wledge of the dualistic model and their typical gene mutations and use the immunoprofile as a diagnostic tool, in concert with the histomorphologic features to specify the tumor type, particularly in difficult cases such as in the differentiation between high-grade endometrioid carcinoma and serous carcinoma (Table 3). 7.2. Endometrial Stromal Sarcoma versus Undifferentiated Sarcoma The distinct molecular alteration described in the majority of endometrial stromal sarcomas is the t(7;17)(p15;q21) leading to the formation of fusion gene JAZF1/JJAZ1, which can be detected by RT-PCR or FISH assays. Thus, in the problematic cases in which the differentiatial diagnosis is between endometrial stromal sarcoma and undifferentiated sarcoma, we look for the fusion gene to make this distinction.

as is the t(7;17)(p15;q21) leading to the formation of fusion gene JAZF1/JJAZ1, which can be detected by RT-PCR or FISH assays. Thus, in the problematic cases in which the differentiatial diagnosis is between endometrial stromal sarcoma and undifferentiated sarcoma, we look for the fusion gene to make this distinction. 7.3. Uterine Smooth Muscle Neoplasm versus Endometrial Stromal Tumors Uterine smooth muscle neoplasm is defined as a mesenchymal tumor composed of cells with smooth muscle differentiation, particularly highly cellular leiomyomas may have morphologically overlapped features of endometrial stromal tumors. According to histologic criteria for differential, immunostainings may help to correct the final diagnosis, particularly in difficult cases. Neoplastic endometrial stromal cells typically express vimentin, muscle-specific and smooth muscle actin and may be positive for desmin. In addition, CD10, initially thought to be a specific marker for endometrial stromal tumors, can be demonstrated in uterine smooth muscle tumors, commonly in highly cellular leiomyomas and leiomyosarcomas. Other antibodies that give positive staining in smooth muscle tumors useful in this differential diagnosis includes h-caldesmon, histone deacetylase 8 (HDAC8), smooth muscle myosin and oxytocin receptor [39]. However, none of these markers can completely specify the smooth muscle/endometrial stroma lineage of the tumor, a panel of the antibodies should be used [39].

smooth muscle tumors useful in this differential diagnosis includes h-caldesmon, histone deacetylase 8 (HDAC8), smooth muscle myosin and oxytocin receptor [39]. However, none of these markers can completely specify the smooth muscle/endometrial stroma lineage of the tumor, a panel of the antibodies should be used [39]. The molecular alterations in smooth muscle tumor are complex, especially in leiomyosarcomas. The translocation t(12;14)(q15;q23-24) has been noted in a high proportion of leiomyomas [41, 58]. By CGH, leiomyosarcomas have the most frequent losses including 10q, 11q, 13q, and 2p while the most common gains are Xp, 1q, 5p, 8q, 12q, 17p and 19p [41, 59, 60]. There are a variety of genetic changes and mutations inclusive of TP53 and MDM2 expression associated with progression of leiomyosarcomas [41]. LOH of 10q is found in more than half of leiomyosarcomas [41]. Leiomyosarcomas exhibit a significantly higher frequency of allelic loss (52%) compared with benign leiomyomas (18%) and smooth muscle tumors of uncertain malignant potential (21%) [41].

d MDM2 expression associated with progression of leiomyosarcomas [41]. LOH of 10q is found in more than half of leiomyosarcomas [41]. Leiomyosarcomas exhibit a significantly higher frequency of allelic loss (52%) compared with benign leiomyomas (18%) and smooth muscle tumors of uncertain malignant potential (21%) [41]. 8. Therapeutic Considerations: Molecular Targeted Therapy Development of targeted anticancer drugs is the direct result of knowledge of the molecular profile of endometrial neoplasms. Drug targets may focus on genes that affect apoptosis, signal transduction, epigenetic modification, drug resistance, protein folding and degradation, cell cycle progression, hormone receptor activity, and angiogenesis [4]. The drugs that comprise targeted therapy include small molecular weight inhibitors, monoclonal antibodies, antisense and gene therapy [61]. At this time, essentially only endometrial carcinomas have been tested with targeted therapy. Carcinosarcomas and endometrial stromal sarcomas are relatively uncommon neoplasms, and there has little experience with specific therapies for these tumors, though there is definitely future potential.

and gene therapy [61]. At this time, essentially only endometrial carcinomas have been tested with targeted therapy. Carcinosarcomas and endometrial stromal sarcomas are relatively uncommon neoplasms, and there has little experience with specific therapies for these tumors, though there is definitely future potential. 8.1. mTOR Inhibitors The phosphatidylinositol-3-kinase (PI3K)-serine/threonine kinase (AKT)-mammalian target of the rapamycin (mTOR) signaling pathway plays a central role in the regulation of cell growth, proliferation, and apoptosis. In in vitro studies, cells with PTEN inactivation in endometrioid carcinoma are sensitive to mTOR inhibitors, since the loss of PTEN leads to constitutive activation of downstream components, which in turn up-regulates mTOR activity [62]. Potential therapies targeting the mTOR pathway include the mTOR inhibitors temsirolimus (CCI-779), everolimus (RAD001), and deforolimus (AP23573) [4]. In a phase II study of temsirolimus activity in patients with advanced or recurrent endometrial cancer, 5 of 19 (26%) evaluable patients had a partial response and 12 (63%) had stable disease [62, 63]. In addition to mTOR inhibitors, other agents targeting components of the mTOR-AKT-PI3K-PTEN pathway have also been developed, including enzastaurin (a PI3K inhibitor) and triciribine (an AKT inhibitor) [4].

metrial cancer, 5 of 19 (26%) evaluable patients had a partial response and 12 (63%) had stable disease [62, 63]. In addition to mTOR inhibitors, other agents targeting components of the mTOR-AKT-PI3K-PTEN pathway have also been developed, including enzastaurin (a PI3K inhibitor) and triciribine (an AKT inhibitor) [4]. 8.2. EGFR Inhibitors/Anti-HER2/neu Epidermal growth factor receptor (EGFR) family members (ERBB1 (EGFR or HER1), ERBB2 (HER2/neu), ERBB3 (HER3), ERBB4 (HER4)) are tyrosine kinase receptors that are activated by binding to epidermal growth factor (EGF)-like growth factor, leading to downstream phosphorylation or dephosphorylation of signaling molecules that involved in cell cycle and apoptosis [63]. Sixty to 80% of endometrial carcinomas overexpress EGFR [4]. In addition, EGFR expression has been described in approximately 70% of endometrial stromal sarcomas [41]. EGFR overexpression has been reported in high grade carcinomas with deep myometrial invasion, positive peritoneal washings and poor survival [61, 63]. The anti-EGFR agents result in down regulation of the MAPK and PI3K/AKT signaling pathways. However, the anti-tumor activity has been described in a minority of the patients treated. Antagonists to EGFR include small molecule tyrosine kinase inhibitors (gefitinib, erlotinib, and lapatinib) and the anti-EGFR monoclonal antibody cetuximab [62]. Experimental observation data have been shown that EGFR inhibitors could be more effective in endometrioid endometrial carcinoma than in uterine papillary serous carcinoma [63].

include small molecule tyrosine kinase inhibitors (gefitinib, erlotinib, and lapatinib) and the anti-EGFR monoclonal antibody cetuximab [62]. Experimental observation data have been shown that EGFR inhibitors could be more effective in endometrioid endometrial carcinoma than in uterine papillary serous carcinoma [63]. As described above, HER2/neu gene overexpression and amplification have been found in up to 80% of nonendometrioid endometrial carcinoma, and in 10–30% of endometrioid endometrial carcinoma. The usage of trastuzumab, a monoclonal antibody directed against HER2/neu, has been tested in endometrial carcinomas. Villella's group found 5 out of 19 (26%) patients with papillary serous carcinoma showed HER2/neu overexpression. One of 5 positive HER2/neu patients with advanced disease treated with trastuzumab achieved a complete response and a second patient's disease stabilized [63]. However in another study, Gynecologic Oncology Group (GOG)-0181-B, investigated trastuzumab in advanced, recurrent, or persistent endometrial cancer, and its preliminary results showed minimal activity, even in cancers with high overexpression of HER2/neu [4]. Several other monoclonal antibodies targeting members of the ERBB/HER family, including pertuzumab, cetuximab, and panitumumab, are currently being investigated [4].

ecurrent, or persistent endometrial cancer, and its preliminary results showed minimal activity, even in cancers with high overexpression of HER2/neu [4]. Several other monoclonal antibodies targeting members of the ERBB/HER family, including pertuzumab, cetuximab, and panitumumab, are currently being investigated [4]. 8.3. Antiangiogenics Vascular endothelial growth factors (VEGF) expression has been found in 56–100% of endometrial carcinomas [63] and has been correlated with high histologic grade, deep myometrial invasion, angiolymphatic invasion, nodal metastasis, and short disease-free survival [63, 64]. VEGF, particularly VEGF-A, plays a key role in angiogenesis and increased permeability of tumor-associated blood vessels. Monoclonal antibodies targeting VEGF, bevacizumab and sorafenib, have been developed. Kamat and coworkers [64] injected Ishikawa cell line into uterine horn of nude mice in one group and Hec-1A cell lines in the other group and treated the mice with docetaxel and/or bevacizumab. The combination of both agents had a greater efficacy in tumor growth inhibition than a single agent. Currently, GOG-229-E is being studied in a phase II trial of single agent bevacizumab in patients with recurrent endometrial carcinoma [64].

es in the other group and treated the mice with docetaxel and/or bevacizumab. The combination of both agents had a greater efficacy in tumor growth inhibition than a single agent. Currently, GOG-229-E is being studied in a phase II trial of single agent bevacizumab in patients with recurrent endometrial carcinoma [64]. 9. Conclusion Knowledge of the molecular profiles of endometrial neoplasms assists in the diagnosis, prognosis, and treatment of endometrial neoplasms. Endometrial carcinoma can be broadly divided into two categories based on clinical behavior and morphologic phenotype, with good correlation to the molecular findings. Type I endometrial carcinoma represents an estrogen-related tumor, which usually arises in the setting of endometrial hyperplasia and have good prognosis. They are associated with a number of well-described genetic alterations including mutations of PTEN, KRAS, β-catenin, PIK3CA, and inactivation of DNA mismatch repair. Targets for molecular therapy in endometrial carcinoma include agents that inhibit components of the AKT-PI3K-PTEN pathway. Type II endometrial cancers are not estrogen-related and have poor prognosis. Mutations of p53 are present in approximately 90% of this tumor type. Carcinosarcoma is considered to be a high-grade carcinoma with sarcomatous differentiation and a high frequency of C-MYC mutations and LOH of p53. The majority of endometrial stromal nodules and stromal sarcomas seem to originate from the abnormal JAZF1/JJAZ1 gene fusion. The molecular biology of undifferentiated endometrial sarcomas is still not clearly delineated. In the near future; additional molecular studies should further elucidate the unclear pathogenesis and provide new targets for diagnosis and treatment.

al sarcomas seem to originate from the abnormal JAZF1/JJAZ1 gene fusion. The molecular biology of undifferentiated endometrial sarcomas is still not clearly delineated. In the near future; additional molecular studies should further elucidate the unclear pathogenesis and provide new targets for diagnosis and treatment. Figure 1 The prototypes for the dualistic model of endometrial carcinoma. Type I endometrioid endometrial carcinoma shows glands lined by stratified neoplastic columnar cells (a), ×100; and type II serous carcinoma showing papillary structures and high nuclear grade (b), ×100. Figure 2 A progression model for endometrioid carcinoma. Tumor initiation and progression are characterized by acquisition of various molecular alterations. PTEN alterations appear central to the initiation of proliferative lesions that then acquire mutations in other cancer-causing genes (e.g., DNA mismatch repair genes, KRAS, β-catenin) in the carcinogensis. An alternative pathway bypasses atypical hyperplasia and low-grade carcinoma to high-grade carcinoma by p53 mutation and HER2/neu amplification. NE, normal endometrium; EH, endometrial hyperplasia without hyperplasia, AH, atypical endometrial hyperplasia; EIC, endometrial intraepithelial carcinoma; LG-ECC, low grade endometrioid endometrial carcinoma; HG-ECC, high grade endometrioid endometrial carcinoma.

de carcinoma by p53 mutation and HER2/neu amplification. NE, normal endometrium; EH, endometrial hyperplasia without hyperplasia, AH, atypical endometrial hyperplasia; EIC, endometrial intraepithelial carcinoma; LG-ECC, low grade endometrioid endometrial carcinoma; HG-ECC, high grade endometrioid endometrial carcinoma. Figure 3 A progression model for nonendometrioid (type II) carcinomas. p53 mutations play a critical role in the conversion of atrophic endometrium to an intraepithelial form of serous carcinoma. NE, normal endometrium; EIC, endometrial intraepithelial carcinoma; NEEC, non-endometrioid endometrial carcinoma. Figure 4 Carcinosarcoma is composed of two malignant components, carcinomatous and sarcomatous. The epithelial component is usually high grade carcinoma for example, serous/clear cell type. The mesenchymal part comprises either homologous (a), ×100 or heterologous element for example, cartilage or bone. Chondrosarcomatous element (∗) is present in (b), ×100. Figure 5 Endometrial stromal sarcoma, low grade is circumscribed from the surrounding myometrium (a), ×40; and higher magnification of endometrial stromal sarcoma shows round uniform tumor cells resembling the stroma of proliferative endometrium with low mitotic rate (b), ×200. Table 1 Clinical and pathological characteristics of type I and type II endometrial carcinoma [1, 2, 4–13].

Figure 5 Endometrial stromal sarcoma, low grade is circumscribed from the surrounding myometrium (a), ×40; and higher magnification of endometrial stromal sarcoma shows round uniform tumor cells resembling the stroma of proliferative endometrium with low mitotic rate (b), ×200. Table 1 Clinical and pathological characteristics of type I and type II endometrial carcinoma [1, 2, 4–13]. Type I Type II Proportion of endometrial carcinomas 4/5 1/5 Menstrual status Pre- and perimenopausal Postmenopausal Endocrine-metabolic disturbance Present Absent Estrogen-associated Yes No Background endometrium Hyperplasia Atrophy Histological type Endometrioid Serous, clear cell Tumor grade Low High Depth of myometrial invasion Superficial Deep Behavior Stable/indolent Progressive/aggressive Table 2 Genetic alterations of type I and type II endometrial carcinomas, reported in percentages (references). Type I Type II PTEN inactivation Up to 83% [1, 4, 11, 12] 11% [1, 2, 12] PIK3CA mutation 26–36% [7, 9] 5% [7] KRAS mutation 10–30% [1, 2, 4, 7–12, 17] 0–10% [2, 12] β-catenin /CTNNB1 mutation 14–44% [7, 8] 0–5% [1, 7, 10, 11] Microsatellite instability 20–45% [1, 7–10] 0–11% [8, 9] p53 mutation 10–20% [1, 4, 6, 7, 10, 11, 13, 17, 18] 90% [1, 2, 4, 6, 7, 10–13, 17] HER2/neu amplification 10–30% [1, 4, 10, 17] 18–80% [13] p16 inactivation 10% [1, 4, 7, 10, 11] 40–45% [4, 7, 10] E-cadherin loss 10–20% [1, 4, 7, 10, 11] 60–90% [4] Table 3 Typical immunoprofile of type I endometrioid carcinoma and type II serous carcinoma.

, 4, 6, 7, 10, 11, 13, 17, 18] 90% [1, 2, 4, 6, 7, 10–13, 17] HER2/neu amplification 10–30% [1, 4, 10, 17] 18–80% [13] p16 inactivation 10% [1, 4, 7, 10, 11] 40–45% [4, 7, 10] E-cadherin loss 10–20% [1, 4, 7, 10, 11] 60–90% [4] Table 3 Typical immunoprofile of type I endometrioid carcinoma and type II serous carcinoma. Endometrioid carcinoma Serous carcinoma Estrogen and progesterone receptors + − PTEN − + β-catenin + − p53 − + HER2/neu − + +: positive result, −: negative result

1. Introduction Endometrial cancer is the most common cancer of the female reproductive tract with 150,000 new cases diagnosed annually worldwide. Approximately 90% of endometrial cancers are sporadic, and the remaining 10% are hereditary. Bokhman have generally categorized endometrial cancer into two broad groups of tumors using both clinical and histopathological variables: estrogen-dependent endometrioid endometrial carcinomas (EECs), or type I, and non-endometrioid endometrial carcinomas (NEECs), or type II tumors (Table 1) [1]. It should be noted that this model is not strict, and only a minority of endometrial cancer may exhibit shared characteristics. For example, mixed serous and endometrioid tumors are being increasingly recognized. Approximately 70% to 80% of new cases are classified as EECs, and other 10% to 20% are designated as NEEC tumors [1]. EECs are strongly associated with the estrogen-related pathway and arise in association with unopposed estrogen stimulation [2]. In contrast, NEECs are unrelated to the estrogen pathways and arise in the background of atrophic endometrium [3]. EECs typically occur in premenopausal and younger postmenopausal women and are usually low-grade and have a favorable outcome, whereas NEECs occur in older postmenopausal women. In addition, NEECs tend to predict a high tumor grade and poor patient prognosis [4, 5]. The first pathway is associated with endometrioid histopathology, and the second is linked to the serous and clear cell subtypes. The precursors of these subtypes are known as atypical hyperplasia and endometrial intraepithelial carcinoma (EIC), respectively. Clear cell cancer, classified as an NEEC, is associated with atypical hyperplasia as well as EIC.

endometrioid histopathology, and the second is linked to the serous and clear cell subtypes. The precursors of these subtypes are known as atypical hyperplasia and endometrial intraepithelial carcinoma (EIC), respectively. Clear cell cancer, classified as an NEEC, is associated with atypical hyperplasia as well as EIC. Recent reports suggest that histological differences may be associated with distinct molecular genetic alterations. Molecular genetic evidence indicates that endometrial carcinomas are likely to develop as the result of a multistep process of oncogenic activation and tumor suppressor inactivation (Table 2) [6]. 2. Gain-of-Function Genetic Events The genes implicated in the gain-of-function events are oncogenes. The important genes related to endometrial oncogenesis or progressions are the K-ras,B-raf, Her2/neu, β-catenin, AKT, and FGFR2 oncogenes. 2.1. K-ras and B-raf

Recent reports suggest that histological differences may be associated with distinct molecular genetic alterations. Molecular genetic evidence indicates that endometrial carcinomas are likely to develop as the result of a multistep process of oncogenic activation and tumor suppressor inactivation (Table 2) [6]. 2. Gain-of-Function Genetic Events The genes implicated in the gain-of-function events are oncogenes. The important genes related to endometrial oncogenesis or progressions are the K-ras,B-raf, Her2/neu, β-catenin, AKT, and FGFR2 oncogenes. 2.1. K-ras and B-raf K-ras proto-oncogene mutations are detected in approximately 10%–30% of endometrioid carcinomas [7]. K-ras mutations have been identified in endometrial hyperplasias, although at a lower frequency than in carcinomas [8–10]. According to these studies, the gain of the K-ras function may represent an early event in endometrioid-type tumorigenesis. During tumorigenesis, activated RAS is usually associated with enhanced proliferation, transformation, and cell survival. Conversely, K-ras mutations occur with equal frequency in tumors with and without hyperplasia, and the epidemiologic results seem to suggest that K-ras activation is associated with malignant progression of endometrial tumors without the need for transition via hyperplasia [11]. In contrast to endometrioid carcinomas, K-ras mutations are extremely rare among serous and clear cell carcinomas [12, 13].

yperplasia, and the epidemiologic results seem to suggest that K-ras activation is associated with malignant progression of endometrial tumors without the need for transition via hyperplasia [11]. In contrast to endometrioid carcinomas, K-ras mutations are extremely rare among serous and clear cell carcinomas [12, 13]. A correlation between colon cancer development and Ras/Raf point mutations in the MAP kinase pathway drives the malignant transformation of colon cancer. In contrast, only a few reports have shown B-raf mutations in patients with endometrial cancer. Feng et al. identified a B-raf mutation in 21% of patients with endometrial cancers and suggest that the mutation correlated with decreased hMLH1 expression [14]. In contrast, Salevesen et al. described a B-raf mutation in only 2% of endometrial cancers; and Kawaguchi et al. and Mizumoto et al. reported no mutation in the patients with endometrial cancer [15–17]. Therefore, a consensus about the role of B-raf mutation in the development of endometrial cancer has not yet been developed. 2.2. Her2/neu

A correlation between colon cancer development and Ras/Raf point mutations in the MAP kinase pathway drives the malignant transformation of colon cancer. In contrast, only a few reports have shown B-raf mutations in patients with endometrial cancer. Feng et al. identified a B-raf mutation in 21% of patients with endometrial cancers and suggest that the mutation correlated with decreased hMLH1 expression [14]. In contrast, Salevesen et al. described a B-raf mutation in only 2% of endometrial cancers; and Kawaguchi et al. and Mizumoto et al. reported no mutation in the patients with endometrial cancer [15–17]. Therefore, a consensus about the role of B-raf mutation in the development of endometrial cancer has not yet been developed. 2.2. Her2/neu Her2/neu (erbB2) is an oncogene that encodes a transmembrane receptor tyrosine kinase involved in cell signaling. Either the overexpression or gene amplification of Her2/neu proto-oncogene activates receptor and soluble tyrosine kinases. Her2/neu overexpression is detected in about 10%–20% of Grades 2 and 3 endometrioid carcinoma [9, 18, 19]. These studies suggest that Her2/neu overexpression in endometrioid carcinoma characterizes late progression and differentiation events. Her2/neu overexpression is detected in approximately 9%–30% of serous carcinomas [20]. Elucidation of the role of Her2/neu in these pathogenic tumor types, therefore, requires further study. 2.3. β-Catenin

Her2/neu (erbB2) is an oncogene that encodes a transmembrane receptor tyrosine kinase involved in cell signaling. Either the overexpression or gene amplification of Her2/neu proto-oncogene activates receptor and soluble tyrosine kinases. Her2/neu overexpression is detected in about 10%–20% of Grades 2 and 3 endometrioid carcinoma [9, 18, 19]. These studies suggest that Her2/neu overexpression in endometrioid carcinoma characterizes late progression and differentiation events. Her2/neu overexpression is detected in approximately 9%–30% of serous carcinomas [20]. Elucidation of the role of Her2/neu in these pathogenic tumor types, therefore, requires further study. 2.3. β-Catenin β -catenin, a component of the E-cadherin family of proteins, is essential for cell differentiation and maintenance of normal tissue architecture, and plays an important role in signal transduction. β-catenin also acts as a downstream transcriptional activator in the Wnt signal transduction pathway. A β-catenin mutation results in the stabilization of proteins that are degradation resistant, thus resulting in cytoplasmic and nuclear β-catenin accumulation and constitutive target gene activity. The accumulation of β-catenin is demonstrated by immunohistochemistry. Several studies have analyzed endometrial cancers, showing that nuclear accumulation of β-catenin is significantly more common in endometrioid lesions (31% to 47%) compared to nonendometrioid histologies (0% to 3%) [21]. In another report, β-catenin nuclear accumulation was more frequent in endometrial hyperplasias than in endometrial carcinoma samples, suggesting a β-catenin role in the early development of this tumor type [22]. In fact, alterations in β-catenin have been described in endometrial hyperplasia that contains squamous metaplasia or morules. Koul et al. found that all β-catenin mutated tumors were estrogen-receptor (ESR) positive and most were progesterone-receptor (PgR) positive, thus suggesting a dependence on estrogen stimulation during endometrial carcinogenesis [11]. In contrast, there is no correlation between β-catenin mutations and Microsatellite Instability (MI) or K-ras or PTEN mutations.

d tumors were estrogen-receptor (ESR) positive and most were progesterone-receptor (PgR) positive, thus suggesting a dependence on estrogen stimulation during endometrial carcinogenesis [11]. In contrast, there is no correlation between β-catenin mutations and Microsatellite Instability (MI) or K-ras or PTEN mutations. 2.4. AKT The phosphatidylinositol 3-kinase (PI3K) AKT pathway is activated in many human cancers and plays a key role in cell proliferation and survival. PIK3CA mutations frequently occur with other genetic alterations such as Her2/neu, K-ras, and PTEN in several types of tumors. Endometrial cancer is known to possess various genes alterations which activate the PI3K-AKT pathway. The frequency of mutations for PIK3CA in endometrial cancer is reported to be 28% [23]. However, Shoji et al. reported that AKT1 (E17K) mutations were detected in 2 out of 89 tissue samples and 0 out of 12 cell lines [24]. They suggested that AKT1 mutations might be mutually exclusive from other PI3K-AKT activating alterations, although PIK3CA mutations frequently coexist with other gene aberrations. Additional mutations in AKT family members in endometrial cancers were reported in AKT2 (D399N, 426T, and 141T) and in AKT3 (E438D) [25]. Taken together, studies found that 5 out of 41 endometrial cancers have mutations in AKT family members at a frequency of approximately 12%.

ntly coexist with other gene aberrations. Additional mutations in AKT family members in endometrial cancers were reported in AKT2 (D399N, 426T, and 141T) and in AKT3 (E438D) [25]. Taken together, studies found that 5 out of 41 endometrial cancers have mutations in AKT family members at a frequency of approximately 12%. 2.5. FGFR2 Alterations in the fibroblast growth factor receptor 2 (FGFR2) gene causes the receptors to become active, leading to cell proliferation. Byron et al. reported mutations in FGFR2 in 10% of primary uterine tumor samples [26]. Mutations were observed in 16% of the endometrioid histology subtype tumors. In primary endometrioid endometrial cancers, FGFR2 and K-ras mutations were mutually exclusive. Conversely, FGFR2 mutations were seen together with PTEN loss-of-function mutations. The authors also showed that endometrial cancer cell lines with activating FGFR2 mutations are selectively sensitive to the pan-FGFR inhibitor, PD173074 [27]. In addition, upregulation of FGF2 mRNA expression was observed in endometrial cancer specimens [28]. These data suggest that investigation of these agents may be therapeutically beneficial for endometrial cancer patients.

ith activating FGFR2 mutations are selectively sensitive to the pan-FGFR inhibitor, PD173074 [27]. In addition, upregulation of FGF2 mRNA expression was observed in endometrial cancer specimens [28]. These data suggest that investigation of these agents may be therapeutically beneficial for endometrial cancer patients. 3. Loss-of-Function Genetic Events 3.1. PTEN Endometrial carcinomas are characterized by a variety of genetic alterations, but the most frequent alteration is in the PTEN gene. PTEN, located at chromosome 10q23, encodes a protein and lipid phosphatase which behaves as a tumor suppressor gene. PTEN inactivation is induced by mutations that lead to a loss of expression and is induced to a lesser extent by a loss of heterozygosity. The PTEN protein has both lipid and protein phosphatase activities, with each serving different functions. The lipid phosphatase activity of PTEN induces cell cycle arrest at the G1/S checkpoint. In addition, the upregulation of proapoptotic mechanisms involving AKT-dependent mechanisms is mediated through PTEN, as is the downregulation of anti-apoptotic mechanisms through Bcl-2 [29–31]. PTEN further acts in opposition to PI3K to control levels of phosphorylated AKT [23, 32]. A PI3K mutation is seen in 36% of endometrioid endometrial cancers and is common in tumors that also carry the PTEN mutation. The protein phosphatase activity of PTEN is involved in the inhibition of focal adhesion formation, cell spread, and migration, as well as the inhibition of growth-factor-stimulated MAPK signaling [33]. The PTEN gene, which acts as a tumor suppressor gene, is present in individuals and causes increased cancer susceptibility, including those with Cowden's syndrome. PTEN mutations are the most frequent genetic lesions in endometrial adenocarcinomas of the endometrioid subtype. PTEN mutations are reported in 25%–83% of tumors, more frequently in endometrioid carcinomas and microsatellite unstable tumors, and are, thus, the most frequent genetic alteration reported in cancers [34]. PTEN gene alterations are associated with metastatic behavior and advanced stage in other cancer types. In contrast, the loss of PTEN function is an early event in endometrial tumorigenesis. Several groups have described a concordance between MI status and PTEN mutations; the mutations occur in 60%–86% of MI-positive endometrial carcinoma EEC cases, but only occur in 24%–35% of MI-negative tumors.

in other cancer types. In contrast, the loss of PTEN function is an early event in endometrial tumorigenesis. Several groups have described a concordance between MI status and PTEN mutations; the mutations occur in 60%–86% of MI-positive endometrial carcinoma EEC cases, but only occur in 24%–35% of MI-negative tumors. Genetic alterations that account for PTEN protein inactivation include various mutations, a loss of heterozygosity (LOH), or promoter hypermethylation, with mutations occurring the most frequently [30]. PTEN promoter methylation is observed in 19% of cancers and is significantly associated with metastatic disease [35]. Kim et al. reported that PTEN and K-ras double-mutant mice (Ptend/dK-rasG12D) exhibited dramatically accelerated endometrial cancer development compared to cancers formed from a single PTEN or K-ras gene mutation [36]. These results suggest a synergistic effect of dysregulation of the PTEN and K-ras signaling pathways during endometrial tumorigenesis.

K-ras double-mutant mice (Ptend/dK-rasG12D) exhibited dramatically accelerated endometrial cancer development compared to cancers formed from a single PTEN or K-ras gene mutation [36]. These results suggest a synergistic effect of dysregulation of the PTEN and K-ras signaling pathways during endometrial tumorigenesis. 3.2. P53 The p53 gene is located on chromosome 17 and is important in preventing the propagation of cells with damaged DNA. p53 mutations or TP53 overexpression is twice as frequent in tumors without hyperplasia (estrogen unrelated) than in those with hyperplasia (estrogen related) [11, 37]. This is consistent with other data in which the most striking genetic alteration, present in about 90% of serous carcinomas (estrogen-unrelated NEEC), is a p53 mutation [38]. In other reports, statistically significant correlations were observed between p53 alterations and non-endometrioid histology type, high-grade tumors, and the absence of the progesterone receptor [39]. On the other hand, p53 genetic alterations were observed in 17% of endometrioid carcinomas, which were primarily Grade 3 [40]. The exact mechanisms causing this mutation are still not well characterized. In response to DNA damage, nuclear P53 accumulates and causes cell cycle arrest by inhibiting Cyclin D1 phosphorylation of the Rb gene and thereby promotes apoptosis. Therefore, mutated P53 results in a nonfunctional protein that accumulates in the cell and acts as a dominant negative inhibitor of wild-type P53, leading to propagation of aberrant cells. p53 mutations in endometrioid carcinoma are a late event during progression or differentiation. P53 alterations play a relatively minor role in clear cell type endometrial carcinoma in comparison to the serous type [41]. p53 mutations are also rarely observed in ovarian clear cell adenocarcinomas in comparison to endometrioid adenocarcinomas [42]. As a result, it is possible that the pathogenesis of clear cell carcinoma in the female genital tract arises from a unique pathway [43].

trial carcinoma in comparison to the serous type [41]. p53 mutations are also rarely observed in ovarian clear cell adenocarcinomas in comparison to endometrioid adenocarcinomas [42]. As a result, it is possible that the pathogenesis of clear cell carcinoma in the female genital tract arises from a unique pathway [43]. 4. Genomic Instability

trial carcinoma in comparison to the serous type [41]. p53 mutations are also rarely observed in ovarian clear cell adenocarcinomas in comparison to endometrioid adenocarcinomas [42]. As a result, it is possible that the pathogenesis of clear cell carcinoma in the female genital tract arises from a unique pathway [43]. 4. Genomic Instability The most important types of genomic instability in endometrial cancers are MI and chromosomal aneuploidy. DNA mismatch repair (MMR) deficiency, detected as MI, is the most common molecular phenotype in endometrioid cancer, as PTEN tumor suppressor gene mutations. MI is seen in cancers (colonic, endometrial, and others) of patients with hereditary nonpolyposis colon cancer (HNPCC) and is also present in 28% of sporadic endometrioid cancers but is not present in serous cancers [40]. MI is distributed almost equally among the three histopathological tumor grades of endometrioid cancers. However, MI is rare in the clear cell type [44]. HNPCC patients with endometrial cancers have an inherited germline mutation in MLH-1, MSH-2, MSH-6, or PMS-2, but endometrial cancer only develops after the instauration of a deletion or mutation in the contralateral MLH-1, MSH-2, MSH-6, or PMS-2 allele. Following this, the deficient MMR (MLH-1, MSH-2, MSH-6, or PMS-2) causes the acquisition of MI and the development of the tumor. Inactivation of the mismatch repair gene MLH1 by methylation of the promoter seems to be the most frequent cause of MI in sporadic endometrioid carcinomas, followed by a loss of the expression of other two mismatch repair genes, the MSH2 and MSH6 genes. The mechanism for the inactivation of MSH2 is still not clear, as promoter methylation and mutations are rare. MSH6 inactivation is usually caused by a mutation.

st frequent cause of MI in sporadic endometrioid carcinomas, followed by a loss of the expression of other two mismatch repair genes, the MSH2 and MSH6 genes. The mechanism for the inactivation of MSH2 is still not clear, as promoter methylation and mutations are rare. MSH6 inactivation is usually caused by a mutation. Aneuploidy is frequent in serous cancers, and is uncommon in endometrioid cancer. When present, aneuploidy is exhibited predominantly by Grade 3 tumors [45, 46]. These data suggest that a different type of genomic instability is associated with the different histopathological-type tumors. However, in some reports, no significant correlations were found to exist with either the K-ras or p53 mutations [7, 11, 47]. Telomeric attrition triggers genomic instability in certain cancer types. Both EEC and NEEC cells have short telomeres in endometrial cancer. However, only NEECs are significantly associated with critical telomere shortening compared to adjacent morphologically normal epithelium, thus suggesting that telomere shortening contributes to the initiation of NEECs but not EECs [48]. The authors also proposed a model in which telomere attrition gives rise to the initiation of NEECs and the progression of EECs.

ed with critical telomere shortening compared to adjacent morphologically normal epithelium, thus suggesting that telomere shortening contributes to the initiation of NEECs but not EECs [48]. The authors also proposed a model in which telomere attrition gives rise to the initiation of NEECs and the progression of EECs. 5. Genetics Events outside the Cancer Pathway Genetic variation acting either within or outside of the cancer cell may determine the outcome of interaction with exogenous or endogenous carcinogens. Endometrial stimulation by estrogens without the differentiating effects of the progestins is a primary etiologic factor associated with the development of endometrial hyperplasia and carcinoma [3]. There is evidence that estrogens and some of their metabolites are involved in the endometrial cancer pathogenesis. Estrogens and some of their derivatives are genotoxic and induce DNA damage, which if not removed could, thus, contribute to an increased risk of malignancy. Defects in the estrogen metabolism can result in defective apoptosis, DNA repair, and proliferation [49, 50]. Estrogens mediate their effects via the estrogen receptors (ESRs), estrogen receptor alpha (ESR1) and estrogen receptor beta (ESR2), which activate its metabolic pathways. The polymorphisms of ESR1 and ESR2 suggest an association with an increasing risk of developing endometrial cancer [51]. Cytochrome P450 1B1 CYP1B1 is a constitutively expressed and inducible enzyme with a central role in the oxidative metabolism of a wide range of endogenous and exogenous compounds including many carcinogens [52, 53]. Saini et al. reported that CYP1B1 depletion in endometrial cancer cells leads to decreased cellular proliferation and induced G0-G1 cell cycle arrest, thus suggesting that CYP1B1 inhibition in endometrial cancer cells could be a useful therapeutic approach [54]. Progesterone or its synthetic form has been used as a primary treatment or palliative treatment of advanced and recurrence endometrial cancer, because progesterone inhibits estrogen-induced endometrial proliferation. In addition, the loss of progesterone-mediated Wnt signaling inhibition in the endometrium plays a rate-limiting role in tumor onset and progression [55].

a primary treatment or palliative treatment of advanced and recurrence endometrial cancer, because progesterone inhibits estrogen-induced endometrial proliferation. In addition, the loss of progesterone-mediated Wnt signaling inhibition in the endometrium plays a rate-limiting role in tumor onset and progression [55]. 6. Inherited Predisposition 6.1. Lynch Syndrome Lynch syndrome, or hereditary non-polyposis colorectal cancer (HNPCC), is characterized by an increased risk for colorectal cancer. Endometrial cancer is the most common malignancy in patients with Lynch syndrome or HNPCC [56]. Lynch syndrome is caused by an inherited mutation in the MMR gene family, such as MLH1, MSH2, MSH6, PMS1, or PMS2 [57]. The age at diagnosis of Lynch syndrome associated endometrial cancer is approximately 2 decades younger than that for sporadic endometrial cancers [58]. Parc et al. demonstrated that 34% of young patients with endometrial cancer (median age 46) were associated with MI, 57% of the MI positive group showed an absence of hMLH1 expression, 19% showed an absence of hMLH2 expression, and 23.8% demonstrated a normal expression of both proteins, while 9.5% of all patients were diagnosed with Lynch syndrome [59]. In another report, the development of the latter tumors of Lynch syndrome is significantly associated with MSH2/MSH6 protein complex deficiency [60].

9% showed an absence of hMLH2 expression, and 23.8% demonstrated a normal expression of both proteins, while 9.5% of all patients were diagnosed with Lynch syndrome [59]. In another report, the development of the latter tumors of Lynch syndrome is significantly associated with MSH2/MSH6 protein complex deficiency [60]. 6.2. Familial Site-Specific Endometrial Carcinoma The clustering of endometrial carcinoma alone, termed as familial site-specific endometrial carcinoma, may constitute a separate entity. Eight percent of this group have been reported to have germline MMR mutations [61]. This mutation rate is lower than that of Lynch syndrome with endometrial cancer patients, of whom 15% show MMR mutations [62]. The difference in MMR, mutations, therefore suggests the existence of different genetic alteration pathways in familial site-specific endometrial carcinoma.

to have germline MMR mutations [61]. This mutation rate is lower than that of Lynch syndrome with endometrial cancer patients, of whom 15% show MMR mutations [62]. The difference in MMR, mutations, therefore suggests the existence of different genetic alteration pathways in familial site-specific endometrial carcinoma. 7. Malignant Mixed Mullerian Tumors (MMMTs) Carcinosarcomas (malignant mixed mullerian tumors, or MMMTs) are currently excluded from uterine sarcoma and classified as metaplastic carcinoma, and many studies include these as NEECs [63]. However, endometrial carcinoma and MMMTs develop along distinctive molecular genetic pathways and exhibit different biological features. In MMMT, p53 alterations occur early, during progression, just prior to clonal expansion and acquisition of genetic diversity [64]. In addition, changes in the AKT/β-catenin pathway may be essential for both the establishment and maintenance of phenotypic characteristics of MMMTs, playing key roles in the regulation of E-cadherin through transactivation of the Slug E-cadherin repressor gene [65]. Vaidya et al. reported that according to the discrepancy in survival the patients of MMMT should not be included in studies of endometrial cancers [66]. From this viewpoint, future studies will identify factors to classify these diseases.

of E-cadherin through transactivation of the Slug E-cadherin repressor gene [65]. Vaidya et al. reported that according to the discrepancy in survival the patients of MMMT should not be included in studies of endometrial cancers [66]. From this viewpoint, future studies will identify factors to classify these diseases. 8. De-Differentiation of Endometrioid Tumors Mixed serous and endometrioid tumors have serous components that may be related to the “de-differentiation” of endometrioid tumors. This concept would explain the presence of overlapping EEC and NEEC features, both morphological and molecular in some tumors [67].

of E-cadherin through transactivation of the Slug E-cadherin repressor gene [65]. Vaidya et al. reported that according to the discrepancy in survival the patients of MMMT should not be included in studies of endometrial cancers [66]. From this viewpoint, future studies will identify factors to classify these diseases. 8. De-Differentiation of Endometrioid Tumors Mixed serous and endometrioid tumors have serous components that may be related to the “de-differentiation” of endometrioid tumors. This concept would explain the presence of overlapping EEC and NEEC features, both morphological and molecular in some tumors [67]. 9. Epigenetic Changes Aberrant CpG island hypermethylation in promoter regions occurs in many cancer-related genes, including those associated with cell cycle control, apoptosis, and DNA repair. Usually, unmethylated CpG islands become methylated, causing transcriptional silencing in cancer cells. Banno et al. reported that the frequencies of aberrant hypermethylation were 40.4% in hMLH1, 22% in APC, 14% in E-cadherin, and 2.3% in RAR-β in endometrial cancer specimens [68]. However, no aberrant DNA methylation was found in the p16 gene. Other genes inactivated by promoter hypermethylation in endometrial cancer include PgR [69], the cell cycle control genes 14-3-3 sigma [70], homeobox gene HOXA11, thrombospondin-2 gene (THBS2) [71], paternally expressed gene 3 (PEG3) [72], as well as the detoxifying enzyme glutathione S-transferase P1 (GSTP1) [73]. The impact of methylation on these genes in endometrial cancer development has not been well established. In endometrial cancers, differential DNA methylation patterns are detected in EICs and NEECs, suggesting divergent epigenetic backgrounds and unique tumorigenic pathways [74]. Promoter hypermethylation is a frequent event in EIC but not NEECs [75]. Many of the tumor suppressor pathways that are mutated in EIC can also be inactivated by hypermethylation.

l DNA methylation patterns are detected in EICs and NEECs, suggesting divergent epigenetic backgrounds and unique tumorigenic pathways [74]. Promoter hypermethylation is a frequent event in EIC but not NEECs [75]. Many of the tumor suppressor pathways that are mutated in EIC can also be inactivated by hypermethylation. 10. The Future The goal of screening endometrial cancers is to identify all patients who have a risk for developing this disease. Therefore clarification of the molecular and genetic mechanisms of development or progression of this disease is required. Understanding the genetic changes underlying cancer development or progression in the different histological subtypes is important for discovery of new targets for both diagnosis and therapy for individual patients. Acknowledgments This work was supported in part by Special Subsidies from the Subsidies for ordinary expenses of private schools from the Promotion and Mutual Aid Corporation for Private Schools of Japan, as well as research funding from the JAOG Ogyaa Donation Foundation (JODF) and the Takeda Science Foundation. Table 1 Clinical and pathological features of endometrial carcinoma.

Acknowledgments This work was supported in part by Special Subsidies from the Subsidies for ordinary expenses of private schools from the Promotion and Mutual Aid Corporation for Private Schools of Japan, as well as research funding from the JAOG Ogyaa Donation Foundation (JODF) and the Takeda Science Foundation. Table 1 Clinical and pathological features of endometrial carcinoma. Type I (EEC) Type II (NEEC) Age Pre- and perimenopausal Postmenopausal Behavior Stable Progressive Grade Low High Hyperplasia-precursor Present Absent Unopposed estrogen Present Absent Myometrial invasion Minimal Deep Specific Subtypes Endometrioid carcinoma Non-endometrioid carcinoma Prevalence 70–80% 10–20% Risk factors Obesity, anovulation, nulliparity and exogenous estrogen exposure In atropic endometrium Table 2 Genetics features of endometrial carcinoma. EEC NEEC Gain-of Function K-ras 15–30% 0–5% Her2/neu 10–20% 9–30% β-Catenin 31–47% 0–3% Loss-of Function PTEN 35–50% 10% P53 10–20% 90% Genomic instability (microsatellite) 20–40% 0-5%

1. Introduction Hysteroscopy is the gold standard procedure for uterine cavity exploration [1]. However, the World Health Organization (WHO) recommends hysterosalpingography (HSG) alone for management of infertile women [2]. The explanation for this discrepancy is that HSG provides information on tubal patency or blockage. Office hysteroscopy is only recommended by the WHO when clinical or complementary exams (ultrasound, HSG) suggest intrauterine abnormality [3] or after in vitro fertilization (IVF) failure [4]. Nevertheless, many specialists feel that hysteroscopy is a more accurate tool because of the high false-positive and falsenegative rates of intra uterine abnormality with HSG [1, 5, 6]. This explains why many specialists use hysteroscopy as a first-line routine exam for infertility patients regardless of guidelines. The aim of this retrospective study is to describe hysteroscopy findings in a population of 557 infertile patients. 2. Materials and Methods We analyzed retrospectively 557 patients referred for hysteroscopy for incapacity to conceive lasting at least 1 year or prior to IVF, from November 2002 to July 2006. This population represents one third of hysteroscopies on that period. All hysteroscopies were performed by the same operator (JLM).

terials and Methods We analyzed retrospectively 557 patients referred for hysteroscopy for incapacity to conceive lasting at least 1 year or prior to IVF, from November 2002 to July 2006. This population represents one third of hysteroscopies on that period. All hysteroscopies were performed by the same operator (JLM). Procedures lasted approximately two minutes without anesthesia or cervical preparation in an office gynecology setting. Diagnostic video-assisted hysteroscopy was performed using a flexible hysteroscope (flexible hysteroscope, Olympus HYF-P, Paris, France) with an outer diameter of 3.1 mm. Procedures were not video recorded. The uterine cavity was expanded under manual hydrostatic pressure (saline solution). Hysteroscopy was performed with a standard sequence, inspecting the endocervical canal, uterine cavity, endometrium, and tubal ostia. Findings were recorded using a standard report. Statistical analyses were performed using Stata 9.2 for Windows (StataCorp LP, TX, USA) and Statistica 6.0 (StatSoft, OK, USA). 3. Results On 557 successive patients, hysteroscopy could not be performed in one case because of pain. We observed no perforation, hemorrhagic, or metabolic complications. The mean age at hysteroscopy was 35.3 years (21–44 years). Indication for hysteroscopy was pre IVF in 78.8% (439/557) and infertility in 21.2% (118/557). Women investigated with hysteroscopy were nulliparous in 73.4% (409/557), primiparous in 21.4% (119/557), and multiparous in 5.2% (29/557).

3. Results On 557 successive patients, hysteroscopy could not be performed in one case because of pain. We observed no perforation, hemorrhagic, or metabolic complications. The mean age at hysteroscopy was 35.3 years (21–44 years). Indication for hysteroscopy was pre IVF in 78.8% (439/557) and infertility in 21.2% (118/557). Women investigated with hysteroscopy were nulliparous in 73.4% (409/557), primiparous in 21.4% (119/557), and multiparous in 5.2% (29/557). Hysteroscopy was normal in 60.5% (337/557) and among women with abnormal results, 20% showed more than one abnormality (44/220). 3.1. Cervico-Isthmic Abnormalities (Figure 1) Cervico-isthmic abnormalities were present in 4.3% (24/557) of patients with 13 cases of polyps (2.3%), 9 stenosis (1.8%), 2 adhesions (0.4%). The stenosis was complete and did not allow to complete the procedure in 4 cases. 3.2. Uterine Cavity Abnormalities (Figure 2) Uterine cavity was seen and normal in 72% of the cases (401/557). It could not be explored in 4 cases because of complete cervical stenosis. Observed abnormalities were the following. Adenomyosis aspect: 17 cases (3.1%). Images compatible with adenomyosis were small openings in the endometrial surface, dark blue colour cystic lesions, rigid, tight tubal ostium (erecta), or T form uterus. Intrauterin adhesion (IUA), synechiae: 22 cases (3.9% of all hysteroscopies). Septate uterus: 4 uterine septa (0.7% of all hysteroscopies). Hypoplasia and uterus unicornis: 15 hypoplasia (2.8% of all hysteroscopies), uterus unicornis in 2 cases. Sub mucous myoma: 13 cases (2.3% of all hysteroscopies).

Adenomyosis aspect: 17 cases (3.1%). Images compatible with adenomyosis were small openings in the endometrial surface, dark blue colour cystic lesions, rigid, tight tubal ostium (erecta), or T form uterus. Intrauterin adhesion (IUA), synechiae: 22 cases (3.9% of all hysteroscopies). Septate uterus: 4 uterine septa (0.7% of all hysteroscopies). Hypoplasia and uterus unicornis: 15 hypoplasia (2.8% of all hysteroscopies), uterus unicornis in 2 cases. Sub mucous myoma: 13 cases (2.3% of all hysteroscopies). Deformed cavity from intramural myoma: 17 cases (3.1% of all hysteroscopies); unique (10 cases), multiple (7 cases). Endometrial polyp: 54 cases (9.7% of all hysteroscopies). Unique in 30 cases (5.4% of all hysteroscopies). Their location was corporeal (19 cases) or cornual (11 cases). Multiples polyps were observed in 4.3% of all hysteroscopies (24 cases). Trophoblastic retention: 7 cases (1.3% of all hysteroscopies) showed images compatible with a trophoblastic retention. Criteria used were a previous pregnancy with miscarriage, no ultrasound control of uterine vacuity, and a typical macroscopic aspect. Those were localized in the utero tubal junction (4 cases), or occupying the whole uterine cavity (one case). Findings according to age are given in Figure 2. 3.3. Ostial Abnormalities A tubal ostium could not be seen in 15 cases, due to cornual adhesions (2 cases), trophoblastic retention (1 case), intra uterine adhesion (6 cases), inflammation (2 cases) or minor hemorrhage (1 case), endometrial hyperplasia (1 case), or unicornuate uterus (2 cases).

Findings according to age are given in Figure 2. 3.3. Ostial Abnormalities A tubal ostium could not be seen in 15 cases, due to cornual adhesions (2 cases), trophoblastic retention (1 case), intra uterine adhesion (6 cases), inflammation (2 cases) or minor hemorrhage (1 case), endometrial hyperplasia (1 case), or unicornuate uterus (2 cases). Both tubal ostia could not be seen in 9 cases, due to cornual adhesions (2 cases), retention (2 cases), intra uterine adhesion (1 case), inflammation (2 cases) or minor hemorrhage (1 case), and endometrial hyperplasia (1 case). 3.4. Endometrial Abnormalities Findings are given in Figure 3. Endometrial inflammation was characterized by the presence of areas of red endometrium flushed with a white central point, localized or scattered throughout the cavity. Continuous analysis with logistic regression shows that the risk to have an hysteroscopic abnormal finding increases with age (P < .001, OR = 1.076, IC = 1.04–1.12). Each year the risk is multiplied by 1.08. The risk is therefore multiplied by a factor of 1.5 every 5 years (Figure 4).

Endometrial inflammation was characterized by the presence of areas of red endometrium flushed with a white central point, localized or scattered throughout the cavity. Continuous analysis with logistic regression shows that the risk to have an hysteroscopic abnormal finding increases with age (P < .001, OR = 1.076, IC = 1.04–1.12). Each year the risk is multiplied by 1.08. The risk is therefore multiplied by a factor of 1.5 every 5 years (Figure 4). 4. Discussion We found that first-line office hysteroscopy for infertility shows abnormal findings in 40% of woman. This proportion increased with age, ranging from 30% at 30 years to more than 60% after 42 years. These findings are based on a large cohort of infertile women, with homogeneous age distribution. Hysteroscopies were performed consecutively by a single-trained operator. All investigations were performed using a flexible minihysteroscope which provides high patient acceptance since it makes hysteroscopy a painless and well-tolerated procedure. However, symptoms, clinical examination, ultrasound findings, HSG or hormonal blood sampling results characteristics were not available in our population. Moreover, there was no control group of fertile women to compare our findings with. Patients were referred from many hospitals and private clinics, with no homogeneity in infertility investigations prior to hysteroscopy. Finally, the absence of video recording did not allow control of findings by a different operator. No possibility of re-evaluation of the findings represents an important weakness of this study. However, experience of the single operator who performed all hysteroscopies and the use of a standard report to record abnormal findings limit the impact of such a bias.

not allow control of findings by a different operator. No possibility of re-evaluation of the findings represents an important weakness of this study. However, experience of the single operator who performed all hysteroscopies and the use of a standard report to record abnormal findings limit the impact of such a bias. The previously published data show large ranges of abnormal finding rates from one study to another (7.2% to 64%) [7–16]. These differences could be explained by the type of hysteroscopic distension medium and/or hysteroscopic technique used, modifying the surgeon's perception of intrauterine filling defects [17]. Results could also be influenced by the characteristics of the population: age of the population, hormonal status, ethnic factor, type of infertility (primary or secondary) and indications for hysteroscopy (infertility alone, association with clinical, echographic or hysterosalpingographic abnormalities, prior to IVF ect.).

ts could also be influenced by the characteristics of the population: age of the population, hormonal status, ethnic factor, type of infertility (primary or secondary) and indications for hysteroscopy (infertility alone, association with clinical, echographic or hysterosalpingographic abnormalities, prior to IVF ect.). Dicker et al. founded higher rates of abnormal findings in elderly women (above 40 years old). Abnormalities such as submucous myomas, endometrial hyperplasia, and polyps were more frequent in this population, while in younger patients other uterine lesions such as adhesions and tubal ostia occlusion were more common [7]. When comparing hysteroscopic abnormalities before and after 38 years of age, Magos et al. [13] did not show a significant difference (51% of abnormal finding before 38 years and 43% after, P = .38).This result might be explained by the high rate of endometritis in their population (17.2%), which was more frequently observed in younger woman.

copic abnormalities before and after 38 years of age, Magos et al. [13] did not show a significant difference (51% of abnormal finding before 38 years and 43% after, P = .38).This result might be explained by the high rate of endometritis in their population (17.2%), which was more frequently observed in younger woman. Many studies describe the incidence of abnormal findings with hysteroscopy in infertile women or prior to IVF, but none give the proportion of these women who could benefit from an adapted treatment based on hysteroscopic findings. It is difficult to draw direct connections between hysteroscopic findings and benefits from a specific treatment based on these findings. Treatments for some abnormalities are suspected beneficial in infertile women. These are intrauterine adhesions, congenital uterine malformations, endometrial polyps, and uterine myomas [18]. Chronic endometrial inflammation and micropolyps have also been related to infertility and recurrent miscarriages [19]. It is not clear yet if abnormal hysteroscopic findings, by guiding infertility treatments, increase pregnancy rates. In our population we founded abnormal hysteroscopic findings in 40% of the infertile women, and 75% of these abnormalities could be related to infertility and benefit from a specific treatment. La Sala et al. suggest hysteroscopy as a routine exam in infertile woman because it would be economically advantageous, in regard to costs of assisted reproductive technology [14].

ndings in 40% of the infertile women, and 75% of these abnormalities could be related to infertility and benefit from a specific treatment. La Sala et al. suggest hysteroscopy as a routine exam in infertile woman because it would be economically advantageous, in regard to costs of assisted reproductive technology [14]. 5. Conclusions Rates of abnormal findings in unselected infertile patient who underwent diagnostic hysteroscopy ranged from 30% at 30 years to more than 60% after 42 years. Risk of abnormal finding was multiplied by a factor of 1.5 every 5 years. Our data are an additional argument to propose office hysteroscopy as part of first line exams in infertile woman, regardless of age. Figure 1 Rates of cervical abnormalities in 557 infertile women during office hysteroscopy according to age. Figure 2 Rates of intrauterine abnormalities in 557 infertile women during office hysteroscopy according to age. Figure 3 Rates of endometrial abnormalities in 557 infertile women during office hysteroscopy according to age. Figure 4 Rates of abnormal findings in 557 infertile women during office hysteroscopy according to age.

1. Introduction Cervical cancer affects a large number of women with an estimated 450,000 new cases per year globally; approximately 10,000 are diagnosed annually in the U.S. [1]. The vast majority of women will be diagnosed at early clinical stage when cure rates are high. Early-stage (IA2 to IIA) carcinoma of the cervix can be treated by radical hysterectomy (RH) or primary radiotherapy (RT) with similar outcomes (five-year overall survival 87%–92%) [2]. The choice of therapeutic modality is based on patient comorbidities, and patient or physician preference. Benefits to surgical treatment include simultaneous lymphadenectomy for surgical staging [3] with possible therapeutic benefit [4], preservation of ovarian function in premenopausal women, and improved coital function, as compared to RT [5–7].

lity is based on patient comorbidities, and patient or physician preference. Benefits to surgical treatment include simultaneous lymphadenectomy for surgical staging [3] with possible therapeutic benefit [4], preservation of ovarian function in premenopausal women, and improved coital function, as compared to RT [5–7]. However, a subset of women treated with primary surgery will require adjuvant postoperative chemoradiation due to findings which confer a high risk of recurrence. These include cancer spread to lymph nodes, invasion into the parametria, and positive surgical margins [8]. In order to avoid the combined morbidity of both methods of treatment, many surgeons will abandon the radical hysterectomy intraoperatively (termed abandoned or aborted radical hysterectomy) if there are findings of disseminated disease, such as positive lymph nodes. This occurs in approximately 8%–10% of radical hysterectomies for early-stage cancer [9, 10], and primary treatment with chemoradiation follows. This is an area of controversy in gynecologic oncology, and there is no consensus regarding the most appropriate management. Further, data on preoperative identification of women who may be at risk of an abandoned hysterectomy is lacking.

or early-stage cancer [9, 10], and primary treatment with chemoradiation follows. This is an area of controversy in gynecologic oncology, and there is no consensus regarding the most appropriate management. Further, data on preoperative identification of women who may be at risk of an abandoned hysterectomy is lacking. The finding of extra-cervical spread at the time of exploration for radical surgery presents a challenging decision: proceed with planned hysterectomy knowing that the patient will require further postoperative therapy or abandon the procedure for primary chemoradiation. Therefore, we embarked on a review of cervical cancer patients at our institution that had an abandoned radical hysterectomy for preoperative clinical characteristics, morbidities and survival.

ctomy knowing that the patient will require further postoperative therapy or abandon the procedure for primary chemoradiation. Therefore, we embarked on a review of cervical cancer patients at our institution that had an abandoned radical hysterectomy for preoperative clinical characteristics, morbidities and survival. 2. Materials and Methods IRB approval was obtained to review charts of patients who underwent surgical exploration for early-stage cervical cancer (Stage IB1-IIA) for the intent of radical hysterectomy over a 10-year period at the University of Washington Medical Center. We identified patients whose planned hysterectomy was abandoned due to intraoperative findings of metastatic spread and abstracted data from medical records on demographics, laboratory values, preoperative radiologic imaging, operative reports, pathology specimens, complications, recurrence, and overall survival. In addition, we identified a group of early-stage patients who underwent a radical hysterectomy and were found to have the high risk feature of positive lymph nodes postoperatively to compare morbidities and outcomes of therapy (completed group). Statistical analysis was performed by Fisher's exact method between two variables and Log rank test to compare survival probabilities.

o underwent a radical hysterectomy and were found to have the high risk feature of positive lymph nodes postoperatively to compare morbidities and outcomes of therapy (completed group). Statistical analysis was performed by Fisher's exact method between two variables and Log rank test to compare survival probabilities. 3. Results Between 1993 and 2003, 268 women with early-stage (IA2 to IIA) cervical cancer presented for primary surgical management with radical hysterectomy at our institution. On review of operative reports, nineteen patients (7%) had intraoperative abandonment of their planned radical hysterectomy. The median age was 42 years old (range 29–85) and approximately half were smokers (n = 9, 47%) with median pack year history of 18. Most were stage IB1 (63%) and squamous histology (79%) (Table 1). Looking at other preoperative characteristics, the most common presenting symptom was vaginal bleeding (n = 12, 68%), whereas only three (16%) of patients were diagnosed following referral for an abnormal Pap smear. The median number of years since prior Pap smear was four (range 1–60). Median preoperative hematocrit and hemoglobin were 34.5 and 12.0, respectively (range 29–41, 9.6–14.1). Preoperative imaging by CT scan was obtained in 12/19 (63%). Of these, 50% (n = 6) had findings suspicious for pelvic lymphadenopathy (Table 1). No patients had preoperative PET imaging during this study period.

60). Median preoperative hematocrit and hemoglobin were 34.5 and 12.0, respectively (range 29–41, 9.6–14.1). Preoperative imaging by CT scan was obtained in 12/19 (63%). Of these, 50% (n = 6) had findings suspicious for pelvic lymphadenopathy (Table 1). No patients had preoperative PET imaging during this study period. The reasons for abandonment were positive pelvic lymph nodes (84%), positive paraaortic lymph nodes (16%) and/or peritoneal spread (16%) by intraoperative frozen section. Of the 16 patients who had abandonment for positive pelvic lymph nodes, 25% were found to have positive para-aortic or common iliac lymph nodes on final pathology. In total, 7 of 19 (37%) had positive para-aortic or common iliac lymph nodes. The mean number of lymph nodes removed was 23 (17 pelvic and 6 para-aortic, range 0–82) and the mean number of positive lymph nodes was 2.7 (range 0–9). Most women received definitive postoperative chemoradiation therapy (12/19, 63%) or radiation alone (4/19, 32%). One patient with adenocarcinoma spread to scalene nodes underwent palliative chemotherapy and died of disease within 12 months. Five patients were treated with concurrent 5-FU and cisplatin, one with gemcitbine and cisplatin, and six with cisplatin alone. The median radiation dose was 45 Gy (range 30.6–50.4) external beam whole pelvic and 85 Gy (range 71–93) to point A with either LDR or HDR. Following radiation therapy, only three patients (16%) underwent an adjuvant hysterectomy, and one (33%) had residual disease.

mcitbine and cisplatin, and six with cisplatin alone. The median radiation dose was 45 Gy (range 30.6–50.4) external beam whole pelvic and 85 Gy (range 71–93) to point A with either LDR or HDR. Following radiation therapy, only three patients (16%) underwent an adjuvant hysterectomy, and one (33%) had residual disease. We compared clinical characteristics, morbidity, and mortality to a group of early-stage patients (n = 44) during the same time period that had a completed radical hysterectomy, but were found postoperatively to have positive pelvic lymph nodes (completed group). All received external beam radiation or chemoradiation following radical surgery, without brachytherapy. They were similar in demographic makeup such as age, stage, and histology (Table 1). However, they were more than twice as likely to present with an abnormal Pap smear (32%), and time from prior Pap smear was median 2.5 years. Other preoperative characteristics such as weight and hemoglobin were not appreciably different between the two groups (Table 1). Next, we evaluated major complications, both operative-and radiation-related. There were five major complications in the abandoned group (26%): two operative-related (intraoperative stroke and bowel perforation postoperatively) and three radiation-related (severe lymphedema, severe radiation proctitis, requiring colostomy, and rectovaginal fistula). In the completed group, there were fifteen major complications (34%), five operative and 10 radiation-related (Table 2). The difference in morbidity was not significant.

rforation postoperatively) and three radiation-related (severe lymphedema, severe radiation proctitis, requiring colostomy, and rectovaginal fistula). In the completed group, there were fifteen major complications (34%), five operative and 10 radiation-related (Table 2). The difference in morbidity was not significant. In the abandoned group, there were seven patients (37%) who experienced recurrence. Three (43%) had positive para-aortic or common iliac lymph nodes, in addition to positive pelvic lymph nodes. Recurrence site was pelvic in four patients and distant in three. Of those with recurrence, only two (29%) were able to be salvaged with exenteration and five (71%) have died. Overall survival rate is 73% after two years follow-up. In the completed group, eight patients had recurrence (18%). Recurrence site was pelvic in four and distant in four. Of this group, none were are able to salvaged and one died of other causes, for an overall survival of 80%. By Kaplan-Maier estimates, the progression-free and overall survival difference between the two groups is not significant at the twoyear interval (Figures 1 and 2).

ce site was pelvic in four and distant in four. Of this group, none were are able to salvaged and one died of other causes, for an overall survival of 80%. By Kaplan-Maier estimates, the progression-free and overall survival difference between the two groups is not significant at the twoyear interval (Figures 1 and 2). 4. Comment In our series of 268 women with early-stage cervical cancer, 19 (7%) had an abandoned hysterectomy when unexpected spread of disease was found outside the cervix at time of surgical exploration. We found patients with abandoned hysterectomy presented more often with vaginal bleeding (68%), as opposed to an abnormal Pap (16%), which suggests that may the disease be further along in the pathogenesis. Additionally, published reports have documented length of time from prior Pap smear as an association with more advanced disease and poorer outcome [11]. In our group of patients who had an abandoned hysterectomy, the average time from prior Pap smear was four years. Therefore, increasing time from prior Pap smear may also be an indicator of finding metastatic spread at time of surgical exploration. The association of anemia and poor prognosis in patients with cervical cancer has been well established in the literature [12–14]. We therefore hypothesized that women at risk of an abandoned hysterectomy may be anemic on presentation as a marker of more advanced disease. However, median preoperative hemoglobin and hematocrit were not markedly low in our group.

s in patients with cervical cancer has been well established in the literature [12–14]. We therefore hypothesized that women at risk of an abandoned hysterectomy may be anemic on presentation as a marker of more advanced disease. However, median preoperative hemoglobin and hematocrit were not markedly low in our group. The use of preoperative imaging with CT or PET-CT is increasingly becoming the standard of care in the US to help identify metastatic disease [15, 16]. Although our study cohort spanned a time prior to more widespread preoperative imaging (only 63% of patients received), we identified suspicious lymphadenopathy in 50% of the patients who underwent a CT scan. Suspicious lymph nodes on CT alone are typically reported by size criteria as greater than 10 mm. Therefore, lymph nodes that have normal size may still harbor “microscopic” spread and will be missed on CT. PET-CT may be able to overcome some of these limitations and currently is approved for preoperative workup of metastatic disease in the US. Overall PET-CT has been shown to have a sensitivity and specificity of 75% and 96% for detecting nodal metastasis [17]. However, some investigators have found lower sensitivity in women with early-stage disease [18], but we still recommend patients with newly diagnosed invasive cervical cancer to undergo a PET-CT to aid in treatment planning if feasible.

e a sensitivity and specificity of 75% and 96% for detecting nodal metastasis [17]. However, some investigators have found lower sensitivity in women with early-stage disease [18], but we still recommend patients with newly diagnosed invasive cervical cancer to undergo a PET-CT to aid in treatment planning if feasible. Despite best efforts to identify preoperatively patients who have metastatic disease, there will still be patients who on surgical exploration have metastatic disease. Controversy exists as to the best management; whether to proceed with radical hysterectomy or abandon for definitive chemoradiation. When there is disseminated disease to the peritoneum or abdomen, the prognosis is poor enough that there is little doubt that radical surgery will contribute significantly to improving outcome. Additionally, the finding of grossly positive para-aortic lymph nodes confers worse prognosis with five year overall survival of rates of 30%–40%. Performing radical surgery in this setting is unlikely to improve and may delay definitive treatment for recovery time. However, it is recommended to attempt surgical resection of grossly enlarged lymph nodes to improve radiation response and possibly improve survival [4, 19].

year overall survival of rates of 30%–40%. Performing radical surgery in this setting is unlikely to improve and may delay definitive treatment for recovery time. However, it is recommended to attempt surgical resection of grossly enlarged lymph nodes to improve radiation response and possibly improve survival [4, 19]. The finding of positive pelvic lymph nodes when exploring for radical surgery presents a unique dilemma. There are no prospective clinical trials randomizing patients into abandoned versus completed radical hysterectomy when unexpected metastatic disease to pelvic lymph nodes is found. Proponents for abandoning argue that patients suffer less radiation toxicity to small bowel, rectum, and bladder when the uterus is left in place, and have shorter interval to recovery and definitive treatment. Those in favor of completing the hysterectomy argue that removal of the primary tumor may reduce the recurrence risk with better pelvic control and allow for improved survival.

toxicity to small bowel, rectum, and bladder when the uterus is left in place, and have shorter interval to recovery and definitive treatment. Those in favor of completing the hysterectomy argue that removal of the primary tumor may reduce the recurrence risk with better pelvic control and allow for improved survival. The first report that described the abandoned (or aborted) radical hysterectomy in cervical cancer was from 1990, in which 15 women with aborted radical hysterectomy were matched with 15 women who had completed hysterectomy, but were found to have positive nodes on pathologic examination after surgery. They excluded all patients with positive para-aortic lymph nodes. Still, survival rate was 45% in the abandoned group versus 30% in the completed group, with only one case of radiationrelated morbidity (7%), which is low compared to most reports [9]. Hopkins and Morley reported in 1991 on 14 patients that had an abandoned hysterectomy, mostly for positive lymph nodes, (57%) and compared to a group of 26 patients found to have involved lymph nodes postoperatively. The overall survival was 50% in the abandoned group compared to 70% in the completed. Morbidities were higher (20%) for the completed compared to the aborted group(16%) [20].

d hysterectomy, mostly for positive lymph nodes, (57%) and compared to a group of 26 patients found to have involved lymph nodes postoperatively. The overall survival was 50% in the abandoned group compared to 70% in the completed. Morbidities were higher (20%) for the completed compared to the aborted group(16%) [20]. In total, there have been seven retrospective series of patients with abandoned hysterectomies reported in the literature. Table 3 summarizes major findings between the studies [9, 10, 20–24]. In general, the reported overall survival ranges from 31%–83% at five years. The earlier studies have a lower survival rate possibly due to lack of chemotherapy paired with radiation. The reported morbidities vary greatly (range 7 to 48%) and are fraught with reporting error, as these all are retrospective and patients are commonly lost to follow-up.

survival ranges from 31%–83% at five years. The earlier studies have a lower survival rate possibly due to lack of chemotherapy paired with radiation. The reported morbidities vary greatly (range 7 to 48%) and are fraught with reporting error, as these all are retrospective and patients are commonly lost to follow-up. In our study, we compared patients that had an abandoned hysterectomy to a group of women that had undergone radical hysterectomy, but had positive lymph nodes postoperatively for morbidities, recurrence rates and survival. There are clear limitations to using this control group, as women found to have only microscopically positive lymph nodes would inherently be expected to have higher survival rates than women with gross or suspicious lymph nodes. In our abandoned group, major toxicities related to treatment were common (26%), but lower in the completed group (34%), although this was not statistically significant. The increase appeared to be mostly due to increased severe radiation-related toxicity (16% versus 23%). Only two prior studies have reported on morbidity in the completed (surgery + radiation) group, which ranged from 22%–29% [20, 24]. One limitation of all of these retrospective studies, including ours, is that the reporting of morbidity is not standardized and can be underreported.

on-related toxicity (16% versus 23%). Only two prior studies have reported on morbidity in the completed (surgery + radiation) group, which ranged from 22%–29% [20, 24]. One limitation of all of these retrospective studies, including ours, is that the reporting of morbidity is not standardized and can be underreported. Progression free survival in our abandoned group was lower than what would be expected for early-stage disease (63%), although several patients were able to be salvaged by exenteration for an overall survival of 73%. Our findings were similar to previously reported studies (Table 3). A major limitation of this study is that it was underpowered to detect a difference in survival between the abandoned group and completed group. In conclusion, individuals with clinically early-stage cervical cancer who present with bleeding or long interval from prior pap smear should be considered at risk for more advanced disease. Preoperative imaging with CT or PET-CT may help to evaluate for metastatic spread. If faced with unexpected metastatic spread to pelvic lymph nodes at time of surgical exploration, many surgeons will choose to abandon the radical hysterectomy for primary chemoradiation therapy. However, completion of the hysterectomy in this setting may not worsen morbidity or effect overall survival. As it is unlikely there will ever be a prospective randomized trial to fully answer this question, our study suggests that either option is viable and should be discussed with the patient during the preoperative counseling.

pletion of the hysterectomy in this setting may not worsen morbidity or effect overall survival. As it is unlikely there will ever be a prospective randomized trial to fully answer this question, our study suggests that either option is viable and should be discussed with the patient during the preoperative counseling. Figure 1 Cumulative probability of recurrence free at two years of followup. Figure 2 Overall survival at five years for abandoned hysterectomy and completed hysterectomy groups. Table 1 Patient characteristics. Abandoned (n = 19) Completed (n = 44) P Age (Years)# 42 [41–54, 29–85] 44 [37–54, 21–85] .4207 Weight (kg)# 83 [65–85, 42–85] 71 [60–86, 41–121] .8727 Smoker (Yes) 9 (47%) 27 (61%) .4515 Pack per year# 18 [10–30, 2–52] 20 [10–30, 2–75] .682 Stage .4999 IB 0 (0%) 3 (7%) IB1 12 (63%) 29 (66%) IB2 5 (26%) 10 (23%) IIA 1 (5%) 2 (5%) IIB 1 (5%) 0 (0%) Histology .6747 Squamous 14 (74%) 31 (70%) Adenocarcinoma 4 (21%) 12 (27%) Adenosquamous 1 (5%) 1 (2%) Presenting symptoms .4407 Abnormal pap 3 (16%) 14 (32%) Vaginal bleeding 13 (68%) 23 (52%) Other 3 (16%) 3 (16%) Preoperative labs Hemoglobin# 12 [11.8–12.7, 9.6–14] 12.7 [12.1–13.4, 7.8–14.2] .3385 Hemotocrit# 34.5 [32–37, 29–41] 38 [35.8–39.3, 22–44] .2456 Prior pap (Years) 4 [2–9, 1–60] 2.5 [1–6,1–25] .2691 Preoperative CT scan⋆ No 7 (37%) NR Yes 12 (63%) NR Suspicious 6 (50%) NR #Median [IQR  =  interquartile Interval, Range  =  Minimum-Maximum]. ⋆NR  =  not recorded. Table 2 Surgical-and radiation-related complications.

Preoperative labs Hemoglobin# 12 [11.8–12.7, 9.6–14] 12.7 [12.1–13.4, 7.8–14.2] .3385 Hemotocrit# 34.5 [32–37, 29–41] 38 [35.8–39.3, 22–44] .2456 Prior pap (Years) 4 [2–9, 1–60] 2.5 [1–6,1–25] .2691 Preoperative CT scan⋆ No 7 (37%) NR Yes 12 (63%) NR Suspicious 6 (50%) NR #Median [IQR  =  interquartile Interval, Range  =  Minimum-Maximum]. ⋆NR  =  not recorded. Table 2 Surgical-and radiation-related complications. Complication Abandoned (N = 19) Completed (N = 44) OR, 95% CI and P-value Surgical Stroke 1 — Bowel perforation/obstruction 1 1 Hemorrhage (>1500 cc) — 1 Ureteral injury — 1 Iliac vein laceration — 1 Neurogenic/denervated bladder — 1 Total surgical complications 2/19 = 10% 5/44 = 11% (0.92, [0.08–6.633] and 1) Radiation-related Lymphedema 1 3 Radiation enteritis/proctitis 1 2 Bowel obstruction — 1 Severe diarrhea requiring TPN — 1 Rectovaginal fistula 1 — Radiation cystitis — 2 Vaginal stenosis — 1 Total radiation complications 3/19 = 16% 10/44 = 23% (0.64, [0.10–2.98] and 0.738) Total 5/19 = 26% 15/44 = 34% (0.69, [0.16–2.57] and 0.789) Note: All complications were grade 3 or higher. Some patients had more than one complication. Table 3 Summary of published reports of abandoned hysterectomy. Author Stage N Morbidity+ Recurrence Survival Potter et al. [9] IB-IIA 15 7% 53% 45%–50% Hopkins and Morley [20] IB-IIA 14 16% NR 50% Bremer et al. [21] IB-IIA 26 45% 39% 61% Whitney and Stehman [22] IB 68 19% 65% 31% Leath et al. [23] IB-IIA 23 34% 26% 83% Suprasert et al. [24] IB-IIA 23 48% 26% 59% Richard et al. [10] IB 55 NR NR 71% Gray 2010 IB-IIA 19 26% 37% 73% Total Abandoned 243 7%–48% 26%–65% 31%–83%

Author Stage N Morbidity+ Recurrence Survival Potter et al. [9] IB-IIA 15 7% 53% 45%–50% Hopkins and Morley [20] IB-IIA 14 16% NR 50% Bremer et al. [21] IB-IIA 26 45% 39% 61% Whitney and Stehman [22] IB 68 19% 65% 31% Leath et al. [23] IB-IIA 23 34% 26% 83% Suprasert et al. [24] IB-IIA 23 48% 26% 59% Richard et al. [10] IB 55 NR NR 71% Gray 2010 IB-IIA 19 26% 37% 73% Total Abandoned 243 7%–48% 26%–65% 31%–83% +Morbidities are combination of surgical-and-radiation related.

(p) ISO × Time 0.94 RT × Time 0.12 Interaction 0.85 Body fat (%) Baseline 46.2 ± 4.9 43.0 ± 4.4 46.8 ± 6.7 46.5 ± 5.7 After 9 months 45.6 ± 5.1 43.8 ± 5.8 48.4 ± 6.8 48.0 ± 6.1 ANOVA (p) ISO × Time 0.25 RT × Time 0.02 Interaction 0.20 Table 3 Daidzein and genistein plasma levels at 9 months of intervention (mean ± SD). Parameter Isoflavone Placebo P-values* (n = 35) (n = 36) Daidzein (μmol/dL) 220.4 ± 53.5 125.4 ± 27.9 <.0001 Genistein (μmol/dL) 144.3 ± 50.5 68.1 ± 19.5 <.0001 *Significant difference among groups (P < .05) (independent t-test).

1. Introduction Hereditary angioedema (HAE), a rare autosomal dominant disease, manifests as recurrent episodes of localized edema, which can involve the larynx and lead to upper airway obstruction and even fatal asphyxiation. Gut involvement results in occlusion, anorexia, vomiting, abdominal pain, and ascites. Symptoms usually begin at school age, and diagnosis is based on a familial history and laboratory data. In some cases, however, HAE occurs only during pregnancy, as severe attacks of abdominal pain [1–3]. Moreover, a normal complement C4 concentration [4] was previously considered to rule out HAE. Accordingly, patients without any positive family history or with normal C4 values [5, 6] often did not undergo assessment of C1-inhibitor protein levels and function [7]. The classic forms of HAE, types I and II, are caused by abnormal C1-inhibitor function, due to mutations in the gene encoding this inhibitor (SERPING1) [8, 9]. A third type, initially designated estrogen-dependent HAE, has recently been described in patients with normal C1-inhibitor concentration and function. This new type (called type III HAE) mostly affects women and appears to be triggered by exposure to high estrogen levels. It may also, however, occur before puberty, in postmenopausal women, and, rarely, in men [10–15]. In 2006, Dewald and Bork identified the causative mutations in the coagulation factor XII gene that lead to abnormal kinin generation [16]. In some families, all affected individuals in some families were reported to be women, but in others, men were also affected [11–13, 15]. Although high estrogen levels in these families may have influenced the frequency of attacks, exacerbations, or improvements, few obstetrical complications have been reported. No miscarriage or fetal or neonatal death was mentioned in a recent large cohort [15], although Bouillet et al. did report a case of in utero death in one French family [17].

in these families may have influenced the frequency of attacks, exacerbations, or improvements, few obstetrical complications have been reported. No miscarriage or fetal or neonatal death was mentioned in a recent large cohort [15], although Bouillet et al. did report a case of in utero death in one French family [17]. We report here obstetrical complications observed in two families with a missense mutation in the factor XII gene responsible for recurrent attacks of severe abdominal pain, transient maternal ascites, and fetal and neonatal deaths in two carriers from one of the families.

in these families may have influenced the frequency of attacks, exacerbations, or improvements, few obstetrical complications have been reported. No miscarriage or fetal or neonatal death was mentioned in a recent large cohort [15], although Bouillet et al. did report a case of in utero death in one French family [17]. We report here obstetrical complications observed in two families with a missense mutation in the factor XII gene responsible for recurrent attacks of severe abdominal pain, transient maternal ascites, and fetal and neonatal deaths in two carriers from one of the families. 2. Cases 2.1. Family 1 The patient, a 31-year-old Jewish woman, was followed at our tertiary center during her fourth pregnancy in a series of complicated pregnancies. During the first, ascites developed at 31 weeks of gestation (WG), and the fetus died in utero, of an unidentified cause (no chromosomal or morphological or placental anomalies). The simultaneous occurrence of these two events led us to the conclusion that they were linked. The second and third pregnancies were very painful, with abdominal attacks (pains and vomiting) at 10-day intervals. Cesarean deliveries were performed in both pregnancies, during abdominal crises with ascites, due to the risk of fetal death, and led to the birth, at 33 and 35 WG, of two healthy girls. Preeclampsia, antiphospholipid syndrome, thrombophilia, afibrinogenemia, and factor XIII deficiency were all ruled out. During the last pregnancy, the patient was admitted on several occasions for vomiting, abdominal pain, and ultrasound-documented ascites, associated once with acute kidney injury. Familial Mediterranean fever was ruled out, both by recurrence during colchicine treatment and by genetic analysis. At an extensive reinterview, the patient reported that abdominal symptoms occurred two days after swelling of the face and extremities, that these symptoms occurred only during pregnancy or when she used oral contraception, and generally it resolved within 72 hours. Testing showed that complement C4- and C1-inhibitor antigen levels were normal, whereas C1-inhibitor activity, assessed in an accredited laboratory (VFB) at 30 and 34 WG, was low (33% and 25% of normal values, resp.) (Table 1).

ng pregnancy or when she used oral contraception, and generally it resolved within 72 hours. Testing showed that complement C4- and C1-inhibitor antigen levels were normal, whereas C1-inhibitor activity, assessed in an accredited laboratory (VFB) at 30 and 34 WG, was low (33% and 25% of normal values, resp.) (Table 1). As HAE was suspected, infusion of a C1-inhibitor concentrate protocol (500 U, twice monthly) was initiated. After three courses of uneventful and efficacious treatment, a severe abdominal attack with abundant ascites at 35.5 WG led to the infusion of C1-inhibitor concentrate (1000 U Berinert P; CSL Behring, Marburg, Germany) and the extraction immediately thereafter of a healthy girl (2500 g). No episodes occurred either after delivery or during the three-year followup period, and C1-inhibitor function, measured at month 15, was normal (Table 1).

to the infusion of C1-inhibitor concentrate (1000 U Berinert P; CSL Behring, Marburg, Germany) and the extraction immediately thereafter of a healthy girl (2500 g). No episodes occurred either after delivery or during the three-year followup period, and C1-inhibitor function, measured at month 15, was normal (Table 1). The patient reported that her maternal grandmother and a maternal first cousin had complained of repeated episodes of facial swelling. Her mother had reported recurrent abdominal pain and vomiting without skin swelling, both exclusively during pregnancy. This family history is suggestive of HAE. Two maternal aunts were unaffected and had had normal pregnancies, but a third maternal aunt had three pregnancies (two boys and twin girls) complicated by recurrent severe vomiting without skin swelling and unexplained neonatal deaths. Subsequent testing showed her to have a normal concentration of functional C1-inhibitor. One cousin also had recurrent abdominal pains and ascites during her pregnancy. These clinical presentations lead us to the diagnosis of type III HAE.

recurrent severe vomiting without skin swelling and unexplained neonatal deaths. Subsequent testing showed her to have a normal concentration of functional C1-inhibitor. One cousin also had recurrent abdominal pains and ascites during her pregnancy. These clinical presentations lead us to the diagnosis of type III HAE. 2.2. Family 2 The patient, a 37-year-old Arab woman, was transferred to our center at 33 WG during her second pregnancy for suspected HAE. She had previously suffered a spontaneous first-trimester miscarriage of unknown etiology. She complained of edema of the face and larynx and abdominal pains, all occurring only during pregnancy. Her mother had died, possibly of laryngeal edema, during her fourth pregnancy. The patient's serum C4 component and C1-inhibitor antigen concentrations were normal, but her C1-inhibitor activity was 33% of the normal level (Table 1). The delivery of a healthy girl (4370 g) was induced at 41 WG for obstetric reasons after C1-inhibitor perfusion (1000 U). The peripartum period was uneventful, and functional C1-inhibitor levels reached normal values.

tigen concentrations were normal, but her C1-inhibitor activity was 33% of the normal level (Table 1). The delivery of a healthy girl (4370 g) was induced at 41 WG for obstetric reasons after C1-inhibitor perfusion (1000 U). The peripartum period was uneventful, and functional C1-inhibitor levels reached normal values. 2.3. Genetic Analysis Blood samples were taken after written informed consent. DNA was extracted and the gene encoding complement C1-inhibitor was analyzed and shown to be normal. We then searched for the recently described p. Thr328Lys mutation of F12 associated with this type of HAE. Direct sequencing of forward and reverse strands was performed with the BigDye terminator cycle sequencing kit (Applied Biosystems), using the same primers as for PCR amplification (5′-3′ F AAGCGCGGAACTGGGGAC; R CCG GCTGGCCGGAATCTA). This mutation was identified in both cases and in the affected aunt (family 1). Blood samples were not available from the grandmother, mother, unaffected aunts, or cousins of the first case (Figure 1). 3. Discussion This report establishes a diagnosis of atypical HAE during pregnancy in two women, based on clinical signs and a positive family history, despite late onset and normal C4 concentrations, and transient deficiencies of C1-inhibitor function.

This mutation was identified in both cases and in the affected aunt (family 1). Blood samples were not available from the grandmother, mother, unaffected aunts, or cousins of the first case (Figure 1). 3. Discussion This report establishes a diagnosis of atypical HAE during pregnancy in two women, based on clinical signs and a positive family history, despite late onset and normal C4 concentrations, and transient deficiencies of C1-inhibitor function. Most commonly, the symptoms of HAE due to C1-inhibitor deficiency begin at school age, worsen around puberty, and are due to mutations resulting in abnormal C1-inhibitor levels or function, associated with a low serum C4 concentration [1]. A new type of HAE with normal C1-inhibitor concentration and function, occurring almost entirely mostly in women, has been recently described [10, 11]. The symptoms and course of this type of HAE, designated as type III HAE, seem to differ from the classic forms (type I and type II HAE) by its later onset, longer disease-free intervals, more frequent facial swelling and tongue involvement, and less frequent attacks of abdominal pain [12]. Obstetrical complications were not reported in the patients in these studies or in those of a large Italian pedigree [14]. In our patients, HAE was manifested mainly during pregnancy, the clinical features were transient but recurrent maternal ascites, abdominal pain, and laryngeal edema. Fetal and neonatal deaths were observed in two carriers of the F12 gene mutation of one pedigree. To date, only one other in utero death has been described in a woman with HAE caused by an F12 mutation [17].

nancy, the clinical features were transient but recurrent maternal ascites, abdominal pain, and laryngeal edema. Fetal and neonatal deaths were observed in two carriers of the F12 gene mutation of one pedigree. To date, only one other in utero death has been described in a woman with HAE caused by an F12 mutation [17]. We suggest that when ascites occurs during pregnancy and is due neither to an obstetric cause such as preeclampsia or mirror syndrome nor to a standard nonobstetric cause, HAE should be suspected and the patient's family history investigated [18]. Screening to confirm suspected cases of HAE can measure the C4 concentration and C1-inhibitor antigen levels, but if clinical suspicion is strong, C1-inhibitor function must be assessed, followed by genetic tests if necessary [7]. Unfortunately these investigations, performed only in reference laboratories, are time consuming. Furthermore, although most HAE cases are caused by mutations of the C1-inhibitor gene, which leads to C1-inhibitor deficiency and low C4 levels (type I or type II HAE), HAE may also occur in patients with normal C4 levels and normal C1-inhibitor functions [5, 6, 10–16] and in women with transient C1-inhibitor deficiencies during pregnancy, as in our cases and another report [17]. When no mutation was detected in our patients for the gene encoding C1-inhibitor, we searched for a mutation in the coagulation factor XII gene. In 2006, Dewald and Bork demonstrated that mutations of this gene may be related to this novel form of HAE with normal C1-inhibitor levels and function [16]. The occurrence of this mutation raises questions about the mechanism underlying this novel type of HAE, designated as type III. Cichon et al. suggested that plasma factor XII activity might increase, while normal factor XII procoagulant levels are maintained as in our cases (results not shown), and that this could lead to increased bradykinin production and vascular permeability resulting in angioedema. Others, however, could not replicate these results during the symptom-free interval between attacks [18–21]. Furthermore, the positive regulation of factor XII gene transcription by estrogens may explain the strong influence of estrogens on the disease in some women [22]. Activation of the kallikrein-kinin system (or contact system), which induces increased bradykinin formation, appears to be the main mechanism in the pathophysiology of the trigger of HAE C1 inhibitor deficiency symptoms.

ption by estrogens may explain the strong influence of estrogens on the disease in some women [22]. Activation of the kallikrein-kinin system (or contact system), which induces increased bradykinin formation, appears to be the main mechanism in the pathophysiology of the trigger of HAE C1 inhibitor deficiency symptoms. If we consider that the trigger factors are similar in the three types of HAE, the contact system may be also involved in the physiopathology of HAE with F12 gene mutations [15, 23].

ption by estrogens may explain the strong influence of estrogens on the disease in some women [22]. Activation of the kallikrein-kinin system (or contact system), which induces increased bradykinin formation, appears to be the main mechanism in the pathophysiology of the trigger of HAE C1 inhibitor deficiency symptoms. If we consider that the trigger factors are similar in the three types of HAE, the contact system may be also involved in the physiopathology of HAE with F12 gene mutations [15, 23]. In clinical practice, HAE diagnosis should be based on clinical symptoms and confirmed by complement assessment and genetic analyses could be limited to cohort studies or for elucidation of the pathophysiology. Treatment of HAE during pregnancy presents special problems, particularly when HAE is clinically suspected and laboratory results are confusing or not yet available. In cases of a potentially fatal attack, an infusion of C1-inhibitor concentration (1000–1500 U) should be administered immediately even if the diagnosis of HAE is only presumptive (at least in the countries where this treatment is widely available) [4, 7]. Generally, in cases of C1 inhibitor deficiency, a positive response generally occurs within two hours in 95% of attacks; if symptoms persist, an additional infusion may be administered and alternative diagnoses considered [1]. Unfortunately, the efficacy of CI inhibitor concentration for type III HAE caused by mutations in the F12 gene varies substantially from very to not at all effective [15]. If attacks occur long before delivery, prophylaxis should be discussed after any emergency care. Androgens are theoretically contraindicated and are, in any case, ineffective [23]. Tranexamic acid, not recommended, may trigger thrombotic events [4]. Kallicrein inhibitor, bradykinin B2 receptor inhibitors (Icatibant), or recombinant C1-inhibitor may prove useful in the future but are unlikely to be allowed during pregnancy soon [7, 24, 25]. Currently, for type I and type II HAE, regular C1-inhibitor replacement therapy can be offered until delivery (once or twice a week), but it is difficult to predict the requirements of individual woman [4, 7, 26, 27]. For the peripartum period of type I or type II HAE, the safest approach appears to be the systematic administration of a predelivery infusion of 500 or 1000 U of C1 inhibitor concentration [4]. Unfortunately, no recommendations yet exist for type III HAE. However, in view of Bouillet's experience and the efficiency of the administration of 1000 U of C1-inhibitor concentration in the two patients reported here, this predelivery infusion may be useful during the peripartum period [17].

ibitor concentration [4]. Unfortunately, no recommendations yet exist for type III HAE. However, in view of Bouillet's experience and the efficiency of the administration of 1000 U of C1-inhibitor concentration in the two patients reported here, this predelivery infusion may be useful during the peripartum period [17]. Furthermore, because trigger factors appear to be the same as those of C1 inhibitor deficiency, regional analgesia is to be preferred to endotracheal intubation to avoid laryngeal edema if operative delivery is undertaken [4, 15]. Obstetricians and physicians should be aware of the maternal and fetal complications of HAE, a rare disease that may occur only during pregnancy. It may be suspected on clinical grounds and may require specific emergency care or prophylaxis. Figure 1 Mutation in the F12 gene in two pedigrees. The pedigree of the two families is shown. Individuals are identified by Arabic numerals within each generation (roman numerals). Affected individuals are indicated with asterisks. A heterozygous mutation was found in subjects with solid symbols. Electrophoregrams correspond to the DNA sequence surrounding the mutated nucleotide in the F12 gene. In exon 9, a C > G heterozygous missense mutation was noted in both patients. Direct sequencing of forward and reverse strands was carried out with the BigDye terminator cycle sequencing kit (Applied Biosystems), using the same primers as for PCR amplification (5′-3′ F AAGCGCGGAACTGGGGAC; R CCG GCTGGCCGGAATCTA).

gene. In exon 9, a C > G heterozygous missense mutation was noted in both patients. Direct sequencing of forward and reverse strands was carried out with the BigDye terminator cycle sequencing kit (Applied Biosystems), using the same primers as for PCR amplification (5′-3′ F AAGCGCGGAACTGGGGAC; R CCG GCTGGCCGGAATCTA). Table 1 Complement profiles in patients. The diagnosis of patient 1 was reviewed two years after the fourth pregnancy. Complement was quantified for patient 1 at 30 and 34 weeks of gestation (W) and 15 months after the pregnancy (M15) and for patient 2 at 29 weeks and 33 weeks of gestation and 2 months after the pregnancy (M2). Plasma concentrations of C1 inhibitor (C1 inh), C4, and C3 were determined by nephelemetry (Dade Behring). Normal values ranged between 170 and 540 mg/L for C1-inhibitor (C1 Inh), 93 and 380 mg/L for C4, and 660 and 1250 mg/L (±2 SD) for C3. CH50 were determined according to standard procedures. Results are expressed as percent of the CH50 of the reference plasma pool (obtained from one hundred healthy blood donors). C1-inhibitor function (C1 inh fx) was assessed in a chromogenic assay (Technochrom, Biolys, Taluyers, France). Patient 1 Patient 2 Units 30 w 34 w M15 29 w 33 w M2 CH50 70–130 % 155 144 131 114 120 107 C3 660–1250 mg/L 1470 1200 989 1270 1310 926 C4 93–380 mg/L 425 352 320 215 257 214 C1 Inh 170–540 mg/L 202 186 239 148 161 190 C1 inh fx 70–130 % 33 25 101 33 35 84

1. Introduction A decrease in estrogen production in women at menopause is accompanied by changes in body composition which are characterized by body fat increase, primarily at the abdomen, and progressive reduction in muscle mass (sarcopenia) [1]. Sarcopenia is usually associated with functional impairment and physical disability, especially in women, and is the direct cause of reduction in muscle strength [2–4]. Menopause may induce a phase of rapid decreases in aerobic fitness, muscle strength, and balance, especially in sedentary women [5]. Several factors such as diet, lifestyle, as well as metabolic and hormonal parameters influence body composition in postmenopausal women [1, 6, 7]. The importance of physical activity at all stages of life for optimal bone health and muscle mass is established [8]. Aerobic exercise is important in maintaining overall health. Nonetheless, resistance muscle training may be more applicable to site-specific effects of exercise, besides having more favorable effects on maintaining or improving muscle mass and strength. Recently, we studied the effect of 16 weeks of resistance training (60%–80% of one-repetition maximum—1-RM) three times a week in 43 postmenopausal women, and we observed increases in maximal strength and muscle mass, indicating that resistance training may be applied in the rehabilitation or prevention of sarcopenia in postmenopausal women [9]. Besides, the abdominal fat reduction is associated with resistance training in older women [10, 11].

n 43 postmenopausal women, and we observed increases in maximal strength and muscle mass, indicating that resistance training may be applied in the rehabilitation or prevention of sarcopenia in postmenopausal women [9]. Besides, the abdominal fat reduction is associated with resistance training in older women [10, 11]. Isoflavones belong to a class of compounds called phytoestrogens. The primary isoflavones found in soy are genistein, daidzein, and glycitein. These nonsteroidal compounds are structurally [12] and weakly bound to estrogenic receptors [13]. Isoflavones may have a direct effect on muscle via estrogen receptors [14], increase of blood concentration of insulin-like growth factor-1 [15], or through their effects on reducing inflammation [16]. Besides, estrogen receptors are present in adipose tissue [12]. Thus, isoflavones could participate in the regulation of adiposity [12, 17]. However, the details of combined effects of regular resistance exercise and isoflavone on the prevention muscle loss or fat gain in women estrogen deficient remain to be solved. We observed that 16 weeks of supplementation with soy protein associated with resistance training did not result in greater increase in strength or muscle mass compared with placebo in postmenopausal women [10]. In this study, the time of intervention of the isoflavone could have affected the results. Two studies demonstrated association between muscle mass and isoflavone; however, the postmenopausal women received isoflavone for 24 weeks [18, 19].

in strength or muscle mass compared with placebo in postmenopausal women [10]. In this study, the time of intervention of the isoflavone could have affected the results. Two studies demonstrated association between muscle mass and isoflavone; however, the postmenopausal women received isoflavone for 24 weeks [18, 19]. Therefore, clinical trials of longer duration are needed to determine the effect of exercise training combined with isoflavone on body composition. Within this context, the primary purpose of this study was to determine the isolated and combined effects of resistance training and soy isoflavone on body composition (fat and muscle mass) in postmenopausal women.

ion are needed to determine the effect of exercise training combined with isoflavone on body composition. Within this context, the primary purpose of this study was to determine the isolated and combined effects of resistance training and soy isoflavone on body composition (fat and muscle mass) in postmenopausal women. 2. Methods 2.1. Study Design and Participants This clinical randomized, double-blind, placebo-controlled trial included 80 Brazilian women seen at the Climacterium Outpatient Service of the Department of Gynecology of Sao Paulo State University. All subjects were postmenopausal women aged 45–70 years with good overall health, sedentary, spontaneous amenorrhea for at least 12 months, and follicle-stimulating hormone level greater than 40 mIU/mL. Exclusion criteria included (1) high-fiber or high-soy diet, (2) hormone therapy (HT) or phytoestrogens use within the preceding 6 months, (3) noncontrolled high-blood pressure, (4) history of myopathic, neuropathic, and skeletal disorders, (5) history of breast cancer, endometrial carcinoma, cardiovascular disease, and thromboembolic disorders, (6) undiagnosed vaginal bleeding, (7) alcoholism, and (8) chronic gastrointestinal diseases. Subjects were considered to be sedentary when they reported no leisure physical activities besides everyday household tasks. Informed consent was obtained from all participants, and the study was approved by the Research Ethics Committee of Botucatu Medical School, UNESP.

ism, and (8) chronic gastrointestinal diseases. Subjects were considered to be sedentary when they reported no leisure physical activities besides everyday household tasks. Informed consent was obtained from all participants, and the study was approved by the Research Ethics Committee of Botucatu Medical School, UNESP. Initial assessment consisted of case history taking, general and gynecological physical examination, oncotic colpocytology, mammography, and transvaginal ultrasonography. Data collected included information on age, menarche, time since menopause, parity, HT contraindication, weight, height, and waist circumference. After a prestudy period, participants were randomly assigned to one of two groups: ISO, receiving soy isoflavone extract (n = 40) or P, receiving placebo (n = 40). Examiners and subjects had no previous knowledge of group assignment (double-blind). Placebo and active treatment were identical in appearance. The centralized computerized subject randomization process was conducted by a statistician unaware of the study protocol using specific software. Capsule packages were labeled with code numbers and the statistician in charge was the only unblinded person.

ind). Placebo and active treatment were identical in appearance. The centralized computerized subject randomization process was conducted by a statistician unaware of the study protocol using specific software. Capsule packages were labeled with code numbers and the statistician in charge was the only unblinded person. Thus, 40 participants were given 250 mg of standardized soy extract (Glycine max), corresponding to 100 mg/day of isoflavone, twice a day in capsules containing 125 mg of soy extract plus 50 mg of isoflavones each, in glycoside form. The standardized extract contained approximately 50% of genistein and 35% of daidzein. The other 40 participants received two lactose capsules per day. All of the capsules were identical in appearance. Subjects were instructed to return any unused medication during visits for compliance determination. Subsequently, the participants were randomized to resistance training (RT) or No RT. Thus, 80 women were randomly assigned to one of four groups: RT + ISO (n = 20), No RT + ISO (n = 20), RT + P (n = 20), and No RT + P (n = 20). Followup length was nine months and evaluations were performed during the prestudy period, at baseline and at the end of the study. Nine women withdrew from the study (7 in RT and 2 in No RT) because of illness, family problems, or protocol violation (less than 2 weekly RT sessions). The remaining 71 subjects completed the nine-month study, and their data were included in the per-protocol analysis.

y period, at baseline and at the end of the study. Nine women withdrew from the study (7 in RT and 2 in No RT) because of illness, family problems, or protocol violation (less than 2 weekly RT sessions). The remaining 71 subjects completed the nine-month study, and their data were included in the per-protocol analysis. All participants realized alimentary register. Three-day diaries (2 weekdays and 1 weekend day) and 24-hour recalls were completed at baseline of the study to estimate caloric intake and diet composition. Total energy, macronutrients (proteins, fats, and carbohydrates), and calcium contents were determined using “NutWin” (nutrition analysis software). Participants were advised to keep their usual dietary habits. The participants that showed inadequate data were oriented for ideal.

te caloric intake and diet composition. Total energy, macronutrients (proteins, fats, and carbohydrates), and calcium contents were determined using “NutWin” (nutrition analysis software). Participants were advised to keep their usual dietary habits. The participants that showed inadequate data were oriented for ideal. 2.2. Exercise Intervention The resistance-training protocol, as adapted for women over 45 years, consisted in the minimum 2 weekly sessions, on nonconsecutive days, for 9 months, under the supervision of trained personnel. Before training, subjects in the exercise group attended a 4-week adaptation period in order to get acquainted with the protocol. Initially, lighter loads were used and subjects performed one set of 15 repetitions at 40%–50% of 1-RM. Progression was gradual untill 3 sets of 8–12 repetitions at 60%–80% of 1-RM were performed [20]. The protocol consisted of dynamic exercises for both lower and upper limbs for a total of 50–60 minutes. Two exercises for each major muscle group (chest, back, and thigh) and one for each minor muscle group (biceps and triceps) were used, in 3 series of 8–12 repetitions. The load setting for each individual exercise was established by 1-RM testing. Loads were periodically adjusted at the end of each month to maintain loads at 8–12 repetitions maximum, according to strength gain and adaptation to training. Subjects performed one set of each of the following exercises: leg press, leg extension, peck deck, bench press, seated row, lat pull-down, triceps pulley, and biceps curl. Abdominal (3 sets of 30 repetitions) and calf exercises (3 sets of 20 repetitions with body weight) were also included. Air breathing was controlled by expiring during concentric actions and inspiring over eccentric actions in order to prevent apnea. Rest breaks between series and exercises ranged from 1 minute and 30 seconds to 2 minutes. During training sessions, participants were advised to perform an eccentric action in 2 second and a concentric action in 1 second. Subjects were considered as trained if they had worked out at least two times a week, on nonconsecutive days. Attendance was recorded by the trainers. Participants of the no-RT groups were asked to maintain their usual level of physical activity. No cardiovascular training was mentioned.

concentric action in 1 second. Subjects were considered as trained if they had worked out at least two times a week, on nonconsecutive days. Attendance was recorded by the trainers. Participants of the no-RT groups were asked to maintain their usual level of physical activity. No cardiovascular training was mentioned. Muscle strength was assessed in all subjects, by the 1-RM test, for dynamic exercises, after an adaptation period for strength testing (3–5 sessions on nonconsecutive days). The 1-RM was defined as the maximal weight that could be lifted with proper body alignment and the correct lifting technique. Muscle strength assessment included all of exercises at baseline, but only leg extension after nine months; because of cultural aspects related to strength, some women did not accomplish other tests. The 1-RM test was performed using the same apparatus used for resistance training.

nt and the correct lifting technique. Muscle strength assessment included all of exercises at baseline, but only leg extension after nine months; because of cultural aspects related to strength, some women did not accomplish other tests. The 1-RM test was performed using the same apparatus used for resistance training. 2.3. Body Composition Weight variation was assessed by the body mass index (BMI = weight/height2). Height and weight were obtained using a stadiometer (Seca, Brazil) and a standard balance beam scale (Filizola, Brazil), respectively, with subjects wearing lightweight clothing and no shoes. Weight was classified according to the system used by the World Health Organization (2000): 18.5 to 24.9 kg/m2 = normal weight, 25 to 29.9 kg/m2 = overweight, and ≥30.0 kg/m2 = obese [21]. Waist circumference was measured to the nearest 0.5 cm midway between the lowest rib margin and the iliac crest in supine position. Abdominal fat was indirectly assessed by measuring waist circumference, and was considered high when waist was >88 cm.

weight, 25 to 29.9 kg/m2 = overweight, and ≥30.0 kg/m2 = obese [21]. Waist circumference was measured to the nearest 0.5 cm midway between the lowest rib margin and the iliac crest in supine position. Abdominal fat was indirectly assessed by measuring waist circumference, and was considered high when waist was >88 cm. Body composition (fat and lean mass) was assessed by dual-energy X-ray absorptiometry (DXA) at baseline and at nine months using a Hologic QDR-2000 densitometer plus scanner (Hologic, Waltham, MA, USA). To minimize interobserver variation, all scans and tests were carried out by the same investigator. Day-to-day percent coefficient of variations was <1.0% for whole-body. Whole body scans were divided into several regions such as arms, legs, trunk (pelvis, spine, and ribs), and head. Body composition was assessed by using the manual DXA analysis software (version 5.73A for whole body). The arm region was defined as the region extending from the head of the humerus to the distal tip of the fingers. The reference point between the head of the humerus and the scapula was positioned at the glenoid fossa. The leg region was defined as the region extending from the inferior border of the ischial tuberosity to the distal tip of the toes. The sub-whole-body was defined as the region extending from the shoulders to the distal tip of the toes. Reference points that could be clearly visualized on the DXA system terminal were selected.

egion was defined as the region extending from the inferior border of the ischial tuberosity to the distal tip of the toes. The sub-whole-body was defined as the region extending from the shoulders to the distal tip of the toes. Reference points that could be clearly visualized on the DXA system terminal were selected. 2.4. Statistical Analysis Statistical analyses were performed using the Software STATISTICA for Windows (version 6.0). The statistical distribution of the variables was evaluated by the tests of Shapiro-Wilk, Kolmogorov & Smirnov, and Levene. Normally distributed variables were reported as mean ± standard deviations. Differences among groups in baseline characteristics were evaluated using one-way analysis of variance. When differences were detected, a Tukey's post hoc test was performed to determine pairwise differences. Primary analyses included two-way repeated measures analysis of variance to test for interaction and for main effects of soy isoflavone and resistance training. All analyses were performed with no intention-to-treat. Comparison of daidzein and genistein levels between ISO and PL groups was carried out by an independent t-test. Exact P values were obtained from the tests employed. Statistical tests were two tailed and significance was set at 5%.

flavone and resistance training. All analyses were performed with no intention-to-treat. Comparison of daidzein and genistein levels between ISO and PL groups was carried out by an independent t-test. Exact P values were obtained from the tests employed. Statistical tests were two tailed and significance was set at 5%. 3. Results Of the 80 postmenopausal women who were randomized, 71 completed this 9-month trial in the following four groups: RT + ISO (n = 15), No RT + ISO (n = 20), RT + P (n = 18), and No RT + P (n = 18). As expected, dropout rates were higher in the RT groups (n = 7) than in the No RT groups (n = 2) because of the time commitment. However, no differences in dropout rates were observed between ISO (n = 5) and Placebo (n = 4) groups. At baseline, weight and BMI were lowers in the RT + P group compared with No RT + ISO and No RT + P, respectively, but not significantly different in other variables. Subject characteristics, fat, and muscle mass (MM) showed no significant differences among groups (Table 1). No significant differences were observed in energy or macronutrient intake per kg body mass (P > .05). Energy/kg intakes were 24 ± 6, 25 ± 7, 23 ± 6, and 22 ± 6 kcal/kg per day, carbohydrate/kg intakes were 2.5 ± 0.6, 2.2 ± 1.4, 2.2 ± 1.0, and 2.4 ± 0.6 g/kg/day, protein/kg intakes were 1.0 ± 0.4, 1.1 ± 0.5, 1.1 ± 0.6, and 1.1 ± 0.5 g/kg/day, and fat/kg intakes were 0.82 ± 0.32, 0.84 ± 0.29, 0.82 ± 0,34, and 0.83 ± 0.28 g/kg/day for RT + ISO, RT + P, No RT + ISO, and No RT + P, respectively.

, carbohydrate/kg intakes were 2.5 ± 0.6, 2.2 ± 1.4, 2.2 ± 1.0, and 2.4 ± 0.6 g/kg/day, protein/kg intakes were 1.0 ± 0.4, 1.1 ± 0.5, 1.1 ± 0.6, and 1.1 ± 0.5 g/kg/day, and fat/kg intakes were 0.82 ± 0.32, 0.84 ± 0.29, 0.82 ± 0,34, and 0.83 ± 0.28 g/kg/day for RT + ISO, RT + P, No RT + ISO, and No RT + P, respectively. Only groups with RT (RT + ISO and RT + P) increased the strength. However, no interaction between isoflavone and resistance training on muscle strength changes was found (Figure 1). There was a main effect of resistance training for muscle strength such that muscle fitness during the 9-month intervention was improved in the RT groups (35.2%) and reduced in the No RT groups (−1.1%). No main effect of soy isoflavone for muscle strength (ISO versus Placebo groups) was observed. Over body composition (fat and muscle mass), there was a main effect of resistance training (Table 2) such that MM improved in the RT groups (1.4% total-MM and 3.6% limb-MM), and decreased in the No RT groups (−1.8% total-MM and −0.7% limb-MM) during the 9-month intervention. Exercising attenuated fat trunk gains (RT = 6.1% versus No RT = 11.2%, P < .05) and % body fat (RT = 0.2% versus No RT = 2.1%, P < .05). No main effect of soy (ISO versus Placebo groups) was observed (Table 2). There were no significant interactions of isoflavone associated with resistance training on changes in fat mass or MM (Table 2).

ed fat trunk gains (RT = 6.1% versus No RT = 11.2%, P < .05) and % body fat (RT = 0.2% versus No RT = 2.1%, P < .05). No main effect of soy (ISO versus Placebo groups) was observed (Table 2). There were no significant interactions of isoflavone associated with resistance training on changes in fat mass or MM (Table 2). Table 3 shows that daidzein and genistein plasma concentrations significantly differed among groups after 9 months of intervention (P < .001). In the subjects given standardized soy extract, the detectable levels of both isoflavones were significantly higher than in the placebo group. Seven of the 35 soy group subjects (20%) and four of the 36 placebo group subjects (11.1%) reported adverse experiences, most frequently in the gastrointestinal system. No serious adverse event related to isoflavone treatment was reported. 4. Discussion The findings of this investigation assessing the isolated and combined effects of exercise and soy isoflavone for 9 months in postmenopausal women were the following: (1) resistance training favorably changed body composition increasing muscle mass and strength, and attenuating % body fat gains, (2) standardized soy extract (100 mg/day of isoflavone) did not change body composition (fat and muscle mass), and (3) there were no apparent additive or synergistic effects of soy isoflavone and resistance training on body composition.

composition increasing muscle mass and strength, and attenuating % body fat gains, (2) standardized soy extract (100 mg/day of isoflavone) did not change body composition (fat and muscle mass), and (3) there were no apparent additive or synergistic effects of soy isoflavone and resistance training on body composition. Important changes in the body composition of a female occur around the menopausal period. These changes include a decrease in bone, strength, and muscle mass and an increase in fat mass that can be managed by exercise, nutrition, and estrogen replacement, or by their combined intervention [1, 7, 8, 22]. In this study, the subjects engaged in resistance training gained strength and muscle mass (MM), whereas those in the nonexercising groups exhibited loss. Progressive resistance exercise or strength training is considered a promising intervention for reversing the loss of muscle function and the deterioration of muscle structure that is associate with aging [7, 23]. The American College of Sports Medicine currently recommends strength or resistance training as key components of an overall fitness program [20]. There is general consensus that physical exercise is a powerful intervention to obtain long-term benefits in muscle function, to reduce the frequency of falls, and to maintain independence and a high quality of life in older persons [3, 24]. Although the skeletal and muscular systems are structurally interdependent, both adapt to mechanical loading; however, the nature of their relationship is still unknown. The factors necessarily required to elicit changes to these systems have been studied. Fat-free has been suggested as a stronger determinant of bone mass with aging than either total mass or fat mass, although fat mass may also be an independent determinant [25].

ever, the nature of their relationship is still unknown. The factors necessarily required to elicit changes to these systems have been studied. Fat-free has been suggested as a stronger determinant of bone mass with aging than either total mass or fat mass, although fat mass may also be an independent determinant [25]. In women, muscle mass and strength decline during perimenopause and this phenomenon seems to be estrogen dependent [1, 7]. Theoretically, estrogen has a direct anabolic action on the skeletal muscle that contains estrogenic receptors [14]. Aubertin-Leheudre et al. [18] investigated whether six months of isoflavone supplementation (70 mg of isoflavones/day) would increase fat-free mass and muscle mass index in obese-sarcopenic postmenopausal women. After 24 weeks, they found that limb and leg fat-free mass and muscle mass index (MMI = appendicular FFM/height2) significantly increased in the isoflavone group, but not in the placebo group. Although the data suggest that isoflavone improves muscle mass, we did not observe association in this study and other ones [10]. A possible explanation would be that these nonsteroidal compounds are weakly bound to estrogenic receptors (<1% of estradiol binding affinity) [13]. Besides, soy isoflavones preferentially bind to β-estrogen receptors that are found in the central nervous system, bones, vascular walls, and the urogenital tract. Unlike estrogens, isoflavones have little affinity with α-estrogen receptors [26], which are found in larger amount in the muscle [14]. Additionally, the form of isoflavones may impact the appropriateness of the dose used in present study. Izumi et al. [27] showed that the isoflavone aglycones were absorbed faster and in greater amounts than their glycosides in humans. Nevertheless, these findings have not been universal [28, 29].

t in the muscle [14]. Additionally, the form of isoflavones may impact the appropriateness of the dose used in present study. Izumi et al. [27] showed that the isoflavone aglycones were absorbed faster and in greater amounts than their glycosides in humans. Nevertheless, these findings have not been universal [28, 29]. Though the weight and IMC have been smaller in RT + P, compared with other groups, the other variables of the corporal composition (muscle mass and body fat) and strength muscle did not show significant differences (Table 1). The postmenopausal women participating in our study were overweight, with an elevate rate of body fat and android fat distribution. Menopausal transition has been associated with a preferential increase in intraabdominal fat that is independent of age and total body fat mass [1], supporting the participation of sex hormones in body fat distribution. In this study, no reduction in body fat was observed after the nine-month intervention. Nonetheless, two weekly sessions of resistance training associated with lower trunk fat gains and lower % body fat gains. Resistance training has been shown to increase resting and total energy expenditure, and to induce decreases in total and abdominal fat [10, 11]. Hunter et al. [11] demonstrated that resistance training reduces the percentage of subcutaneous and intraabdominal fat in women, On the other hand, studies conducted in postmenopausal women showed no reduction in fat tissue [30, 31].

total energy expenditure, and to induce decreases in total and abdominal fat [10, 11]. Hunter et al. [11] demonstrated that resistance training reduces the percentage of subcutaneous and intraabdominal fat in women, On the other hand, studies conducted in postmenopausal women showed no reduction in fat tissue [30, 31]. Little is known about the effects of isoflavones on body composition or fat and lean mass distribution in postmenopausal women. The rationale for studying the effect of soy protein on body fat distribution is that isoflavone may bind to estrogenic receptors of fat and lean tissues promoting gynoid fat deposition [19]. Goodman-Gruen and Kritz-Silverstein [32] investigated the association between dietary isoflavone intake and body composition in 208 participants to the Soy Health Effects Study (aged 45–74 years). They observed a significant inverse relation between soy consumption and body mass index, abdominal circumference and fat mass, as well as no correlation with lean mass, suggesting that an isoflavone-rich diet is associated with reduced body fat and may play a role in the prevention of obesity-related chronic diseases.

They observed a significant inverse relation between soy consumption and body mass index, abdominal circumference and fat mass, as well as no correlation with lean mass, suggesting that an isoflavone-rich diet is associated with reduced body fat and may play a role in the prevention of obesity-related chronic diseases. In agreement with our findings, Moeller et al. [19] assessed body composition and physical activity in 69 perimenopausal women and found that 40 g of soy protein/day did not prevent abdominal fat deposition, but physical activity positively correlated with the regional distribution of fat and muscle mass. In 2007, Maesta et al. [10] assessed the effect of soy protein and progressive resistance training on body composition in postmenopausal women concluding that there were no additive effects of soy and exercise. Soy protein supplementation did not influence body composition indicators. Indeed, the increase in muscle mass and reduction in abdominal fat were correlated with resistance training.

sive resistance training on body composition in postmenopausal women concluding that there were no additive effects of soy and exercise. Soy protein supplementation did not influence body composition indicators. Indeed, the increase in muscle mass and reduction in abdominal fat were correlated with resistance training. The findings herein reported should be interpreted with caution as this study presented some limitations. First, the sample size was relatively small due to the inclusion criteria and nature of the intervention (resistance training) used. Second, although the participants were oriented, the ingestion of food with isoflavone was not controlled. This might explain high daidzein and genistein plasma levels the in the placebo group, interfering in difference between perceptual changes. Finally, the dropout rate ranged from 2.5% in the No RT groups to 20% in the RT groups. These rates are high, but still comparable with those observed in other trials involving exercising [33]. As a matter of fact, the most frequently reported reason for dropping-out and poor compliance was lack of interest to continue with the resistance-training protocol. This study was designed to assess the effect of resistance training combined with soy isoflavone on body composition in the women who completed the program, and not to determine the effectiveness of such intervention in a public health or community setting. Thus, the subjects who dropped out or did not meet compliance requirements were not retested and no intention-to-treat analysis was performed post hoc.

flavone on body composition in the women who completed the program, and not to determine the effectiveness of such intervention in a public health or community setting. Thus, the subjects who dropped out or did not meet compliance requirements were not retested and no intention-to-treat analysis was performed post hoc. In conclusion, our results indicate that standardized soy extract (100 mg/day of isoflavone) did not change body composition (fat and muscle mass) and that soy isoflavone and resistance training had no apparent additive or synergistic effects. Resistance training was found to be a safe and effective intervention for reversing the loss of muscle function by increasing muscle mass and strength and to attenuate fat trunk gains and % body fat in postmenopausal women. Acknowledgments This investigation was supported by Ativus Farmacêutica, Brazil, and grants by Fundação Lucentis de Apoio a Cultura, Ensino, Pesquisa e Extensão. Figure 1 Percent changes in muscle strength (leg extension) from baseline to 9 months in response to exercise and isoflavone. ANOVA (p): ISO × Time = 0.82. RT × Time = 0.02. Interaction = 0.86. Table 1 Baseline clinical characteristics of 71 postmenopausal women (mean ± SD).

Acknowledgments This investigation was supported by Ativus Farmacêutica, Brazil, and grants by Fundação Lucentis de Apoio a Cultura, Ensino, Pesquisa e Extensão. Figure 1 Percent changes in muscle strength (leg extension) from baseline to 9 months in response to exercise and isoflavone. ANOVA (p): ISO × Time = 0.82. RT × Time = 0.02. Interaction = 0.86. Table 1 Baseline clinical characteristics of 71 postmenopausal women (mean ± SD). Variables RT + ISO RT + P No RT + ISO No RT + P P value* (n = 15) (n = 18) (n = 20) (n = 18) Age (years) 55.7 ± 7.0 56.6 ± 8.8 56.0 ± 5.8 55.3 ± 8.0 .99 Postmenopause (years) 7.7 ± 5.3 8.7 ± 6.1 6.5 ± 4.9 7.7 ± 6.2 .64 FSH (mUI/mL) 74.0 ± 27.7 71.8 ± 24.3 63.0 ± 21.1 63.3 ± 21.3 .15 E2 (pg/mL) 23.1 ± 4.8 21.7 ± 4.3 22.5 ± 4.8 23.3 ± 5.8 .59 Weight (kg) 73.2 ± 15.1 62.8 ± 9.8(a) 75.0 ± 12.9(a) 73.7 ± 12.6 .03 Height (cm) 154.9 ± 7.0 154.7 ± 5.6 158.1 ± 5.8 155.6 ± 5.2 .27 WC (cm) 95.7 ± 10.1 87.3 ± 7.7 94.1 ± 12.3 95.1 ± 11.9 .09 BMI (kg/m2) 30.3 ± 4.7 26.0 ± 3.0(a) 30.1 ± 4.7 30.4 ± 5.3(a) .03 MM (kg) 34.0 ± 6.2 30.3 ± 4.1 34.0 ± 4.2 33.6 ± 4.0 .08 MM-Limbs (kg) 13.6 ± 2.8 12.0 ± 1.9 13.7 ± 2.0 13.6 ± 2.0 .08 MM (%) 52.2 ± 4.5 54.3 ± 4.2 51.2 ± 6.2 51.0 ± 5.1 .20 Trunk fat mass (kg) 14.3 ± 4.4 12.3 ± 3.2 15.8 ± 5.4 14.8 ± 4.4 .11 Body fat (%) 46.2 ± 4.9 43.0 ± 4.4 46.8 ± 6.7 46.5 ± 5.7 .12 Strength muscle (kg) 30.7 ± 9.2 31.3 ± 6.8 33.0 ± 10.2 32.7 ± 7.9 .15 FSH: follicle-stimulating hormone; E2: estradiol; WC: waist circumference; BMI: body mass index; MM: muscle mass; SD: standard deviations; ISO: isoflavone; PL: placebo; RT: resistance training.

fat (%) 46.2 ± 4.9 43.0 ± 4.4 46.8 ± 6.7 46.5 ± 5.7 .12 Strength muscle (kg) 30.7 ± 9.2 31.3 ± 6.8 33.0 ± 10.2 32.7 ± 7.9 .15 FSH: follicle-stimulating hormone; E2: estradiol; WC: waist circumference; BMI: body mass index; MM: muscle mass; SD: standard deviations; ISO: isoflavone; PL: placebo; RT: resistance training. *Significant difference among groups (P < .05) (one-way ANOVA),(a)With significantly difference between groups (P < .05). Table 2 Effects of soy (ISO) and/or exercise (RT) on body composition (fat and muscle mass) in the study groups at baseline and at 9 months (mean ± SD).

fat (%) 46.2 ± 4.9 43.0 ± 4.4 46.8 ± 6.7 46.5 ± 5.7 .12 Strength muscle (kg) 30.7 ± 9.2 31.3 ± 6.8 33.0 ± 10.2 32.7 ± 7.9 .15 FSH: follicle-stimulating hormone; E2: estradiol; WC: waist circumference; BMI: body mass index; MM: muscle mass; SD: standard deviations; ISO: isoflavone; PL: placebo; RT: resistance training. *Significant difference among groups (P < .05) (one-way ANOVA),(a)With significantly difference between groups (P < .05). Table 2 Effects of soy (ISO) and/or exercise (RT) on body composition (fat and muscle mass) in the study groups at baseline and at 9 months (mean ± SD). Variables RT + ISO RT + P No RT + ISO No RT + P (n = 15) (n = 18) (n = 20) (n = 18) MM (Kg) Baseline 34.0 ± 6.2 30.3 ± 4.1 34.0 ± 4.2 33.6 ± 4.0 After 9 months 34.6 ± 6.1 30.6 ± 3.7 33.2 ± 3.8 33.1 ± 4.1 ANOVA (p) ISO × Time 0.61 RT × Time <0.01 Interaction 0.19 Limb-MM (kg) Baseline 13.6 ± 2.8 12.0 ± 1.9 13.7 ± 2.0 13.6 ± 2.0 After 9 months 14.2 ± 2.9 12.3 ± 1.9 13.7 ± 2.2 13.4 ± 1.9 ANOVA (p) ISO × Time 0.05 RT × Time <0.00 Interaction 0.67 MM (%) Baseline 52.2 ± 4.5 54.3 ± 4.2 51.2 ± 6.2 51.0 ± 5.1 After 9 months 52.4 ± 4.8 54.3 ± 5.0 50.0 ± 6.0 50.3 ± 5.6 ANOVA (p) ISO × Time 0.79 RT × Time 0.04 Interaction 0.48 Trunk fat mass (kg) Baseline 14.3 ± 4.4 12.3 ± 3.2 15.8 ± 5.4 14.8 ± 4.4 After 9 months 15.4 ± 5.3 12.9 ± 3.7 17.2 ± 6.1 16.7 ± 5.6 ANOVA (p) ISO × Time 0.96 RT × Time 0.04 Interaction 0.28 Whole-body fat mass (kg) Baseline 31.1 ± 10.3 24.8 ± 6.2 32.2 ± 9.7 31.3 ± 8.8 After 9 months 31.8 ± 10.0 25.6 ± 7.4 33.8 ± 10.7 32.8 ± 9.2 ANOVA (p) ISO × Time 0.94 RT × Time 0.12 Interaction 0.85 Body fat (%) Baseline 46.2 ± 4.9 43.0 ± 4.4 46.8 ± 6.7 46.5 ± 5.7 After 9 months 45.6 ± 5.1 43.8 ± 5.8 48.4 ± 6.8 48.0 ± 6.1 ANOVA (p) ISO × Time 0.25 RT × Time 0.02 Interaction 0.20 Table 3 Daidzein and genistein plasma levels at 9 months of intervention (mean ± SD).

1. Introduction Tuboovarian abscess (TOA), typically the end result of acute pelvic inflammatory disease (PID), is a condition characterized by a walled-off inflammatory mass in the pelvis. One-third to one-half of patients diagnosed with a TOA acknowledge a history of PID [1, 2]. PID and TOAs are polymicrobial infections of anaerobic and aerobic bacteria. While Neisseria gonorrhoeae and Chlamydia trachomatis are thought to facilitate the infection, they are rarely recovered from an abscess [3, 4]. The most commonly isolated organisms from TOAs are Escherichia coli and Bacteroides species [5]. TOA is a complication of PID in 15% of cases, and 33% of patients with PID requiring admission have a TOA [5]. Mortality associated with TOA has decreased dramatically over the last 50 years. However, the morbidity associated with TOA remains significant with complications including infertility, ectopic pregnancy, chronic pelvic pain, pelvic thrombophlebitis, and ovarian vein thrombosis [6]. While the majority of TOA respond to antibiotic therapy, in approximately 25% of cases surgery or drainage is indicated [7]. There is some evidence that TOA size is associated with need for intervention. Reed et al. in 1991 showed that 35% of abscesses 7 to 9 cm required surgery and nearly 60% of abscesses ≥10 cm required surgery [8]. The objective of our study was to assess the association of TOA size and need for prolonged hospitalization and surgical intervention. Our hypothesis was that TOA size would be associated with prolonged hospitalization and increased need for surgery or drainage.

and nearly 60% of abscesses ≥10 cm required surgery [8]. The objective of our study was to assess the association of TOA size and need for prolonged hospitalization and surgical intervention. Our hypothesis was that TOA size would be associated with prolonged hospitalization and increased need for surgery or drainage. 2. Methods After obtaining approval from the Washington University in St. Louis Human Research Protection Office, we performed a retrospective review of women diagnosed with TOA admitted to Barnes-Jewish Hospital in Saint Louis, Missouri, from January 1, 1999 until December 31, 2005. Potential subjects were selected by the following ICD-9 codes: PID (614.9) and TOA (614.2). Three hundred seventy-three charts were then reviewed to confirm eligibility. Inclusion criteria included (1) presence of an inflammatory mass in the pelvis, (2) age greater than 17 years, and (3) admission to Barnes-Jewish Hospital. We excluded charts of women who did not have an inflammatory mass identified. Data were abstracted for demographics, diagnostic testing, imaging, treatment, and clinical course.

teria included (1) presence of an inflammatory mass in the pelvis, (2) age greater than 17 years, and (3) admission to Barnes-Jewish Hospital. We excluded charts of women who did not have an inflammatory mass identified. Data were abstracted for demographics, diagnostic testing, imaging, treatment, and clinical course. Imaging was obtained in almost all cases and included ultrasound, CT, or MRI. In just over half of cases more than one imaging modality was used. One abscess that was not imaged preoperatively was identified at laparotomy. Fever was defined as a temperature greater than 38.2°C. Complications were defined as factors that could result in prolonged hospitalization; specifically, complications included intensive care unit (ICU) admission, bowel or bladder injury, blood transfusion, blood loss >1000 mL, sepsis or bacteremia, ileus, readmission, and death. Data were abstracted by three trained abstractors. We created two abscess groups based on maximal diameter of the abscess: ≤8 cm and >8 cm. If bilateral abscesses were noted, we used the larger of the two to quantify the size of the abscess. Women with large and small abscesses were compared by duration of hospitalization, surgical intervention, and complication rate. The initial antibiotic regimens were categorized to include at least (1) cefoxitin and doxycycline, (2) ciprofloxacin and metronidazole, (3) gentamicin and clindamycin, and (4) others.

scess. Women with large and small abscesses were compared by duration of hospitalization, surgical intervention, and complication rate. The initial antibiotic regimens were categorized to include at least (1) cefoxitin and doxycycline, (2) ciprofloxacin and metronidazole, (3) gentamicin and clindamycin, and (4) others. Descriptive statistics were calculated using mean, standard deviation, and percentages. Comparisons and associations were made using Chi-square or Fisher's exact test for categorical variables. T-tests were used for continuous variables; nonparametric tests were used for data that were not normally distributed. Estimates of relative risk were estimated using Poisson regression with robust error variance. This approach results in an unbiased estimate when a binary outcome is common (>10%) [9]. The power calculations with an α error of 0.05 and β error of 0.20 found that the number of TOAs needed to evaluate was 121 assuming that 15% of cases would have TOA < 8 cm and 30% would have TOA > 8 cm. To achieve our target of 121 cases of TOA, we estimated that 370 charts would need to be reviewed given the frequent misdiagnosis of PID. Statistical analyses were conducted using SAS v. 9.2 (SAS Corporation, Cary, NC).

to evaluate was 121 assuming that 15% of cases would have TOA < 8 cm and 30% would have TOA > 8 cm. To achieve our target of 121 cases of TOA, we estimated that 370 charts would need to be reviewed given the frequent misdiagnosis of PID. Statistical analyses were conducted using SAS v. 9.2 (SAS Corporation, Cary, NC). 3. Results Of 373 charts reviewed, 135 patients were identified as having a TOA while 218 patients were excluded based on the lack of presence of an abscess. Demographic and reproductive characteristics are provided in Table 1. The average age was 35.2 years (SD 10.0), and the mean gravidity and parity were 1.96 (SD 1.76) and 1.52 (SD 1.64), respectively. More than two-thirds were black, and 10% gave a history of having had a bilateral tubal ligation.

resence of an abscess. Demographic and reproductive characteristics are provided in Table 1. The average age was 35.2 years (SD 10.0), and the mean gravidity and parity were 1.96 (SD 1.76) and 1.52 (SD 1.64), respectively. More than two-thirds were black, and 10% gave a history of having had a bilateral tubal ligation. Clinical characteristics of the study population are noted in Table 2. The most common reason for presentation was pain (94%) with a median duration of 3 days. Only 27% of patients had a fever on admission. The mean white blood count (WBC) on admission was 14.2 (SD 5.9) with 27% having WBC less than 10,000. Few subjects had positive tests for N. gonorrhoeae (12%) and C. trachomatis (14%). Imaging was obtained in most cases with patients frequently having either a CT (79%) or an ultrasound (70%). MRI was used in only 3 cases (2%). In the only case in which imaging was not obtained, the TOA was diagnosed during her cesarean section and bilateral tubal ligation and confirmed on pathology. Abscess size ranged from 1.5 cm to 16 cm with a mean of 6.7 cm. Most (51%) abscesses were between 5 and 8 cm. Two-thirds of abscesses were unilateral. Of patients with bilateral abscesses, the maximal diameter of the smaller abscess was 8 cm.

her cesarean section and bilateral tubal ligation and confirmed on pathology. Abscess size ranged from 1.5 cm to 16 cm with a mean of 6.7 cm. Most (51%) abscesses were between 5 and 8 cm. Two-thirds of abscesses were unilateral. Of patients with bilateral abscesses, the maximal diameter of the smaller abscess was 8 cm. Comparison was made between those patients with abscesses ≤8 cm or >8 cm (Table 3). Of the 135 patients with a TOA, 112 (83%) had an abscess ≤8 cm and 23 (17%) had an abscess >8 cm. Patients with larger abscesses averaged longer duration of hospitalization, more complications, longer duration febrile, and an increased need for surgery or drainage when compared with those patients with smaller abscesses. Every 1 cm increase in abscess size is associated with an increase in hospitalization by 0.4 days (P =  .001). The mean abscess size for those requiring drainage or surgery was 7.7 cm versus 6.3 cm (P =  .02) for those with successful conservative management. Among the larger abscess group, 7 hysterectomies (30%) were performed with 8 (35%) patients receiving a simple adnexectomy. Patients with abscesses ≤8 cm had fewer complications (9%) than did patients with abscesses >8 cm (35%, P <  .01). Patients often had more than one complication; 18 patients developed 35 complications. The most common complication was blood transfusion which occurred in 14 patients (10%). There were no deaths. Patients with larger abscesses were febrile for longer versus those with smaller abscesses (1.35 versus 1.10 days, P =  .65).

tients often had more than one complication; 18 patients developed 35 complications. The most common complication was blood transfusion which occurred in 14 patients (10%). There were no deaths. Patients with larger abscesses were febrile for longer versus those with smaller abscesses (1.35 versus 1.10 days, P =  .65). Patients were then evaluated by initial antibiotic regimen (Table 4). Fifty-seven patients were initially placed on regimens containing cefoxitin and doxycycline, and of these, 91% were successfully managed conservatively. Patients placed initially on regimens including cefoxitin and doxycycline were less likely to require surgery or drainage. This is in contrast to other antibiotic regimens which had approximate equal need for surgery or drainage in the smaller and larger abscess groups. A regression model was used to assess the need for surgical intervention or drainage controlling for other confounders. The relative risk for cefoxitin/doxycycline regimen was 0.21 (95%, CI 0.09–0.48) adjusting for age and total number of antibiotics. This finding persisted when fever, abscess >8 cm, and days hospitalized were included in the model.

assess the need for surgical intervention or drainage controlling for other confounders. The relative risk for cefoxitin/doxycycline regimen was 0.21 (95%, CI 0.09–0.48) adjusting for age and total number of antibiotics. This finding persisted when fever, abscess >8 cm, and days hospitalized were included in the model. 4. Discussion We found that TOA size is associated with important outcomes including more complications and longer duration of hospitalization as well as an increased need for surgery or drainage when compared with those patients with smaller abscesses. Reed et al. also found that increasing abscess size is associated with increasing need for operative management [8]. We found a 43% failure rate for abscesses >8 cm while Reed showed a 35% failure rate for abscesses 7 to 9 cm and nearly 60% failure rate for abscesses ≥10 cm. In contrast, Gjelland et al. noted that treatment success was not affected by the size of the abscess or the presence of bilateral abscesses in a sample restricted to abscesses with maximal diameter of at least 3 cm. The authors of this study recommended transvaginal drainage as a first-line procedure [10].

ses ≥10 cm. In contrast, Gjelland et al. noted that treatment success was not affected by the size of the abscess or the presence of bilateral abscesses in a sample restricted to abscesses with maximal diameter of at least 3 cm. The authors of this study recommended transvaginal drainage as a first-line procedure [10]. Originally, treatment of TOA was thought to require bilateral oophorectomy and hysterectomy. Medical management with broad spectrum antibiotics is now generally considered as the initial management for unruptured TOAs [11]. However, optimal treatment of TOA remains unclear. The 2006 Center for Disease Control and Prevention Sexually Transmitted Diseases Treatment Guidelines recommends inpatient intravenous antibiotics for at least 24 hours. No specific inpatient antibiotic regimen is suggested. Upon discontinuation of parenteral therapy, the CDC recommends that clindamycin or metronidazole be used with doxycycline for a total of 14 days of treatment [12]. Sweet recommends broad spectrum antibiotic coverage including coverage of Gram-negative anaerobes for at least 24 hours [11]. Oral therapy and hospital discharge are acceptable when the patient has had a favorable clinical response to therapy. Surgery or drainage should be considered when the patient displays a failure to respond in 48 to 72 hours [9].

tibiotic coverage including coverage of Gram-negative anaerobes for at least 24 hours [11]. Oral therapy and hospital discharge are acceptable when the patient has had a favorable clinical response to therapy. Surgery or drainage should be considered when the patient displays a failure to respond in 48 to 72 hours [9]. Drainage procedures have been increasingly evaluated in the management of TOA. Most recommendations for the management of TOA advise drainage after conservative management has failed [9]. However, several studies show that early drainage of all TOAs is safe and improves outcomes [10, 13] and may be appropriate as a primary therapy [14]. Drainage may be accomplished by CT or ultrasound through the abdomen, vagina, rectum, or transgluteal. Gjelland et al. observed in a retrospective review that, in 302 cases of transvaginal drainage of TOA combined with antibiotics, 93% patients avoided surgery with no major procedure-related complications [10]. Perez-Medina randomized 40 women to traditional conservative management versus antibiotic therapy plus early transvaginal drainage in unruptured TOAs and found that 90% had successful early response versus 65% in the control group (P-value ≤  .05) [15]. Moreover, they found that early transvaginal drainage was associated with significantly shorter hospital stays and decreased morbidity. Our study observed only 11 drainage procedures that were split relatively equally between the 2 abscess groups and surgical drainage was not a primary treatment. Thus, we could not assess the effect of drainage on length of hospitalization.

s associated with significantly shorter hospital stays and decreased morbidity. Our study observed only 11 drainage procedures that were split relatively equally between the 2 abscess groups and surgical drainage was not a primary treatment. Thus, we could not assess the effect of drainage on length of hospitalization. Regarding antibiotic regimens for TOA, Wiesenfeld and Sweet summarized the response rates of TOAs to medical therapy and showed a 72% success rate in patients treated with clindamycin and aminoglycoside (n = 101) versus 82% success rate in patients treated with cefoxitin or cefotetan and doxycycline (n = 62) [6]. However, this finding has not been consistently replicated in the literature; another study found this regimen to be less successful [15]. Our data are consistent with the review by Wiesenfeld and Sweet—we found that 9% of patients who were initially treated with cefoxitin and doxycycline required surgical intervention or drainage. These findings should be interpreted with caution, as many of the patients started on the cefoxitin and doxycycline regimen received additional antibiotics.

e review by Wiesenfeld and Sweet—we found that 9% of patients who were initially treated with cefoxitin and doxycycline required surgical intervention or drainage. These findings should be interpreted with caution, as many of the patients started on the cefoxitin and doxycycline regimen received additional antibiotics. The strengths of our study are that we have a large number of TOAs despite its being an infrequent outcome. Management of TOA was specifically evaluated with emphasis on the influence of abscess size as a prognostic marker of outcome. The limitations of our study include a relatively small number of large TOAs and the inability to completely control confounding variables such as disease severity. In addition, our study design is unable to completely control confounding factors that may influence the apparent success rate of cefoxitin and doxycycline. Randomized trials are needed to provide sound evidence regarding appropriate antibiotic therapy for TOAs and whether routine drainage of TOAs can decrease prolonged hospitalizations and improve reproductive outcomes.

control confounding factors that may influence the apparent success rate of cefoxitin and doxycycline. Randomized trials are needed to provide sound evidence regarding appropriate antibiotic therapy for TOAs and whether routine drainage of TOAs can decrease prolonged hospitalizations and improve reproductive outcomes. Acknowledgments This research was supported in part by Grant no. K24 HD01298 (Midcareer Investigator Award in Women's Health Research (Peipert)), by Clinical and Translational Science Awards (UL1RR024992), and by training grants (K12RR023249 and KL2RR024994, Allsworth) from the National Center for Research Resources (NCRR), a component of the National Institutes of Health (NIH) and NIH Roadmap for Medical Research. Its contents are solely the responsibility of the authors and do not necessarily represent the official view of NCRR or NIH. Table 1 History and sociodemographic characteristics of study population. Sociodemographic characteristics N (%) Age 17–19 years 10 (7%) 20–29 years 30 (22%) 30–39 years 51 (38%) 40 years or older 44 (33%) Race/ethnicity White 37 (27%) Black 92 (68%) Hispanic/other 6 (4%) Smoking history Yes 64 (47%) No 45 (33%) Unknown 20 (19%) Reproductive history Gravidity 0 28 (21%) 1-2 65 (48%) 3 or more 39 (29%) Missing 3 (2%) Parity 0 44 (33%) 1-2 54 (40%) 3 or more 32 (24%) Missing 5 (4%) History of GC/CT 38 (28%) Any contraceptive use 27 (20%) Condom use 10 (7%) Tubal ligation 14 (10%) Table 2 Clinical characteristics by presence of TOA. TOA Clinical Characteristics Positive N  (%) Negative (or not documented) N (%)

Figure 3 Example of preprocessed EHG signal (upper plot) before (left) and after (right) nifedipine treatment. The corresponding power spectral densities before (left) and after (right) nifedipine treatment are depicted in the bottom part of the figure with an indication of the peak frequency fP. Table 1 Study population characteristics. Pt Age GA1 at first EHG recording Pregnancy BMI2 Cervical dilatation (cm) Cervical length (cm) PPROM3 UTI4 Nifedipine-recording- interval (hours) Delivery < 1wk 1 29 24 + 1 Singleton 22.3 1 3.1 No No 4 No 2 34 28 + 0 Singleton 27.8 1 2.3 No No 9 No 3 32 25 + 4 Singleton 28.0 0 1.6 No No 10 No 4 29 30 + 4 Twin 22.5 1 1.9 No Yes 8 Yes 5 29 31 + 2 Twin 21.6 1 2.0 Yes Yes 5 Yes 6 28 29 + 5 Twin 27.3 1 2.2 No Yes 0 No 7 34 25 + 6 Twin 23.4 2 2.6 Yes Yes 0 No 8 34 31 + 1 Singleton 25.1 1 ntb No Yes 0 Yes 1 = gestational age 2 = body mas index 3 = preterm prelabor rupture of membranes 4 = urinary tract infection. Table 2 Shift of PSD peak frequency. Group 11 Before nifedipine (Subgroup  1.1) 15 minutes after nifedipine (subgroup 1.2) P-value PDS peak frequency (Hz) 0.367 ± 0.061 0.340 ± 0.040 0.415 ± 0.046 0.375 ± 0.051 0.424 ± 0.079 0.405 ± 0.021 Mean 0.402 ± 0.025 0.  373 ± 0.026 .043* Group 22 Within 24 hours after starting nifedipine (subgroup 2.1) 1 day after finishing nifedipine (subgroup 2.2) P-value PDS peak frequency (Hz) 0.390 ± 0.035 0.371 ± 0.058 0.436 ± 0.046 0.403 ± 0.074 0.434 ± 0.047 0.403 ± 0.041 Mean 0.42 ± 0.021 0.392 ± 0.015 .024*

Gravidity 0 28 (21%) 1-2 65 (48%) 3 or more 39 (29%) Missing 3 (2%) Parity 0 44 (33%) 1-2 54 (40%) 3 or more 32 (24%) Missing 5 (4%) History of GC/CT 38 (28%) Any contraceptive use 27 (20%) Condom use 10 (7%) Tubal ligation 14 (10%) Table 2 Clinical characteristics by presence of TOA. TOA Clinical Characteristics Positive N  (%) Negative (or not documented) N (%) Reason presenting for care Pain 127 (94%) 8 (6%) Median duration of pain 3 days — Abnormal bleeding 22 (16%) 113 (84%) Abnormal discharge 28 (21%) 107 (79%) Information at admission Fever 37 (27%) 98 (73%) Pain 109 (81%) 26 (19%) Vaginal discharge 42 (31%) 93 (69%) Rebound 29 (21%) 106 (79%) Guarding 50 (37%) 85 (63%) Uterine tenderness 36 (27%) 99 (73%) Adnexal tenderness 69 (51%) 66 (49%) Cervical motion tenderness 65 (48%) 70 (52%) Cervical culture 91 (67%) 44 (33%) GC (n = 91) 11 (12%) 77 (85%) CT (n = 91) 13 (14%) 74 (81%) CT performed 106 (79%) 29 (21%) US performed 94 (70%) 41 (30%) MRI performed 3 (2%) 132 (98%) WBC, mean (SD)** 14.2 (5.9) — WBC <10 37 (27%) — 10 to <15 46 (34%) — 15+ 52 (39%) — Unilateral abscess† 89 (66%) — Largest abscess size (cm), mean  (SD) (Note: range = 1.5–16 cm) 6.7 (2.7) — Abscess size 0 to 4 cm 43 (32%) — 5 to 8 cm 69 (51%) — >8 cm 23 (17%) — *87 patients (64%) had no information on HIV in medical record **WBC not documented for 5 patients †Information missing for 21 patients. Table 3 Outcomes and initial antibiotic regimen by abscess size. All patients (N = 135) Abscess ≤8 cm (N = 112) Abscess >8 cm (N = 23) P-value # days hospitalized mean (SD) 4.6 (3.6) 4.4 (3.4) 5.7 (4.2) .09

Information at admission Fever 37 (27%) 98 (73%) Pain 109 (81%) 26 (19%) Vaginal discharge 42 (31%) 93 (69%) Rebound 29 (21%) 106 (79%) Guarding 50 (37%) 85 (63%) Uterine tenderness 36 (27%) 99 (73%) Adnexal tenderness 69 (51%) 66 (49%) Cervical motion tenderness 65 (48%) 70 (52%) Cervical culture 91 (67%) 44 (33%) GC (n = 91) 11 (12%) 77 (85%) CT (n = 91) 13 (14%) 74 (81%) CT performed 106 (79%) 29 (21%) US performed 94 (70%) 41 (30%) MRI performed 3 (2%) 132 (98%) WBC, mean (SD)** 14.2 (5.9) — WBC <10 37 (27%) — 10 to <15 46 (34%) — 15+ 52 (39%) — Unilateral abscess† 89 (66%) — Largest abscess size (cm), mean  (SD) (Note: range = 1.5–16 cm) 6.7 (2.7) — Abscess size 0 to 4 cm 43 (32%) — 5 to 8 cm 69 (51%) — >8 cm 23 (17%) — *87 patients (64%) had no information on HIV in medical record **WBC not documented for 5 patients †Information missing for 21 patients. Table 3 Outcomes and initial antibiotic regimen by abscess size. All patients (N = 135) Abscess ≤8 cm (N = 112) Abscess >8 cm (N = 23) P-value # days hospitalized mean (SD) 4.6 (3.6) 4.4 (3.4) 5.7 (4.2) .09 # days febrile (>38.2) 1.1 (1.8) 1.10 (1.6) 1.35 (2.5) .65 N (%) Requiring drainage/surgery 42 (31%) 32 (29%) 10 (43%) .16 Drainage* 11 (8%) 9 (8%) 2 (9%) .31 Surgery 34 (25%) 26 (23%) 8 (35%) .24 Number of antibiotics used, mean (SD) 2.8 (1.2) 2.7 (1.2) 3.1 (1.3) .20 Antibiotic regimen* Cefoxitin and doxycycline 57 (43%) 50 (45%) 7 (30%) .14 Ciprofloxacin and Flagyl 11 (8%) 7 (6%) 4 (17%) Gentamicin and clindamycin 18 (14%) 13 (12%) 5 (22%) Other 46 (35%) 39 (35%) 7 (30%)

Requiring drainage/surgery 42 (31%) 32 (29%) 10 (43%) .16 Drainage* 11 (8%) 9 (8%) 2 (9%) .31 Surgery 34 (25%) 26 (23%) 8 (35%) .24 Number of antibiotics used, mean (SD) 2.8 (1.2) 2.7 (1.2) 3.1 (1.3) .20 Antibiotic regimen* Cefoxitin and doxycycline 57 (43%) 50 (45%) 7 (30%) .14 Ciprofloxacin and Flagyl 11 (8%) 7 (6%) 4 (17%) Gentamicin and clindamycin 18 (14%) 13 (12%) 5 (22%) Other 46 (35%) 39 (35%) 7 (30%) Any complication 18 (13%) 10 (9%) 8 (35%) <.01 *P-value estimated from Fisher's exact test. Table 4 Outcomes of surgery and drainage by initial antibiotic regimen. No surgery/ drainage Surgery/ drainage P-value Cefoxitin/Doxycycline 52 (91%) 5 (9%) <.0001 Ciprofloxacin/Flagyl 5 (45%) 5 (55%) Gentamicin/Clindamycin 9 (50%) 9 (50%) Other 27 (59%) 19 (41%)

1. Introduction Postaxial longitudinal defect is one of the most common congenital limb reduction defects. This entity includes a large spectrum of abnormalities that may range from severe hypoplasia to complete absence of the fibula and the 5th rays. It is a rare disorder, with an estimated prevalence of 5.7 to 20 cases per 1 million births [1]. Up to now, only few cases of prenatal diagnosis of isolated longitudinal deficiency of the fibula were reported [1–4].

abnormalities that may range from severe hypoplasia to complete absence of the fibula and the 5th rays. It is a rare disorder, with an estimated prevalence of 5.7 to 20 cases per 1 million births [1]. Up to now, only few cases of prenatal diagnosis of isolated longitudinal deficiency of the fibula were reported [1–4]. 2. Case Report A 38-year-old healthy primigravida, with no familiar history of limb defects or exposure to teratogenic drugs, was referred to our ultrasound unit at 22 weeks gestation after the absence of the right fibula was diagnosed during the second trimester ultrasound examination. This diagnosis was confirmed in our department. Further assessment of the ipsilateral lower limb detected a discrete femur shortening (35 mm versus 37 mm of left femur, discrepancy of 5.4%), anteromedial bowing and tibial shortening (27.2 mm versus 33.6 mm of left tibia, discrepancy of 19%), see Figure 1, foot equinovalgus, and absence of the fourth and fifth foot rays and digits. All other long bones (humeri, ulnae, and radii) were symmetric and appropriated in length and configuration for gestational age, as were the hands. No other anomalies were detected, namely, craniosynostosis, omphalocele, renal displasia, neural tube defects, thoracoabdominal schisis, or facial dysmorphies. Amniocentesis revealed a normal female karyotype (46, XX). Fetal echocardiography was normal. Follow-up ultrasound examinations were carried out periodically until birth (Figure 2). Tibial discrepancy increased slightly with a difference of 13 mm (23.6%) at 34 weeks gestation.

oabdominal schisis, or facial dysmorphies. Amniocentesis revealed a normal female karyotype (46, XX). Fetal echocardiography was normal. Follow-up ultrasound examinations were carried out periodically until birth (Figure 2). Tibial discrepancy increased slightly with a difference of 13 mm (23.6%) at 34 weeks gestation. No other antenatal problems occurred, and at 41 weeks gestation a cesarean section was performed due to cephalopelvic disproportion. A 3430 g female newborn was delivered with Apgar scores of 9 and 10 at 1 and 5 minutes, respectively. Neonatal examination (Figure 3) and X-Ray (Figure 4) confirmed the anomalies. Right tibia appeared shortened and bowed anteriomedially, and an ipsilateral equinovalgus foot was present with the absence of the fourth and fifth rays. In spite of limb-length shortening and alterations in limb alignment and stability, normal active mobility of both limbs was observed. No other congenital abnormalities were detected. The newborn was discharged at the fourth postpartum day. The baby was followed up by orthopaedic pediatrics, and there were no perinatal complications. At ten months the infant could sit but not crawl, and she started to walk at the fifteenth month, though she has a normal intellectual development for her age. Orthopaedic surveillance will be maintained and treatment will only be applied when and if needed.

paedic pediatrics, and there were no perinatal complications. At ten months the infant could sit but not crawl, and she started to walk at the fifteenth month, though she has a normal intellectual development for her age. Orthopaedic surveillance will be maintained and treatment will only be applied when and if needed. 3. Discussion Some authors [5, 6] consider postaxial longitudinal defect the most common lower limb congenital deficiency although there are published data [7, 8] that refer to terminal transverse defects as being the most common one. The inexistence of a consensual data may reflect the rarity of limb malformations, differences between populations, and the lack of uniform classification system of limb reduction defects (LRDs). There are many classification systems for LRD published in the literature, such as the Frantz and O' Rahilly classification [9] and the Achterman and Kalamchi classification [10], which is based on clinical and radiographic findings. European Surveillance of Congenital Anomalies (EUROCAT) proposed a more consistent classification, that was reviewed in 2004. Using the EUROCAT classification, our case report can be classified as a postaxial (fibula) longitudinal defect. This entity includes a large spectrum of abnormalities that may range from severe hypoplasia to the complete absence of fibula and the 5th rays. No sex differences were reported, although some studies referred that males are affected twice as often as females. Moreover, unilateral involvement occurred in two thirds of cases with the right side being more frequently affected [5, 6, 10, 11].

om severe hypoplasia to the complete absence of fibula and the 5th rays. No sex differences were reported, although some studies referred that males are affected twice as often as females. Moreover, unilateral involvement occurred in two thirds of cases with the right side being more frequently affected [5, 6, 10, 11]. There is a consensus about the critical embrionary period of limb development, which is between 4 and 8 weeks gestation, but there is no precise understanding of the etiology of this unusual disorder. The majority of the cases appeared sporadically or as isolated events although it may be part of a malformative syndrome related with genetic conditions, teratogenic insults, or vascular disruption [11, 12]. A familiar history of skeletal anomalies may be found by Calzolari et al. in 7.2% of all limb reduction defects (LRDs) [7]. Femur-fibula-ulna complex and proximal femoral focal deficiency should be ruled out during the ultrasound examination, as both can show fibular deficiency. Thus, as soon as some fibular deficiency is found, meticulous search for additional long bones and hands abnormalities should be performed. Since there are only few cases associated with other nonskeletal malformations (0.8%) [11], such as cranial, facial, gastrointestinal, urogenital, cardiac, lung, diaphragmatic, and neural tube abnormalities, attention should also be paid to their ultrasound screening. In our case, these malformations were excluded.

ed. Since there are only few cases associated with other nonskeletal malformations (0.8%) [11], such as cranial, facial, gastrointestinal, urogenital, cardiac, lung, diaphragmatic, and neural tube abnormalities, attention should also be paid to their ultrasound screening. In our case, these malformations were excluded. Fibular deficiency is generally accompanied by a constellation of other ipsilateral lower limb malformations, with a wide variety of manifestations, such as femoral hypoplasia, tibial a/hypoplasia or anterior bowing, knee and ankle deformities and abnormal positions (equinovalgus), and, finally, toe deficiency, absence of one or more lateral rays [10]. The degree of tibial shortening increases as the fibular deficiency becomes more marked, while femoral shortening did not correlate with the fibular deficiency [10]. The correct evaluation of the severity of foot abnormalities, which appeared to be correlated with the severity of fibular deficiency, is important for assessment of orthopaedic impact. The diagnosis of lower limb malformations used to be performed, almost always, after birth. Nowadays, it can be identified if a complete, detailed, and skilled prenatal ultrasound examination of lower limbs is systematically done. This highlights ultrasound accuracy for the diagnosis of the postaxial longitudinal defects. We should point out that it may be difficult to determine whether the bone of distal lower extremity is the fibula or the tibia; tibia articulates with the femur.

al ultrasound examination of lower limbs is systematically done. This highlights ultrasound accuracy for the diagnosis of the postaxial longitudinal defects. We should point out that it may be difficult to determine whether the bone of distal lower extremity is the fibula or the tibia; tibia articulates with the femur. Bone echodensities seen before pregnancy correlate well with neonatal X-Ray densities, allowing a correct classification of this entity, undoubtedly necessary to orthopaedic assessment. That was clearly shown by the excellent agreement observed in our case between prenatal ultrasound and neonatal radiographic results. This condition, although not fatal, has a poor prognosis that depends on the degree of limb deformity and on the chosen treatment. Treatment options should be individualized (tibial lengthening, amputation, epiphysiodesis, or prosthetic rehabilitation), and management should be based on careful interpretation of functional, psychological, and cosmetic needs. To conclude, accurate prenatal ultrasound diagnosis of postaxial longitudinal defect may be performed. The correct approach and advice would be done only after birth, when the impact of the anomaly can be completely evaluated. A multidisciplinary approach including obstetricians, geneticists, neonatologists, and pediatric orthopaedists should always discuss the implications of this situation with the parents wand counsel thoroughly concerning the management, treatment options, and prognosis.

impact of the anomaly can be completely evaluated. A multidisciplinary approach including obstetricians, geneticists, neonatologists, and pediatric orthopaedists should always discuss the implications of this situation with the parents wand counsel thoroughly concerning the management, treatment options, and prognosis. Figure 1 Ultrasound images showing absence of right fibula, bowing of right tibia (a), and tibial discrepancy (right tibia: 27.2 mm and left tibia: 33.6 mm) at 23 weeks gestation (b). Figure 2 Ultrasound images showing absence of right fibula (a) and absence of the fourth and fifth right foot rays and digits (b) at 31 weeks gestation. Figure 3 Neonatal pediatric examination at the 1st postnatal day: right limb shortening and bowing (a) and right foot with three digits (b). Figure 4 Radiograph showing absence of right fibula and shortening and bowing of right tibia at the 7th postnatal day.

Group 11 Before nifedipine (Subgroup  1.1) 15 minutes after nifedipine (subgroup 1.2) P-value PDS peak frequency (Hz) 0.367 ± 0.061 0.340 ± 0.040 0.415 ± 0.046 0.375 ± 0.051 0.424 ± 0.079 0.405 ± 0.021 Mean 0.402 ± 0.025 0.  373 ± 0.026 .043* Group 22 Within 24 hours after starting nifedipine (subgroup 2.1) 1 day after finishing nifedipine (subgroup 2.2) P-value PDS peak frequency (Hz) 0.390 ± 0.035 0.371 ± 0.058 0.436 ± 0.046 0.403 ± 0.074 0.434 ± 0.047 0.403 ± 0.041 Mean 0.42 ± 0.021 0.392 ± 0.015 .024* Group 33 Patients delivering after 1 week after start nifedipine (subgroup 3.1) Patients delivering within 1 week after start nifedipine (subgroup 3.2) P-value PDS peak frequency (Hz) 0.340 ± 0.040 0.407 ± 0.065 0.375 ± 0.051 0.380 ± 0.055 0.405 ± 0.021 0.434 ± 0.047 0.436 ± 0.046 0.390 ± 0.035 Mean 0.389 ± 0.031 0.407 ± 0.022 .458** 1 = patient 1–3 (Table 1) 2 = patient 4–6 (Table 1) 3 = patient 1–8 (Table 1) *calculation by 2-tailed paired student t test **calculation by 2-tailed unpaired student t-test.

1. Introduction Preterm birth is the leading cause of perinatal mortality and morbidity and accounts for approximately half of preterm births [1]. Although the management of threatened preterm labor by tocolytic therapy can prolong gestation [2], effectiveness of tocolytic agents depends on early introduction of therapy [1]. Some risk factors for preterm delivery have been identified, but available prediction methods exhibit poor diagnostic value for detecting preterm labor at an early stage [3–5]. When evaluating women with preterm contractions, the clinicians' daily dilemma is to differentiate between physiological uterine activity and contractions leading to preterm delivery. Monitoring uterine contractions may provide important prognostic information during pregnancy and parturition. No method is currently available for non-invasive monitoring and quantification of uterine contractions [3–5]. It is well established that uterine contractions are the result of the electrical activity propagation through the uterine muscle cells [6, 7]. The electrohysterogram (EHG) is a direct measure of the myometrium electrical activity and it can be measured noninvasively by electrodes positioned on the maternal abdomen [8–10]. The EHG is characterized by intermittent “bursts” of action potentials and each burst corresponds to a mechanical contraction of the uterus. The uterus is quiescent during pregnancy. However, as the delivery time approaches, the uterine electrical activity becomes increasingly synchronous [11].

maternal abdomen [8–10]. The EHG is characterized by intermittent “bursts” of action potentials and each burst corresponds to a mechanical contraction of the uterus. The uterus is quiescent during pregnancy. However, as the delivery time approaches, the uterine electrical activity becomes increasingly synchronous [11]. Some typical changes have been observed in the EHG properties during preterm and term delivery [12–15]. In particular, the shifting of the EHG signal energy to higher frequency as delivery approaches is the most experienced phenomenon observed in the literature [11–14, 16]. In particular, the peak frequency, that is, the frequency corresponding to the maximum of the EHG signal power spectral density (PSD), is the EHG parameter that is most commonly tested [17]. As uterine preterm contractions are the most frequently recognized precursor of preterm delivery, inhibiting contractions (tocolysis) has been the focus of therapeutic approaches for prevention of preterm birth.

Some typical changes have been observed in the EHG properties during preterm and term delivery [12–15]. In particular, the shifting of the EHG signal energy to higher frequency as delivery approaches is the most experienced phenomenon observed in the literature [11–14, 16]. In particular, the peak frequency, that is, the frequency corresponding to the maximum of the EHG signal power spectral density (PSD), is the EHG parameter that is most commonly tested [17]. As uterine preterm contractions are the most frequently recognized precursor of preterm delivery, inhibiting contractions (tocolysis) has been the focus of therapeutic approaches for prevention of preterm birth. As the delivery time approaches, the number of cell gap junctions and voltage-dependant calcium channels increases and form an electrical syncytium leading to an increased uterine activity [18]. Currently, nifedipine, which directly blocks the calcium channel, would be in fact the first choice tocolytic agent to postpone delivery [2, 19]. Nifedipine is reported to be highly effective [2]; however, no placebo-controlled trials were performed and nifedipine is not registered for use as a tocolytic agent [20]. In some rare cases, serious maternal side effects are reported due to overdosage of nifedipine [21]. Therefore, more evidence is needed about the efficacy and safety of this drug for the inhibition of labor.

no placebo-controlled trials were performed and nifedipine is not registered for use as a tocolytic agent [20]. In some rare cases, serious maternal side effects are reported due to overdosage of nifedipine [21]. Therefore, more evidence is needed about the efficacy and safety of this drug for the inhibition of labor. The aims of this study are (1) providing further insight into the use of the electrohysterogram (EHG) for monitoring uterine electrical contractions in preterm labor; (2) evaluating the nifedipine-induced changes in PSD peak frequency in women with preterm contractions. 2. Material and Methods Data were collected at the Máxima Medical Center, a tertiary care teaching hospital. The study was approved by the medical ethical committee of the hospital. All participating women signed an informed consent. 2.1. EHG Signal Recording The EHG was recorded by eight disposable contact Ag-AgCl electrodes (2 cm interelectrode distance) placed on the abdomen as shown in Figure 1. Prior to electrode attachment the skin was prepared by gentle rubbing with fine abrasive paper. A reference electrode was placed on the left hip. In order to obtain an efficient rejection of the electromagnetic interference, a driven-right-leg ground electrode and actively shielded cables were employed. The EHG signals were recorded and digitized at 1000 Hz, 20 bit resolution, by an M-PAQ amplifier (Maastricht Instruments Ltd., the Netherlands); a 16-channel system for physiological measurements with programmable gain and sampling frequency.

driven-right-leg ground electrode and actively shielded cables were employed. The EHG signals were recorded and digitized at 1000 Hz, 20 bit resolution, by an M-PAQ amplifier (Maastricht Instruments Ltd., the Netherlands); a 16-channel system for physiological measurements with programmable gain and sampling frequency. All included patients received nifedipine. For each patient, the EHG was recorded twice, before and after giving nifedipine. The period of each recording varied between 20 and 100 minutes per patient. A conventional tocogram was simultaneously recorded for medical prescription. The use of the tocogram was not specifically required for the study, but was employed for confirming the EHG contraction detection. 2.2. EHG Signal Processing In order to improve the signal quality, seven bipolar signals were obtained by subtraction of the signal recorded at electrode five (Figure 1). The recorded signals where then visually inspected to assess the signal quality. If judged of good quality, all seven channels were used for further processing.

gnal Processing In order to improve the signal quality, seven bipolar signals were obtained by subtraction of the signal recorded at electrode five (Figure 1). The recorded signals where then visually inspected to assess the signal quality. If judged of good quality, all seven channels were used for further processing. Additional noise due to respiration, maternal electrocardiogram, and skeletal electromyogram caused by abdominal muscle contraction was removed by band-pass filtering between 0.34 Hz and 1 Hz [16]. To this end, a four-order band-pass Butterworth filter was employed. The signal was then downsampled up to 20 Hz. The down-sampling was made possible by the low frequency content of the signal and reduced significantly the computational time. Contraction bursts of EHG activity were detected on the EHG signal by the method described in [22] and validated by comparison with the simultaneously recorded tocogram. Similarly to many previous studies on the prediction of pre-term labor by electrohysterography, the power spectral density (PSD) was calculated on each bipolar signal by computing the Fast Fourier Transform (FFT) [8, 12, 16, 18]. Only signal segments within the detected EHG electrical contraction burst were used and no information derived from the quiescent period between contractions was analyzed.

hysterography, the power spectral density (PSD) was calculated on each bipolar signal by computing the Fast Fourier Transform (FFT) [8, 12, 16, 18]. Only signal segments within the detected EHG electrical contraction burst were used and no information derived from the quiescent period between contractions was analyzed. The PSD was separately calculated for three contraction bursts during the first period of measurement, before treatment, and three contraction bursts during the second period of measurement, after treatment. The frequency at which the PSD was maximum, that is, the peak frequency (fP), was singled-out. For each channel, the values of fP were averaged over the three recorded bursts of electrical activity. Average and standard deviation of resulting fP for the different periods of recording were then analyzed and compared. 2.3. Study Population The study protocol has been carried out in addition to the general used procedures for threatened preterm labor. Women admitted to the hospital for spontaneous preterm contractions, at risk for extreme preterm delivery, that is, gestational age between 24th and 32nd weeks, were included in the study. Exclusion criteria were age under 18 years, body mass index above 30, cervical dilatation over 3 cm. Additionally were excluded from the study those pregnancies complicated by pre-eclampsia or HELLP syndrome, placenta praevia, fetal anomalies or maternal trauma precipitating the symptoms.

Women admitted to the hospital for spontaneous preterm contractions, at risk for extreme preterm delivery, that is, gestational age between 24th and 32nd weeks, were included in the study. Exclusion criteria were age under 18 years, body mass index above 30, cervical dilatation over 3 cm. Additionally were excluded from the study those pregnancies complicated by pre-eclampsia or HELLP syndrome, placenta praevia, fetal anomalies or maternal trauma precipitating the symptoms. Immediately after admission to the hospital, patients were assessed by a clinician in order to evaluate the risk of preterm delivery. The criteria for preterm labor were regular contractions at a rate of 4 in 20 minutes or 8 in 1 hour with at least one of the following: progressive effacement or dilatation of the cervix over time or dilatation of the cervix greater than or equal to 2.0 cm. If they were judged to be at risk, they were treated according to the hospital guidelines following the national guideline for tocolysis of the Dutch Association for Obstetrics and Gynecology [23]. Three primiparous and five multiparous women attending the hospital for preterm contractions underwent multichannel EHG recording. The mean gestational age on admission was 28.2 ± 2.8 weeks. All singleton pregnancies received nifedipine oros 120 mg daily for the first 48 hours. After that, the dosage was lowered gradually over a 4-day period.

Three primiparous and five multiparous women attending the hospital for preterm contractions underwent multichannel EHG recording. The mean gestational age on admission was 28.2 ± 2.8 weeks. All singleton pregnancies received nifedipine oros 120 mg daily for the first 48 hours. After that, the dosage was lowered gradually over a 4-day period. The two twin gestations received nifedipine oros 90 mg as the maximum dosage in order to decrease the risk of acute lung oedema [24]. Before giving nifedipine oros, in all pregnancies a load up dose with nifedipine 20 or 40 mg was given. betamethasone for fetal lung maturation was given to 7 patients, in the other patient betamethasone was already completed in the referring hospital.

mum dosage in order to decrease the risk of acute lung oedema [24]. Before giving nifedipine oros, in all pregnancies a load up dose with nifedipine 20 or 40 mg was given. betamethasone for fetal lung maturation was given to 7 patients, in the other patient betamethasone was already completed in the referring hospital. Three different analyses were done by comparing the PSD peak frequency among different patient groups. The first comparison (three subjects, Group 1) was done between the value of fP derived from EHG signals recorded before (Subgroup 1.1) and 15 minutes after administration of nifedipine (Subgroup 1.2). For those patients who already started nifedepine treatment in the referring hospital before the first EHG measurement session (three subjects, Group2), the comparison was done between the first measurement, that is, within 24 hours after starting nifedipine oros 9 (Subgroup 2.1), and one day after finishing the tocolytic treatment (Subgroup 2.2). After the effects of betamethasone were expected to be occurred, that is, 48 hours after the first injection, tocolytic treatment was finished in a cutback plan over a 4-day period. The third analysis included all patients (Group3) and aimed at a retrospective comparison of the fP measured within 24 hours after starting treatment with nifedipine on women who delivered within one week (Subgroup 3.1) and the fP measured within 24 hours after starting treatment with nifedipine on women who delivered after more than one week (removed: from the EHG measurement) (Subgroup 3.2).

ve comparison of the fP measured within 24 hours after starting treatment with nifedipine on women who delivered within one week (Subgroup 3.1) and the fP measured within 24 hours after starting treatment with nifedipine on women who delivered after more than one week (removed: from the EHG measurement) (Subgroup 3.2). Of the included patients, five women were diagnosed with lower urinary tract infection at 2–4 days before attending the hospital and four of them were still being treated with antibiotics when nifedipine was given. Patient's characteristics are summarized in Table 1. 2.4. Statistical Analysis Sigma-Stat software (SPPS Inc, Chicago, IL) was implemented for statistical comparison of groups. The PSD peak frequencies in all channels for every single patient and all separate bursts were considered for analysis. Statistical analysis was performed using a 2-tailed t-student. A P value of less than  .05 was considered statistically significant. 3. Results In all patients, the uterine contractions detected by the EHG signal analysis corresponded to the uterine mechanical activity as recorded by tocodynamometer. In Figure 2 an example of these typical recordings is given for one patient. In general, the peak frequency as well as the uterine activity decreased in all patients after treatment with nifedipine (Table 2).

3. Results In all patients, the uterine contractions detected by the EHG signal analysis corresponded to the uterine mechanical activity as recorded by tocodynamometer. In Figure 2 an example of these typical recordings is given for one patient. In general, the peak frequency as well as the uterine activity decreased in all patients after treatment with nifedipine (Table 2). In three patients (Group 1), the EHG signal was recorded before (62 bursts analyzed in total) and 15 minutes (time expected that effect of nifedipine will occur) after administration of nifedipine (53 bursts analyzed in total). As shown in Table 2 (group 1), averaging the results in these three patients, a significant 7.2% (P < .05) decrease in PSD peak frequency was observed (Figure 3). In three other patients (Group 2), a comparison was made between the EHG signal recorded within 24 hours after starting treatment with nifedipine (56 bursts analyzed in total) and the EHG signal recorded 1 day after finishing tocolytic treatment after the 6-day period of treatment (63 bursts analyzed in total). Also in this case, a significant (P < .05) decrease in PSD peak frequency was observed in all patients (Table 2, Group 2). On average, the PSD peak frequency decreased of 6.7%.

total) and the EHG signal recorded 1 day after finishing tocolytic treatment after the 6-day period of treatment (63 bursts analyzed in total). Also in this case, a significant (P < .05) decrease in PSD peak frequency was observed in all patients (Table 2, Group 2). On average, the PSD peak frequency decreased of 6.7%. Finally, from the eight patients in which the EHG was recorded (Group 3), three patients delivered within 1 week after treatment with nifedipine (Subgroup 3.2). Frequency analysis of the signals recorded from these patients (3 patients, 56 bursts analyzed in total), to those patients delivering after more than 1 week (Subgroup 3.2, 5 patients, 98 bursts analyzed in total), showed, on average, a higher fP for Subgroup 3.1 as shown in Table 2. However, the significance of this difference was poor (P = .46).

nals recorded from these patients (3 patients, 56 bursts analyzed in total), to those patients delivering after more than 1 week (Subgroup 3.2, 5 patients, 98 bursts analyzed in total), showed, on average, a higher fP for Subgroup 3.1 as shown in Table 2. However, the significance of this difference was poor (P = .46). 4. Discussion The devices and methods currently used for predicting preterm delivery, such as vaginal examination, tocodynamometer, fetal fibronectin, and ultrasound measurement of cervical length, do not provide accurate diagnosis or prediction of preterm labor [25]. Intrauterine pressure catheters are reliable but limited by their invasiveness and the need for ruptured membranes. External uterine monitors, such as tocodynamometers, are uncomfortable, inaccurate, and depend on the personal interpretation of the examiner. Biological tests, such as fetal fibronectin, can be used as a prognostic marker, although with poor positive predictive values [5]. Even cervical length changes may not be an accurate indicator of true labor, as a large percentage of women with established cervical change do not deliver preterm even without tocolytic treatment [26]. As current methods are not capable of discriminating contractions, most obstetricians either treat all patients having preterm contractions or wait for cervical change. Delay in the diagnosis of preterm labor may result in a lower efficacy of tocolytic drugs [27, 28], while giving tocolysis to all patients exhibiting uterine activity is not free of risks for mother and fetus [21, 28, 29].

obstetricians either treat all patients having preterm contractions or wait for cervical change. Delay in the diagnosis of preterm labor may result in a lower efficacy of tocolytic drugs [27, 28], while giving tocolysis to all patients exhibiting uterine activity is not free of risks for mother and fetus [21, 28, 29]. Uterine contractions are the direct consequence of the generation and propagation of electrical activity at the myometrium [6, 7]. It has been demonstrated that the EHG signal recorded on the skin surface is representative of the electrical activity initiating the mechanical contraction of the uterus [7]. Being a measure of the first cause of a contraction, the EHG signal is potentially the best predictor of delivery. Furthermore, it is noninvasively and can therefore be employed for both singleton and multiple gestations during pregnancy and delivery. For the prediction of labor by electrohysterography previous studies mainly focused on the analysis of the frequency properties of the EHG signal [8, 9, 14, 15]. There is general agreement that the peak frequency derived from the PSD of the signal increases as the measurement-to-delivery time interval decreases [13, 16]. In particular, Buhimschi et al. [12] detected a significant shift from low to high PSD peak frequency in nonlaboring preterm patients as compared to laboring preterm patients. Doret et al. [11] performed a study in rats and showed the PSD peak frequency to be the earliest change to occur in the EHG signal as preterm delivery approached.

imschi et al. [12] detected a significant shift from low to high PSD peak frequency in nonlaboring preterm patients as compared to laboring preterm patients. Doret et al. [11] performed a study in rats and showed the PSD peak frequency to be the earliest change to occur in the EHG signal as preterm delivery approached. The shift of the EHG signal spectral content to higher frequencies can be explained by the underlying physiology. The frequency of action potentials within a burst is, in fact, a direct measure of the rate of the depolarization/repolarization process in the myometrial cells, a process governed by calcium ion-influx across ion channels [30]. When modifications in the ion channels of the myometrial cell plasma membrane initiate labor, the uterus becomes more excitable [30, 31], the signal propagation distance and contraction strength increase, and, as a result, higher frequencies within bursts of activity [13, 16] are expected.

lux across ion channels [30]. When modifications in the ion channels of the myometrial cell plasma membrane initiate labor, the uterus becomes more excitable [30, 31], the signal propagation distance and contraction strength increase, and, as a result, higher frequencies within bursts of activity [13, 16] are expected. Nifedipine is a calcium channel blocker. Therefore, it can be expected to act directly on the propagation of uterine activity. Lower frequencies within the EHG bursts after tocolytics can be associated to a lower level of electrical activity propagation and, ultimately, to the effectiveness of the treatment. The majority of tocolytic agents attempt to paralyze the myometrium, without addressing the root stimulus of preterm labor in a cause-specific way. Conversely, nifedipine acts by inhibiting the influx of calcium ions through the cell membrane and the release of intracellular calcium from the sarcoplasmatic reticulum. As a result, a decrease in intracellular free calcium occurs and leads to the inhibition of myosin light-chain kinase-mediated phosphorylation, which is calcium-dependant, and inducing a relaxation of the myometrium [32]. Moreover, nifedipine can be a preferable treatment for patients at risk for preterm delivery because it can be administered by oral route at any gestational age and severe side effects are reported to be very rare [20].

Nifedipine is a calcium channel blocker. Therefore, it can be expected to act directly on the propagation of uterine activity. Lower frequencies within the EHG bursts after tocolytics can be associated to a lower level of electrical activity propagation and, ultimately, to the effectiveness of the treatment. The majority of tocolytic agents attempt to paralyze the myometrium, without addressing the root stimulus of preterm labor in a cause-specific way. Conversely, nifedipine acts by inhibiting the influx of calcium ions through the cell membrane and the release of intracellular calcium from the sarcoplasmatic reticulum. As a result, a decrease in intracellular free calcium occurs and leads to the inhibition of myosin light-chain kinase-mediated phosphorylation, which is calcium-dependant, and inducing a relaxation of the myometrium [32]. Moreover, nifedipine can be a preferable treatment for patients at risk for preterm delivery because it can be administered by oral route at any gestational age and severe side effects are reported to be very rare [20]. In the present study, the PSD peak frequency decreased in all patients after treatment with nifedipine, also for those patients close to delivery. The same results were obtained by analyzing the EHG signal of patients who were recorded within 24 hours of first treatment compared to their EHG signal after end of treatment. A disadvantage of this study is that there is no control group. However, all of the patients we included in the study were at high risk for delivering very preterm.

In the present study, the PSD peak frequency decreased in all patients after treatment with nifedipine, also for those patients close to delivery. The same results were obtained by analyzing the EHG signal of patients who were recorded within 24 hours of first treatment compared to their EHG signal after end of treatment. A disadvantage of this study is that there is no control group. However, all of the patients we included in the study were at high risk for delivering very preterm. The PSD fP tendency highlighted in the present study suggests that nifedipine can be an effective tocolytic agent for inhibiting electrical activity propagation and therefore the effectiveness of uterine contraction when occurring preterm. In this case, the ultimate goal is in fact delaying the delivery in order to allow medical intervention or, in the most desirable case, to reach the term of pregnancy. In the present study, three patients delivered in 1 week, even if they had been treated with nifedipine. All these women had an asymptomatic urinary tract infection. It is well established that treatment with antibiotics is associated with prevention of preterm birth [33]. In our study, four of the five patients were treated with antibiotics and two of them did not deliver within one week. The effectiveness of tocolytics in the presence of urinary tract infections needs further to be defined.

established that treatment with antibiotics is associated with prevention of preterm birth [33]. In our study, four of the five patients were treated with antibiotics and two of them did not deliver within one week. The effectiveness of tocolytics in the presence of urinary tract infections needs further to be defined. The reason of the failure of tocolytics in preventing preterm delivery is still an unsolved issue for gynecologists. However, it has been suggested that the effectiveness of tocolytics is highly dependent on early initiation of the therapy. Therefore, timely recognition of the process leading to the delivery is fundamental. Garfield et al. [18] and Linhart et al. [26] demonstrated, in fact, that parturition is a two-step process consisting of a conditioning (preparatory) phase, followed by active labor. During the conditioning phase, there is a progression of uterine contractility from an inactive to an active state. In the myometrium, the preparatory process involves changes in transduction mechanisms and the synthesis of several new proteins. At some point of the conditioning step, the process becomes irreversible and leads to active labor and, ultimately, to delivery [18]. It is known that agents used to intervene in labor may be of greatest value during the preparatory phase in the progression to delivery [1]. Therefore, it could be suggested that the women in this study who failed to respond to therapy were treated too late and that during the EHG recording, an irreversible change to the active phase of labor might already have occurred.

y be of greatest value during the preparatory phase in the progression to delivery [1]. Therefore, it could be suggested that the women in this study who failed to respond to therapy were treated too late and that during the EHG recording, an irreversible change to the active phase of labor might already have occurred. On average, the women who did not respond to the therapy and delivered within one week from the measurement showed a higher fP than the other patients. However, not all patients within one week from delivery showed higher fP than the rest; finding a cut-off value for an accurate classification of the two populations was therefore not possible. Due to the limited size of the study population, these preliminary results, even if obtained by analyzing a large number of bursts, might be affected by the different treatment-to-measurement time. We showed that the PSD peak frequency decreased in all women after treatment with nifedipine, although, due to the small number of women in the study and variable times of EHG recording (in five of the eight women, EHG recordings could only be made after the start of nifedipine treatment), this could not be linked with a significant increase in time to delivery.

sed in all women after treatment with nifedipine, although, due to the small number of women in the study and variable times of EHG recording (in five of the eight women, EHG recordings could only be made after the start of nifedipine treatment), this could not be linked with a significant increase in time to delivery. Previous studies on a more extended database on animals and women not treated with tocolytics concluded that the peak frequency derived from the PSD of the EHG signal increases as the measurement-to-delivery interval decreases [17]. Noticeably, these conclusions were also mostly derived by mean differences for all the patients. However, ultimately, due to a poor significance of the difference, a cut-off value of peak frequency has not been found yet. Therefore, further studies are required in order to develop methods that can reliably differentiate the uterine status of labor. Longitudinal studies in pregnant women could support the determination of the EHG properties related to the electrophysiological process leading to preterm labor. Further research on the differentiation of the uterine electrical activity in the non-prepared state before labor and the active state of labor would be necessary; objective criteria to evaluate the state of the uterine preparedness for labor can in fact improve management of preterm labor.

ss leading to preterm labor. Further research on the differentiation of the uterine electrical activity in the non-prepared state before labor and the active state of labor would be necessary; objective criteria to evaluate the state of the uterine preparedness for labor can in fact improve management of preterm labor. 5. Conclusion In this study, the efficacy of nifedipine as a tocolytic drug on the uterine activity was evaluated by monitoring and analysis of the EHG signal. Treatment with nifedipine leads to a statistically significant shift to lower PDS peak frequency in the EHG. As the PDS peak frequency is expected to rise when delivery approaches, from these preliminary results, it could be concluded that nifedipine is an effective tocolytic agent for suppressing preterm contractions. However, further studies are needed to relate the decrease in the myometrium electrical activity to the physiological effects leading to the effectiveness of the tocolitic treatment. Even with accurate tocolysis, some patients do not respond to therapy and preterm delivery cannot not be inhibited. In these patients, an irreversible change to the active phase of labor has probably occurred. As tocolytic agents may be more effective during certain periods in the progression to delivery, it is important to develop methods that could reliably diagnose the risk of preterm birth at early stages.

nnot not be inhibited. In these patients, an irreversible change to the active phase of labor has probably occurred. As tocolytic agents may be more effective during certain periods in the progression to delivery, it is important to develop methods that could reliably diagnose the risk of preterm birth at early stages. This study confirmed that the EHG signal analysis is a reliable technique for non-invasive monitoring the uterine contraction and has great potential for predicting labor, especially for patients at high risk for preterm delivery. However, considering also the previous literature on the topic, our results suggest that the analysis of the EHG signal frequency alone may not provide a universal cut-off value for assessing whether the irreversible process leading to labor has already started or not. Since the spreading of the electrical activity through the myometrium is the first cause of labor, the analysis of the EHG signal propagation, for example, in terms of direction and velocity [34–36], might provide an important contribution to the development of methods that can objectively evaluate the state of the uterine preparedness for labor and to identify the patients at risk for preterm labor. Contribution to Authorship Maartje P. G. C. Vinken worked on conception and design of data, acquisition of data, and analysis and interpretation of data; C. Rabotti on analysis and interpretation of data, critical revision of article; M. Mischi on critical revision of article and statistical analysis; Judith O. E. H. van and G. S. Oei on the critical revision of article.

rked on conception and design of data, acquisition of data, and analysis and interpretation of data; C. Rabotti on analysis and interpretation of data, critical revision of article; M. Mischi on critical revision of article and statistical analysis; Judith O. E. H. van and G. S. Oei on the critical revision of article. Acknowledgments Part of this work was supported by the Dutch Technology Foundation STW. The study was approved by the medical ethical committee of the Maxima Medical Centre, METC-number 0650. Figure 1 Electrode position and numbering on the abdomen. Prior to electrode positioning, the half length of the uterus was measured. The interelectrode distance was 2 cm for all electrodes. Ref = reference electrode, DRL = ground electrode. Figure 2 Tocographic and EHG recording of a patient with preterm uterine contractions. The figure shows an example of the temporal correlation between the uterine electrical and mechanical activity. Figure 3 Example of preprocessed EHG signal (upper plot) before (left) and after (right) nifedipine treatment. The corresponding power spectral densities before (left) and after (right) nifedipine treatment are depicted in the bottom part of the figure with an indication of the peak frequency fP. Table 1 Study population characteristics.

1. Introduction Twin discordance is a complication that is unique to multiple gestations. It is defined as discordance in the growth pattern, genotype, and phenotype of the twins. It can also just refer to a birth weight discordance of more than 20% between the twins [1]. Perinatal outcomes are worse in discordant twins than in concordant twins. Discordant twins have been found to have a significant increase in perinatal death, congenital anomalies, severe intracranial haemorrhage, neurological morbidity, neonatal asphyxia, and respiratory distress [2, 3]. It is important to study the causes of twin discordance in order to find a treatment and improve the prognosis. Abnormal genomic imprinting may be one of the important causes that results in twin discordance [4]. Genomic imprinting is a special phenomenon that operates outside of Mendel's law and directs the nonrandom monoallelic expression of specific autosomal genes according to their parental origin. Loss of imprinting (LOI) is an epigenetic alteration associated with the expression of the normally repressed parental allele or silence of the normally active allele. Imprinted genes play a key role in fetal and placental growth, development of particular lineages, and behaviour [5, 6]. It has been reported that an abnormal adjustment of genomic imprinting may result in congenital anomalies, some genetic diseases, and tumours [7, 8].

allele or silence of the normally active allele. Imprinted genes play a key role in fetal and placental growth, development of particular lineages, and behaviour [5, 6]. It has been reported that an abnormal adjustment of genomic imprinting may result in congenital anomalies, some genetic diseases, and tumours [7, 8]. Insulin-like growth factor II (IGF-II) is one of the crucial imprinted genes related to fetal growth and placental development. It has 4 promoters and is expressed by the paternal allele. The expression of IGF-II is very complex since its imprinting is periodic, as well as tissue and promoter specific. Placental IGF-II can control fetal growth by modulating placental development and nutritional transfusion [9]. The LOI of the IGF-II gene is associated with fetal anomalies, while mutations of the IGF-II gene promoter result in fetal growth restriction [5]. Few studies have examined the changes of IGF-II in twin pregnancies, and there are no reports dealing with IGF-II imprinting and promoter regulation in twin placenta. It is uncertain whether IGF-II gene imprinting and promoter regulation in the placenta are associated with twin discordance. To clarify the role of imprinted IGF-II genes in the pathogenesis of twin discordance, the imprinting and promoter usage of IGF-II were studied.

printing and promoter regulation in twin placenta. It is uncertain whether IGF-II gene imprinting and promoter regulation in the placenta are associated with twin discordance. To clarify the role of imprinted IGF-II genes in the pathogenesis of twin discordance, the imprinting and promoter usage of IGF-II were studied. 2. Materials and Methods 2.1. Placenta Collection Human placentas were collected after birth in the first Affiliated Hospital of Sun Yat-sen University (Guangzhou, China). Eighty sets of twins were enrolled, including 55 sets (64.71%) of normal twins (T0 group), 17 sets (20.00%) of twins with weight discordance (T1 group), and 13 sets (15.29%) of twins with phenotype discordance (T2 group). Weight discordance was defined as a birth weight discordance of more than 20% between the twins. Phenotype discordance was defined as being present when only one of the twins had a malformation or both twins had different abnormalities. Detailed description of T2 group was listed in Table 1. For each twin, tissue samples were collected from under the umbilical cord even when the twins shared the same placenta. The tissues were washed with a saline solution to remove maternal blood, then frozen in liquid nitrogen, and stored at −80°C until analysis.

2. Materials and Methods 2.1. Placenta Collection Human placentas were collected after birth in the first Affiliated Hospital of Sun Yat-sen University (Guangzhou, China). Eighty sets of twins were enrolled, including 55 sets (64.71%) of normal twins (T0 group), 17 sets (20.00%) of twins with weight discordance (T1 group), and 13 sets (15.29%) of twins with phenotype discordance (T2 group). Weight discordance was defined as a birth weight discordance of more than 20% between the twins. Phenotype discordance was defined as being present when only one of the twins had a malformation or both twins had different abnormalities. Detailed description of T2 group was listed in Table 1. For each twin, tissue samples were collected from under the umbilical cord even when the twins shared the same placenta. The tissues were washed with a saline solution to remove maternal blood, then frozen in liquid nitrogen, and stored at −80°C until analysis. 2.2. Preparation of Genomic DNA and RNA Approximately 0.1 g of the frozen tissues was homogenized using Teflon tissue grinders. Genomic DNA (gDNA) was extracted using phenol and chloroform. Total RNA isolation was performed using TRI-reagent as suggested by the manufacturer (Molecular Research Center Inc, Cincinnati, OH, USA). The quality and quantity of the extracted RNA was assessed using an Ultraviolet Bioanalyzer (Eppendorf, Hamburg, Germany). Total RNA was treated with 10 Units of DNase I (Fermentas, Vilnius, Lithuania) for 30 minutes at 37°C to eliminate any genomic DNA residue.

rer (Molecular Research Center Inc, Cincinnati, OH, USA). The quality and quantity of the extracted RNA was assessed using an Ultraviolet Bioanalyzer (Eppendorf, Hamburg, Germany). Total RNA was treated with 10 Units of DNase I (Fermentas, Vilnius, Lithuania) for 30 minutes at 37°C to eliminate any genomic DNA residue. 2.3. Identification of Genomic Polymorphisms To assess the heterozygosity of the IGF-II gene, Apa I polymorphism within exon 9 of IGF-II was screened. For PCR amplification of the IGF-II gene, the following primers were used: sense strand (P1F), 5′CTTGGACTTTGAGTCAAATTGG-3′, and antisense strand (P1R), 5′-CCTCCTTTGGTCTTACTGGG-3′. For each 50 μl PCR, 2 μl gDNA, 5 μl 10×PCR buffer, and 0.5 U Taq DNA Polymerase (TaKaRa, Dalian, China), 1 μl sense and antisense primer (10 pmol) and 1 μl dNTPs (2.5 mM) were amplified using an initial denaturation step at 94°C for 7 minutes, followed by 35 cycles of denaturation at 94°C for 30 s, annealing at 60°C for 40 s with a 40 s extension, and a final extension at 72°C for 9 minutes. The amplification resulted in a gene fragment of 236 bp whose allele had an ApaI restriction site at 173 bp. The PCR products were digested with the restriction enzyme ApaI (New England Biolabs, Ipswich, UK) and loaded onto 3% agarose gel, which was followed by ethidium bromide staining.

xtension at 72°C for 9 minutes. The amplification resulted in a gene fragment of 236 bp whose allele had an ApaI restriction site at 173 bp. The PCR products were digested with the restriction enzyme ApaI (New England Biolabs, Ipswich, UK) and loaded onto 3% agarose gel, which was followed by ethidium bromide staining. 2.4. Allele-Specific Expression of IGF-II Genes Allele-specific expression of the IGF-II genes was examined using reverse transcription-PCR (RT-PCR) followed by enzyme digestion as described above. cDNA was synthesized using about 1 μg RNA and reverse transcriptase reagents (RevertAid H Minus First Strand cDNA Synthesis Kit, Fermentas, Vilnius, Lithuania) as suggested by the manufacturer. The resulting cDNA samples were heated at 70°C for 10 minutes to inactivate the reverse transcriptase, and then used for PCR amplification as described above. The primers used for RT-PCR of IGF-II were P1F and P1R as described above. The PCR products were digested with ApaI (yielding either a 236 bp fragment or a 173 and 63 bp fragment), electrophoresed through a 3% agarose gel, and visualized with ethidium bromide.

se, and then used for PCR amplification as described above. The primers used for RT-PCR of IGF-II were P1F and P1R as described above. The PCR products were digested with ApaI (yielding either a 236 bp fragment or a 173 and 63 bp fragment), electrophoresed through a 3% agarose gel, and visualized with ethidium bromide. 2.5. Promoter-Specific Expressions of IGF-II Promoter-specific expressions of IGF-II were quantified using the Absolute QPCR SYBR Green Mix plus ROX vial Kit (Abgene, Epsom, UK) and the OpticonTM 2 continuous fluorescence detection system (MJ Research, Boston, MA, USA). The primers used in real-time quantitative PCR (RT-PCR) for each promoter specific expression of IGF-II are listed in Table 2. The OpticonTM 2 was programmed as follows: preincubation and denaturation of template cDNA for 15 minutes at 94°C; followed by 40 cycles of amplification: 96°C for 15 s, 66°C for 40 s (P1) or 69°C for 20 s (P2) or 60°C for 30 s (P3) or 63°C for 30 s (P4), and 72°C for 40 s (P1) or 20 s (P2) or 30 s (P3, P4). The melting curve analysis was performed at 65°C to 98°C, 0.2°C /read, 2 seconds hold. 2.6. Statistics The statistical analysis was performed using the χ2 test, Fisher's exact test, Student's t-test, and ANOVA, as appropriate. A P value of less than  .05 was considered significant.

2.5. Promoter-Specific Expressions of IGF-II Promoter-specific expressions of IGF-II were quantified using the Absolute QPCR SYBR Green Mix plus ROX vial Kit (Abgene, Epsom, UK) and the OpticonTM 2 continuous fluorescence detection system (MJ Research, Boston, MA, USA). The primers used in real-time quantitative PCR (RT-PCR) for each promoter specific expression of IGF-II are listed in Table 2. The OpticonTM 2 was programmed as follows: preincubation and denaturation of template cDNA for 15 minutes at 94°C; followed by 40 cycles of amplification: 96°C for 15 s, 66°C for 40 s (P1) or 69°C for 20 s (P2) or 60°C for 30 s (P3) or 63°C for 30 s (P4), and 72°C for 40 s (P1) or 20 s (P2) or 30 s (P3, P4). The melting curve analysis was performed at 65°C to 98°C, 0.2°C /read, 2 seconds hold. 2.6. Statistics The statistical analysis was performed using the χ2 test, Fisher's exact test, Student's t-test, and ANOVA, as appropriate. A P value of less than  .05 was considered significant. 3. Results 3.1. Imprinting of IGF-II For the imprinting study, DNA specimens from the placenta were first analyzed for heterozygosity using IGF-II gene polymorphisms at ApaI. Then, RT-PCR products were analyzed for LOI by biallelic expression of IGF-II ( Figure 1). The incidence of LOI of the IGF-II gene was 43.8% in the T2 group placenta, which was higher than that in the T0 and T1 groups (17.6% and 15.8%, resp., P < .05) (Table 3). All LOI occurred in abnormal twins except in one twin pair in which one twin was large for gestational age.

llelic expression of IGF-II ( Figure 1). The incidence of LOI of the IGF-II gene was 43.8% in the T2 group placenta, which was higher than that in the T0 and T1 groups (17.6% and 15.8%, resp., P < .05) (Table 3). All LOI occurred in abnormal twins except in one twin pair in which one twin was large for gestational age. There were 4 cases of twin-twin transfusion syndrome, including 1 acardiac twin. Among these 4 cases, 3 sets were IGF-II homozygous, and 1 set, which was a Stage III case of twin-twin transfusion syndrome, was IGF-II heterozygous. Polyhydramnios occurred in the recipient twin at 23 gestational weeks, and the maximal vertical amniotic fluid pocket was 16.5 cm. The donor twin, whose bladder was not visible, was a “stuck twin”, and the reverse end-diastolic velocity of the umbilical artery could be seen. The woman decided to abort the pregnancy. After induction, the imprinting status was analyzed. The donor twin had normal imprinting expression, while the recipient twin had LOI.

e donor twin, whose bladder was not visible, was a “stuck twin”, and the reverse end-diastolic velocity of the umbilical artery could be seen. The woman decided to abort the pregnancy. After induction, the imprinting status was analyzed. The donor twin had normal imprinting expression, while the recipient twin had LOI. 3.2. Promoter Usage of IGF-II The usage of IGF-II promoters was similar in the T0, T1, and T2 groups. In each group, among the 4 promoters, the transcripts of IGF-II gene promoter 3 were the highest. The transcripts of IGF-II gene promoter 3 in group T2 were significantly lower than in the T0 and T1 groups (P < .05). The transcripts of IGF-II gene promoter 3 were lower in twin placenta with LOI than in those with normal imprinting (P < .05). Intrapair transcriptions of the 4 IGF-II gene promoters did not differ among the T0, T1, and T2 groups (P > .05). There was a positive correlation between the transcription of IGF-II gene promoter 3 and the transcription of IGF-II gene promoter 4 (r = 0.229, P < .05); there were no correlations with other promoters (P > .05).

ir transcriptions of the 4 IGF-II gene promoters did not differ among the T0, T1, and T2 groups (P > .05). There was a positive correlation between the transcription of IGF-II gene promoter 3 and the transcription of IGF-II gene promoter 4 (r = 0.229, P < .05); there were no correlations with other promoters (P > .05). 4. Discussion The paternally expressed, imprinted gene, IGF-II, located on chromosome 11 p15.5, encodes an autocrine growth factor that plays an important role in embryonic growth. LOI of the IGF-II gene results in generalized constitutional overgrowth, malformation, and a predisposition to the development of specific embryonal tumours, most commonly Wilm's tumour. Our data shows that the incidence of LOI of the IGF-II gene was higher in T2 than in T0 and T1 group placenta, and that all LOI occurred in abnormal twins, except in one case. The incidence of LOI of the IGF-II gene was similar in the T0 and T1 groups. Ravenel et al. [10], observed that there was virtually complete segregation of the intralobar nephrogenic rest (ILNR)-like and perilobar nephrogenic rest (PLNR)-like tumours, depending on the imprinting status; 9 (90%) of 10 PLNR-like tumours had LOI, but only 1 (6.7%) of 15 ILNR-like tumours had LOI, although the ILNR-like and PLNR-like tumours were both Wilm's tumours. Thus, it is possible that the pathology involved in phenotype discordant twins and weight discordant twins is different, and that LOI of the IGF-II gene may be closely related with phenotype discordant twins, as in PLNR-like tumours. Among the weight discordant twins, most of the small fetuses were selective fetal growth restriction and the larger ones were appropriate for gestational age, not overgrowth. Furthermore, LOI of the IGF-II gene is much more relevant to classical cases of Beckwith-Wiedemann syndrome (BWS) or Silver-Russell syndrome (SRS) than to sporadic growth deficiencies or overgrowth. These might be the possible reason why LOI of the IGF-II gene in placentae is not a major contributor to weight discordance.

thermore, LOI of the IGF-II gene is much more relevant to classical cases of Beckwith-Wiedemann syndrome (BWS) or Silver-Russell syndrome (SRS) than to sporadic growth deficiencies or overgrowth. These might be the possible reason why LOI of the IGF-II gene in placentae is not a major contributor to weight discordance. Placental IGF-II regulates the growth and transport capacity of the placenta, thereby controlling the supply of nutrients. It may also directly regulate the growth rate of fetal tissues, thereby controlling fetal nutrient demand. LOI of the IGF-II gene in the placenta might alter the balance between placental and fetal growth and lead to fetal abnormalities. It is reported that LOI of the IGF-II gene was relevant to malformation like macroglossia and exomphalos, but not other malformation. The malformation cases in our data are diverse. It is not known whether fetal abnormalities are due to a direct effect of LOI of the IGF-II gene in the placenta, or whether LOI of the IGF-II gene in the placenta reflects LOI of the IGF-II gene in other fetal tissue, even other gene imprinting. Further investigation of methylation of other related imprinting genes is warranted to delineate the possible mechanism of the diverse malformation. And a larger series of phenotype discordant twins should be followed up, in which each type of specific malformation is represented by more cases.

n other gene imprinting. Further investigation of methylation of other related imprinting genes is warranted to delineate the possible mechanism of the diverse malformation. And a larger series of phenotype discordant twins should be followed up, in which each type of specific malformation is represented by more cases. During the life cycle of the organism, genomic imprints change in characteristic ways. They undergo erasure, establishment, and maintenance of methylation imprints at imprinting centers during germ cell and embryonic development. Imprint patterns are maintained as chromosomes duplicate and segregate, although there is genome-wide demethylation after fertilization and a wave of de novo methylation after implantation. Any changes in the microenvironment around the implantation could interfere with the mechanism of maintenance, so twins might be more sensitive to get LOI with the epigenetic risks related to assisted reproductive technologies and monochorionic twins. As the changes occur around the implantation, imprints may sometimes be discordant in twins, even in monozygotic twins.

antation could interfere with the mechanism of maintenance, so twins might be more sensitive to get LOI with the epigenetic risks related to assisted reproductive technologies and monochorionic twins. As the changes occur around the implantation, imprints may sometimes be discordant in twins, even in monozygotic twins. In our data, among 4 cases of twin-twin transfusion syndrome, 3 sets were homozygous for IGF-II, and 1 set was heterozygous (the donor twin had normal imprinting expression, while the recipient twin had LOI). There have been no reports dealing with the relationship between twin-twin transfusion syndrome and LOI. Bajoria et al. [11], found that fetal IGF-II levels in recipient twins with TTTS were higher than those in donor twins. It was also reported that IGFs may be involved in endothelial dysfunction [12]. Thus, more research is needed to determine whether LOI of the IGF-II gene induced the abnormality of the vessels in the twin placenta that leads to twin-twin transfusion syndrome or was just related to the complications, such as twin-twin transfusion syndrome with abnormal vessels.

lved in endothelial dysfunction [12]. Thus, more research is needed to determine whether LOI of the IGF-II gene induced the abnormality of the vessels in the twin placenta that leads to twin-twin transfusion syndrome or was just related to the complications, such as twin-twin transfusion syndrome with abnormal vessels. IGF-II is transcribed from four distinct promoters, P1–P4. It has been found that the fetal liver uses only three promoters (P2, P3, and P4) for IGF-II transcription, with the P3 promoter having the highest activity [13, 14]. Late placental tissues of twin pregnancies show similar use of the IGF-II promoter as in fetal liver. The P3 transcript levels were higher than those of P4, while the P2 and P1 transcript levels were low or zero. Compared to normal twins, the P3 transcript levels in the T2 group were dramatically lower. This suggests that P3 plays the most important role in the placenta, and that its abnormal activity may affect fetal growth.

transcript levels were higher than those of P4, while the P2 and P1 transcript levels were low or zero. Compared to normal twins, the P3 transcript levels in the T2 group were dramatically lower. This suggests that P3 plays the most important role in the placenta, and that its abnormal activity may affect fetal growth. Usually, P1 has biallelic expression, and P2-4 have paternally monoallelic expression [15, 16]. It has been proposed that IGF-II expression through promoter P1 could explain the biallelic expression patterns in some neoplasms [17, 18]. In these neoplasms, P1 plays the most important role, and most transcripts are from P1 rather than P2-4. However, we found that, when there was LOI of the IGF-II gene in twin placenta, the activity of P3, not P1, was greatly changed. This suggests that methylation changes of P3 induce biallelic expression and decrease P3 activity. The mechanism of LOI of the IGF-II gene in phenotype discordant twins may be different from that in some neoplasms. For phenotype discordant twins, it was the biallelic expression of P3 and not P1 that caused the discordance. In conclusion, our data suggests that the promoter 3 specific LOI of the IGF-II gene may be closely related with phenotype discordance, not weight discordance. Further investigation of methylation of other related imprinting genes and more phenotype discordant twins was warranted to delineate the possible mechanism of the diverse malformation. Following up the weight discordance twins with LOI will find out whether they are predisposed to embryonal tumours.

eight discordance. Further investigation of methylation of other related imprinting genes and more phenotype discordant twins was warranted to delineate the possible mechanism of the diverse malformation. Following up the weight discordance twins with LOI will find out whether they are predisposed to embryonal tumours. Acknowledgments This work was supported by the Natural Science Foundation of Guangdong Province (Project. no. 4009370) and Project of Science and Technology of Guangdong Province (Project. no. 2008B080701022). Figure 1 Polymorphisms and imprinting of the IGF-II gene in the placenta. (a) polymorphisms of IGF-II: 1 had type AA homozygosity; 4 had type BB homozygosity; 2,3,5,6 had type AB heterozygosity; and M, marker. (b) imprinting of IGF-II: 1, loss of imprinting AB; 2–4, normal imprinting A; 5, normal imprinting B; and M, marker. Table 1 Detail description of phenotype discordant in T2 group.

Figure 1 Polymorphisms and imprinting of the IGF-II gene in the placenta. (a) polymorphisms of IGF-II: 1 had type AA homozygosity; 4 had type BB homozygosity; 2,3,5,6 had type AB heterozygosity; and M, marker. (b) imprinting of IGF-II: 1, loss of imprinting AB; 2–4, normal imprinting A; 5, normal imprinting B; and M, marker. Table 1 Detail description of phenotype discordant in T2 group. No. Chorion Twin A Twin B Abnormity PM* I** Abnormity PM* I** 1 MCMA Transposition of the great vessels, ventricular septal defect AB AB N AB AB 2 DCDA Mass on right back AB AB N AB B 3 DCDA Hydatidiform mole AB AB N AB A 4 DCDA Intrauterine death AB AB N AB B 5 DCDA Anencephaly, spinal bifida, omphalocele AB AB N BB 6 MCDA Thoracogastroschisis, ventricular septal defect, aortic overriding AB AB N AB B 7 DCDA Fetal hydrops AB A N AA 8 DCDA Hypospadias AB A N AA 9 DCDA Abdominal-wall defect, spinal bifida AB A N BB 10 DCDA Double outlet right ventricle AB B N AB B 11 DCDA Omphalocele AA N AA 12 MCDA Acardiac twin AA N AA 13 MCDA Hypospadias AA N AA *AA, BB, and AB are three different gene polymorphisms; AA/BB reflects homozygosity, and AB reflects heterozygosity. **A, B, and AB are three different imprinting states; A/B is the normal imprinting expression, while AB reflects loss of imprinting. Abbreviations: Polymorphisim (PM), Imprinting (I), monochorionic monoamniotic twin (MCMA), dichorionic diamniotic twin (DCDA), monochorionic diamniotic twin (MCDA), and normal (N). Table 2 Primers for promoter-specific expression of the IGF-II gene and internal control.

No. Chorion Twin A Twin B Abnormity PM* I** Abnormity PM* I** 1 MCMA Transposition of the great vessels, ventricular septal defect AB AB N AB AB 2 DCDA Mass on right back AB AB N AB B 3 DCDA Hydatidiform mole AB AB N AB A 4 DCDA Intrauterine death AB AB N AB B 5 DCDA Anencephaly, spinal bifida, omphalocele AB AB N BB 6 MCDA Thoracogastroschisis, ventricular septal defect, aortic overriding AB AB N AB B 7 DCDA Fetal hydrops AB A N AA 8 DCDA Hypospadias AB A N AA 9 DCDA Abdominal-wall defect, spinal bifida AB A N BB 10 DCDA Double outlet right ventricle AB B N AB B 11 DCDA Omphalocele AA N AA 12 MCDA Acardiac twin AA N AA 13 MCDA Hypospadias AA N AA *AA, BB, and AB are three different gene polymorphisms; AA/BB reflects homozygosity, and AB reflects heterozygosity. **A, B, and AB are three different imprinting states; A/B is the normal imprinting expression, while AB reflects loss of imprinting. Abbreviations: Polymorphisim (PM), Imprinting (I), monochorionic monoamniotic twin (MCMA), dichorionic diamniotic twin (DCDA), monochorionic diamniotic twin (MCDA), and normal (N). Table 2 Primers for promoter-specific expression of the IGF-II gene and internal control. Name Sequence (5′-3′) Length P1 (F) AG TCC TGA GGT GAG CTG CTG 181 bp P2 (F) ACC GGG CAT TGC CCC CAG TC 277 bp P3 (F) ACA TTC GGC CCC CGC GAC TC 201 bp P4 (F) TCC TCC TCC TCC TGC CCC AG 118 bp P1-4(R) CAA TGC AGC ACG AGG CGA AG B-actin(F) GCG GGA AAT CGT GCG TGA CAT T 228 bp B-actin(R) GAT GGA GTT GAA GGT AGT TTG GTG Table 3 Polymorphisms and imprinting of the IGF-II gene in placenta [n (%)].

) ACC GGG CAT TGC CCC CAG TC 277 bp P3 (F) ACA TTC GGC CCC CGC GAC TC 201 bp P4 (F) TCC TCC TCC TCC TGC CCC AG 118 bp P1-4(R) CAA TGC AGC ACG AGG CGA AG B-actin(F) GCG GGA AAT CGT GCG TGA CAT T 228 bp B-actin(R) GAT GGA GTT GAA GGT AGT TTG GTG Table 3 Polymorphisms and imprinting of the IGF-II gene in placenta [n (%)]. Group Polymorphisim Imprinting n AA/BB* AB* n A/B** AB** T0 group 110 41 (37.3) 69 (62.7) 68 56 (82.4) 12 (17.6) T1 group 34 15 (44.1) 19 (55.9) 19 16 (84.2) 3 (15.8) T2 group 26 10 (38.5) 16 (61.5) 16 9 (56.2) 7 (43.8)# *AA, BB, and AB are three different gene polymorphisms; AA/BB reflects homozygosity, and AB reflects heterozygosity. **A, B, and AB are three different imprinting states; A/B is the normal imprinting expression, while AB reflects loss of imprinting. In the T0 group, the imprinting status could not be detected in one case with AB gene polymorphism.# The frequency of loss of imprinting of the IGF-II gene was statistically significantly different between the T0 group and the T2 group, P = .035.

1. Introduction Endometrial carcinoma is the most common type of female genital tract malignancy. It is estimated that 42,160 cases of endometrial carcinoma were diagnosed in the United States in 2008 and 7780 women would die from the disease [1]. Since the primary symptom is abnormal uterine bleeding in postmenopausal women, most patients would have a better chance of survival if diagnosed at an early stage of the disease. However, there still remain a lot of challenges in the clinical treatment of endometrial carcinoma. At the diagnostic stage, the condition of the disease can range from excellent prognosis with high curability to aggressive disease with poor outcome. In this paper, our goals are to discuss current challenges in the management of endometrial carcinoma and to provide an overview of the new approaches that would help overcome these challenges.

the condition of the disease can range from excellent prognosis with high curability to aggressive disease with poor outcome. In this paper, our goals are to discuss current challenges in the management of endometrial carcinoma and to provide an overview of the new approaches that would help overcome these challenges. 2. Pathological and Biologic Type Pathological examination is the cornerstone in diagnosing endometrial carcinoma. There are different types of endometrial carcinomas, as shown in Table 1. The endometrioid tumors are further classified according to the degree of morphological differentiation. As defined by the International Federation of Gynecology and Obstetrics (FIGO), endometrioid carcinoma of grade 1 consists of well-formed glands, with no more than 5% solid nonsquamous areas (areas of squamous differentiation are not deemed to be solid tumor growth). Carcinomas of grade 2 consist of 6–50% and grade 3 consists of more than 50% solid nonsquamous areas. The tumor is upgraded from grade 1 to 2, or from grade 2 to 3 if striking cytological atypia is found [2].

re than 5% solid nonsquamous areas (areas of squamous differentiation are not deemed to be solid tumor growth). Carcinomas of grade 2 consist of 6–50% and grade 3 consists of more than 50% solid nonsquamous areas. The tumor is upgraded from grade 1 to 2, or from grade 2 to 3 if striking cytological atypia is found [2]. It is considered that the different molecular biology of the different histological type is probably related to different behavior and prognosis. With more understanding about biologic behavior of endometrial carcinoma, we know that histological grading is far from enough to evaluate degrees of malignancy of endometrial carcinomas. Although about 80% of all endometrial carcinomas are of the endometrioid type, several subtypes or variants of endometrioid carcinoma provide more valuable information for guiding therapy. Most of all, special subtypes may be associated with higher death rate, for example, uterine papillary serous tumors and clear cell carcinoma. On the basis of their Pathological and biologic features, endometrial carcinomas are classified into 2 subtypes [2].

inoma provide more valuable information for guiding therapy. Most of all, special subtypes may be associated with higher death rate, for example, uterine papillary serous tumors and clear cell carcinoma. On the basis of their Pathological and biologic features, endometrial carcinomas are classified into 2 subtypes [2]. About 80% of all endometrial carcinomas are type I carcinoma (endometrioid type), arise from atypical complex hyperplasia, which seems to affect mainly pre- and perimenopausal women and presents with less myometrial invasion, lower grade disease. The type I tends to arise in the setting of prior estrogen stimulation because it is usually estrogen receptor positive and associated with hyperestrogenism [3, 4]. Other associated findings include late onset of menopause, nulliparity, diabetes mellitus, and hypertension. The patients with Type I endometrial carcinoma have a better prognosis since the lesion is limited to the uterus in 70% of the cases; the 5-year survival rate of these patients is more than 85%.

rogenism [3, 4]. Other associated findings include late onset of menopause, nulliparity, diabetes mellitus, and hypertension. The patients with Type I endometrial carcinoma have a better prognosis since the lesion is limited to the uterus in 70% of the cases; the 5-year survival rate of these patients is more than 85%. In contrast, type II tends to occur in elderly postmenopausal women with high risk of relapse and metastatic disease, often with aggressive histologies such as serous or clear cell [3, 4]. Type II endometrial carcinomas appear to be unrelated to high estrogen levels. These tumors are not oestrogen driven and often develop in nonobese women. Type II endometrial carcinomas appear to be associated with endometrial atrophy; the histological type is either poorly differentiated endometrioid or nonendometrioid. A high proportion of tumors, even those with little or no myometrial invasion, have extensive extrauterine spread with complete surgical staging. More than 60% of patients with type II endometrial carcinoma present with advanced disease; 5-year survival is 43% for patients with stage III disease and 3% for patients with stage IV disease. Without adjuvant chemotherapy or vaginal brachytherapy, the recurrence rate is 23% in patients with stage I disease [4].

ging. More than 60% of patients with type II endometrial carcinoma present with advanced disease; 5-year survival is 43% for patients with stage III disease and 3% for patients with stage IV disease. Without adjuvant chemotherapy or vaginal brachytherapy, the recurrence rate is 23% in patients with stage I disease [4]. The molecular basis for different progression of these two subtypes is still unknown. However, a lot of clinical observations exhibited that gene alterations are specific for carcinomas of types I and II, which supports a dualistic model of endometrial carcinogenesis [5–10]. Type I endometrial carcinomas display a high incidence of alterations in KRAS oncogene, PTEN tumor suppressor gene [5, 6, 11–13], the β-catenin gene [14, 15], as well as defects in mismatch repair that results in microsatellite instability [10, 16]. In contrast, type II endometrial carcinomas are more likely to be characterized by p53 mutation and ERBB-2 (HER-2/neu) expression, and less commonly associated with E-cadherin and widespread aneuploidy [17–21]. However, there is some discrepancy in gene alterations report between two types of endometrial carcinomas (i.e., BUB1, CCNB2, MYC, STK15, etc.) [22]. Wong et al. performed an integrated, genome-wide analysis of gene expression in endometrioid adenocarcinomas and compared with normal endometrium controls. Supervised analysis identified 15 genes significantly upregulated and 132 genes downregulated in endometrial carcinoma, as compared with normal control. The gene expression profiles in endometrial carcinoma were classified in mutually dependent 6 function sets, resulting in 10 biological processes according to gene ontology. The gene ontology analysis showed that endometrial carcinogenesis underwent complete down-regulation of integrin binding and cell adhesion activity. Gene pathway analysis revealed the interaction among the genes of interest and its role in the endometrial carcinogenesis. The results from this preliminary study highlight novel molecular features of endometrioid endometrial carcinoma [23].

rwent complete down-regulation of integrin binding and cell adhesion activity. Gene pathway analysis revealed the interaction among the genes of interest and its role in the endometrial carcinogenesis. The results from this preliminary study highlight novel molecular features of endometrioid endometrial carcinoma [23]. These data indicated that distinct patterns of gene expression characterize various histological types of endometrial carcinoma. An understanding of the molecular heterogeneity could potentially lead to better individualization of treatment in the future. Although some inconsistencies between single-gene and the whole-genomic approach have been observed, gene-array studies should be useful to disentangle molecular pathways and to identify potential targets for molecular-based treatments. 3. Diagnostic Approach Endometrial carcinoma presents with abnormal uterine bleeding in 90% of patients. But other diseases could also cause abnormal uterine bleeding such as endometrial hyperplasia, and endometrial polyps. Proper treatment requires adequate preoperative work-up consisting of histopathology confirmation and imaging. The clinical approach to postmenopausal bleeding requires prompt and efficient evaluation to exclude or diagnose carcinoma.

se abnormal uterine bleeding such as endometrial hyperplasia, and endometrial polyps. Proper treatment requires adequate preoperative work-up consisting of histopathology confirmation and imaging. The clinical approach to postmenopausal bleeding requires prompt and efficient evaluation to exclude or diagnose carcinoma. One of the most convenient methods of achieving this is transvaginal ultrasound. Transvaginal ultrasonography can be useful in the triage of patients in whom endometrial sampling was performed but tissue was insufficient for diagnosis. Endometrial thickness is the most valuable parameter to prognosticate both endometrial carcinoma and any endometrial pathology (sensitivities of 90% and 89%, and specificities of 79% and 94% with optimal cutoffs of 9.6 and 7.7 mm, resp.) [24]. The majority of these studies reported that a thin (4-5 mm) endometrial measurement on transvaginal sonography can exclude malignancy in the majority of postmenopausal women with vaginal bleeding. This has a negative predictive value of 96% when the endometrial echo is ≤4 mm thick, whereas an echo >4 mm indicates the need for a biopsy [25].

studies reported that a thin (4-5 mm) endometrial measurement on transvaginal sonography can exclude malignancy in the majority of postmenopausal women with vaginal bleeding. This has a negative predictive value of 96% when the endometrial echo is ≤4 mm thick, whereas an echo >4 mm indicates the need for a biopsy [25]. When scanning demonstrates the possibility of pathology, outpatient hysteroscopy and biopsy are the gold standard for investigating the endometrial cavity [26]. Hysteroscopy, a significantly more accurate diagnostic method for the detection of endometrial pathology than transvaginal ultrasonography (TVS), has better specificity and should be considered for all patients with abnormal uterine bleeding with an endometrial thickness of more than 4 mm. For women showing abnormal or suspicious lesions, it is necessary to perform hysteroscopy with eye-directed biopsy because some cases of endometrial carcinoma are unlikely recognized by ultrasonography with an endometrial thickness less than 4 mm, the possibility of missing is 0.8% [27, 28]. It can be stated that there is a high level of concordance between findings of hysteroscopic studies and the directed endometrial biopsy [29]. But it is a pity that hysteroscopy is not warranted as a first line investigation for postmenopausal bleeding [30].

less than 4 mm, the possibility of missing is 0.8% [27, 28]. It can be stated that there is a high level of concordance between findings of hysteroscopic studies and the directed endometrial biopsy [29]. But it is a pity that hysteroscopy is not warranted as a first line investigation for postmenopausal bleeding [30]. When the diagnosis is confirmed histopathologically, imaging is recommended to identify stages of the disease radiologically prior to surgery. The accuracy/sensitivity/specificity of TVS, CT, and MRI in detecting deep myometrial invasion were 89%/90%/88%, respectively. The sensitivity and accuracy of MRI in detecting deep myometrial invasion were significantly higher than those of TVS and CT [25]. For diagnosis of deep myometrial infiltration, cervical invasion, or both, MRI sensitivity and specificity were 56% and 85%; 47% and 83%; and 67% and 77%, respectively. However MRI has limited value in identifying patients with endometrial carcinoma who are at risk of lymph node metastasis [31].

ose of TVS and CT [25]. For diagnosis of deep myometrial infiltration, cervical invasion, or both, MRI sensitivity and specificity were 56% and 85%; 47% and 83%; and 67% and 77%, respectively. However MRI has limited value in identifying patients with endometrial carcinoma who are at risk of lymph node metastasis [31]. Positron emission tomography (PET) is a new imaging technology in detection of subclinical nodal disease. Several investigators have demonstrated the value of PET in screening endometrial carcinomas [32, 33]. Recently, Signorelli et al. reported that patient-based sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of 18F-FDG PET/CT for detection of nodal disease were 77.8%, 100.0%, 100.0%, 93.1%, and 94.4%, respectively. Nodal lesion site-based sensitivity, specificity, positive predictive value, negative predictive value and accuracy of 18F-FDG PET/CT were 66.7%, 99.4%, 90.9%, 97.2%, and 96.8%, respectively. It seems that 18F-FDG PET/CT is an accurate method for the presurgical evaluation of pelvic nodes metastases. High negative predictive value may be useful in selecting patients who only may benefit from lymphadenectomy in order to minimize operative and surgical complications [34]. Kitajima and his colleagues compared the accuracy of integrated 18F-FDG PET/CT with intravenous contrast medium in detecting pelvic and paraaortic lymph node metastasis in patients of uterine carcinoma with surgical and histopathological findings used as the reference standard. They found that FDG-PET is only moderately sensitive in predicting lymph node metastasis preoperatively in patients with endometrial carcinoma [35]. Horowitz has similar conclusion about the sensitivity and specificity of FDG-PET for detecting pelvic and paraaortic lymph node metastasis in patients with uterine corpus carcinoma before surgical staging. The sensitivity and specificity of FDG-PET were 60% and 98%, respectively. A notable question is this imaging modality should not replace lymphadenectomy, but may be helpful for patients on whom lymphadenectomy cannot be, or was not, performed [36].

sis in patients with uterine corpus carcinoma before surgical staging. The sensitivity and specificity of FDG-PET were 60% and 98%, respectively. A notable question is this imaging modality should not replace lymphadenectomy, but may be helpful for patients on whom lymphadenectomy cannot be, or was not, performed [36]. 4. Treatment Overview The cornerstone of curative for patients with endometrial carcinoma is surgical treatment, including complete hysterectomy, removal of remaining adnexal structures, and appropriate surgical staging in patients considered at risk for extrauterine disease. During the last 10 years interest in endometrial carcinoma has increased considerably and investigations into the following areas have increased our understanding of how we could reduce the risk of acquiring the disease and how we could best use the surgical and nonsurgical treatments available to us: optimal use of adjuvant radiotherapy; effect of hormonotherapy; role of chemotherapy; effectiveness of lymphadenectomy; genetic predisposition to the disease; and influence of less common histotypes.

4. Treatment Overview The cornerstone of curative for patients with endometrial carcinoma is surgical treatment, including complete hysterectomy, removal of remaining adnexal structures, and appropriate surgical staging in patients considered at risk for extrauterine disease. During the last 10 years interest in endometrial carcinoma has increased considerably and investigations into the following areas have increased our understanding of how we could reduce the risk of acquiring the disease and how we could best use the surgical and nonsurgical treatments available to us: optimal use of adjuvant radiotherapy; effect of hormonotherapy; role of chemotherapy; effectiveness of lymphadenectomy; genetic predisposition to the disease; and influence of less common histotypes. 5. Surgical Therapy Treatment has remained relatively unchanged over the last 40 years relying principally on surgery to achieve cure. Survival is heavily dependent on surgical stage, which is determined by the classification system adopted by the FIGO in 1988. The foundation of primary treatment is hysterectomy, during which nodal assessment and surgical staging offer the opportunity for the most accurate assessment/detection of occult extrauterine malignancy in all women whose disease appears clinically confined to the uterus. Although these tenets are universally acceptable, the integration and implementation of these concepts when performing the “proper or appropriate” surgical procedure remain contested.

st accurate assessment/detection of occult extrauterine malignancy in all women whose disease appears clinically confined to the uterus. Although these tenets are universally acceptable, the integration and implementation of these concepts when performing the “proper or appropriate” surgical procedure remain contested. 6. Surgical Staging Surgical staging of endometrial carcinoma was first recommended 20 years ago by the FIGO. The development of surgical staging in the management of endometrial carcinoma has arisen over the last several decades with anticipated benefits including prognostic information, tailoring of adjuvant treatment, and a possible therapeutic effect. Twenty years later, the FIGO Committee introduced changes in the staging criteria [37]. Firstly, The FIGO Committee recognized the favorable prognosis for both the former Stage IA and IB patients and elected to merge these substages. Furthermore, the ambiguity of defining cervical invasion, based on the involvement of the cervical mucosa only, was recognized and the Committee merged the former Stage IIA with Stage I disease. Secondly, the Committee eliminated the isolated positive peritoneal cytology criterion from the new staging system presumably based on the uncertain prognostic importance of isolated positive peritoneal cytology. Thirdly, the Committee incorporated some tumor characteristics (such as positive peritoneal cytology, invasion of the adnexa or vagina, or uterine serosa) by subdividing Stage IIIC patients into 2 different risk categories based on the presence (IIIC2) or absence (IIIC1) of metastatic disease in the paraaortic area.

y. Thirdly, the Committee incorporated some tumor characteristics (such as positive peritoneal cytology, invasion of the adnexa or vagina, or uterine serosa) by subdividing Stage IIIC patients into 2 different risk categories based on the presence (IIIC2) or absence (IIIC1) of metastatic disease in the paraaortic area. Recently, Mariani et al. focused on the examination of paraaortic metastases relative to the inferior mesenteric artery (IMA), and found that 77% of patients with paraaortic node involvement had metastases above the IMA, whereas nodes in the ipsilateral paraaortic area below the IMA and ipsilateral common iliac basin were declared negative in 60% and 71%, respectively. In 25 patients with paraaortic node metastases which gonadal veins were excised, 28% patients had documented metastatic involvement of gonadal veins or surrounding soft tissue. These data indicates the need for systematic pelvic and paraaortic lymphadenectomy up to the renal vessels including consideration of excision of the gonadal veins [38]. 7. Lymphadenectomy As will be referred to shortly, there has been a vigorous debate about the benefits of pelvic (plus or minus paraaortic) lymphadenectomy. Although the assessment of the pelvic and paraaortic lymph nodes has been recommended since 1988, FIGO failed to define either the anatomical extent of the lymphadenectomy or the number of lymph nodes harvested to be considered adequate for the assessment of pelvic and paraaortic node basins. This question is further complicated when people try to assess the adequacy of lymphadenectomy that was performed.

nce 1988, FIGO failed to define either the anatomical extent of the lymphadenectomy or the number of lymph nodes harvested to be considered adequate for the assessment of pelvic and paraaortic node basins. This question is further complicated when people try to assess the adequacy of lymphadenectomy that was performed. There is also lack of consensus on the extent of surgical staging in endometrial carcinoma. Some authors suggest performing complete pelvic and paraaortic lymphadenectomy on all endometrial carcinoma patients because positive lymph nodes (including isolated paraaortic lymph nodes) are common in all grades [39]. It is reported that the carcinoma related survival and the recurrence free survival were better with standard surgery plus lymphadenectomy than with with adjuvant radiotherapy in treating the endometrioid adenocarcinoma type at high risk [40]. Other studies have assessed readily discernible parameters intraoperatively to identify patients having an extremely low probability of lymphatic spread in order to minimize under- and over-treatment [38, 41]. A recent report by Mariani et al. showed that sixty-three (22%) of 281 patients undergoing lymphadenectomy had lymph node metastases: both pelvic and paraaortic in 51%, only pelvic in 33%, and isolated to the paraaortic area in 16%. Furthermore, 77% of patients with paraaortic node involvement had metastases above the inferior mesenteric artery. Conversely, lymphadenectomy does not benefit patients with grade 1 and 2 endometrioid lesions with myometrial invasion ≤50% and primary tumor diameter ≤2 cm [38]. In the most recently published prospective randomized trials that aimed to test the therapeutic benefit of lymphadenectomy, Benedetti Panici reported that both early and late postoperative complications occurred more frequently in patients who had received pelvic systematic lymphadenectomy. Although systematic pelvic lymphadenectomy statistically significantly improved surgical staging, it did not increase disease-free or overall survival rate [42].

Panici reported that both early and late postoperative complications occurred more frequently in patients who had received pelvic systematic lymphadenectomy. Although systematic pelvic lymphadenectomy statistically significantly improved surgical staging, it did not increase disease-free or overall survival rate [42]. Researchers are concerned about the results of ASTEC surgical trial that showed no evidence of benefit in terms of overall or recurrence-free survival for pelvic lymphadenectomy in women of early endometrial carcinoma [43]. However, Amant et al. argued that there are several reasons why the ASTEC trial did not show improved overall survival with routine lymphadenectomy [44]. First, the number of lymph nodes resected was insufficient in many patients. Second, the high rate of inclusion of low-risk patients and the low number of lymph nodes removed are the reasons for the low rate of involved lymph nodes seen in the lymphadenectomy group. Third, the study group did not assess the paraaortic nodes. Fourth, the ASTEC trial was too small to detect an overall survival difference because the expected proportion of isolated pelvic lymph-node recurrences is as low as 2-3% in early endometrial carcinoma.

f involved lymph nodes seen in the lymphadenectomy group. Third, the study group did not assess the paraaortic nodes. Fourth, the ASTEC trial was too small to detect an overall survival difference because the expected proportion of isolated pelvic lymph-node recurrences is as low as 2-3% in early endometrial carcinoma. Without clear standard recommendations, surgical staging will continue to be a confusing topic, with no appropriate quality control. There are still many unanswered questions. Are there a critical minimum number of nodes that should be resected? Do the paraaortic nodes always need to be resected? Should the histologic type of uterine carcinoma determine the extent of lymphadenectomy? Do the modern robotic-assisted or laparoscopic approaches provide surgeons adequate exposure to perform sufficient lymphadenectomy? The ideal surgical staging for endometrial carcinoma remains a subject of active debate. We are hoping that more prospective randomized trials will solve them.

e extent of lymphadenectomy? Do the modern robotic-assisted or laparoscopic approaches provide surgeons adequate exposure to perform sufficient lymphadenectomy? The ideal surgical staging for endometrial carcinoma remains a subject of active debate. We are hoping that more prospective randomized trials will solve them. 8. Laparoscopy The standard surgical surgery of endometrial carcinomas includes total hysterectomy, bilateral salpingo-oophorectomy, and lymphoidectomy. However, the application of laparoscopy in the management of gynecologic malignancy has received much attention and given rise to considerable debate. During the past few years, several investigators have demonstrated that total or vaginally-assisted laparoscopic hysterectomy, associated with laparoscopic pelvic lymphadenectomy, represents a valid alternative to open surgery [45–48]. The potential health gain of performing a laparoscopic hysterectomy instead of an abdominal hysterectomy in patients with early stage endometrial carcinoma is expected in lower rate of intraoperative complications, less blood loss; lower transfusion rate and haemoglobin decrease, shorter hospital stay, as well as a faster return of bowel activity and quicker return to activities in daily life. Nevertheless, laparoscopic hysterectomy does not seem to modify the disease-free survival and the overall survival, laparoscopic approach is an effective procedure for treating early stage endometrial carcinoma [49–51]. Randomized trials and long-term follow-up at various medical centers are necessary to evaluate the overall oncologic outcomes of this procedure.

not seem to modify the disease-free survival and the overall survival, laparoscopic approach is an effective procedure for treating early stage endometrial carcinoma [49–51]. Randomized trials and long-term follow-up at various medical centers are necessary to evaluate the overall oncologic outcomes of this procedure. 9. Radiotherapy Due to the difficulty in detecting cervical involvement preoperatively, treatment paradigms for stage II endometrial carcinoma often call for adjuvant radiotherapy postoperatively [52]. The debate regarding whether postoperative radiotherapy could improve survival has been fueled by multiple retrospective studies which have presented conflicting conclusions.

cal involvement preoperatively, treatment paradigms for stage II endometrial carcinoma often call for adjuvant radiotherapy postoperatively [52]. The debate regarding whether postoperative radiotherapy could improve survival has been fueled by multiple retrospective studies which have presented conflicting conclusions. Several studies suggested that survival rate increases if a surgery is performed in conjunction with adjuvant pelvic radiotherapy, external beam radiotherapy (EBRT) or brachytherapy (BT). For high-risk disease, the standard care has always been pelvic radiotherapy. Clearly, there are advantages as shown in meta-analyses and by the Cochrane group [53, 54]. In the Gynecologic Oncology Group's prospective evaluation of adjuvant radiation, which included patients with occult stage II tumors, radiation decreased the risk of pelvic recurrence [55]. In a report of 162 stage II endometrial carcinoma patients, Cohn et al. noted that the 5-year disease-free survival was improved (94% versus 76%) in patients who underwent radical hysterectomy [52]. Likewise, the studies by Rossi teams came to similar conclusions. They found that women with Stage IIIC endometrial carcinoma receiving adjuvant EBRT and EBRT/BT had improved overall survival compared with patients receiving no additional radiotherapy. When direct extension of the primary tumor was present, the addition of BT to EBRT was even more beneficial [56]. Up to date, Wright and his colleagues examined 1577 women with stage II endometrial adenocarcinoma and analyzed the role of radical hysterectomy and radiation in management of endometrial adenocarcinoma. They found that women who did not receive radiation were 48% more likely to die from their tumors. The benefit of adjuvant radiation is most pronounced in women with high-risk pathologic features who underwent radical hysterectomy [57].

of radical hysterectomy and radiation in management of endometrial adenocarcinoma. They found that women who did not receive radiation were 48% more likely to die from their tumors. The benefit of adjuvant radiation is most pronounced in women with high-risk pathologic features who underwent radical hysterectomy [57]. In contrast, other investigators have been unable to show a survival benefit based on the type of surgical procedure performed. In the paper by Kong, there is undoubtedly a benefit in local control when adjuvant pelvic radiotherapy is given but again no survival advantage. This is further supported by a presentation at ECCO 2007 from Cornes and Johnson in which they showed that there is up to a 10% survival advantage for patients with IC G3 tumors treated with pelvic radiotherapy [58]. They have also shown that for low-risk patients adjuvant EBRT is probably detrimental whilst for intermediate-risk patients although there may be a small benefit for some patients, this is offset by additional morbidity leading to an overall neutral effect. There are also two papers looking at data from the Survival, Epidemiology, and End Results (SEER) database [59, 60]. Both Lee et al. and Chan et al. analyzed the SEER data and showed that patients with high-grade IC G3 tumors appeared to benefit but failed to show any benefit to other patients. The data from a prospective, multicenter randomized trial of 645 evaluable low-risk endometrial carcinoma patients was showed that the impact of postoperative brachytherapy on even the locoregional recurrence rate seems to be limited in patients with low-risk endometrial carcinoma. The overall recurrence rate and survival were similar in postoperative vaginal irradiation and surgery alone groups [61].

endometrial carcinoma patients was showed that the impact of postoperative brachytherapy on even the locoregional recurrence rate seems to be limited in patients with low-risk endometrial carcinoma. The overall recurrence rate and survival were similar in postoperative vaginal irradiation and surgery alone groups [61]. The fresh data of ASTEC/EN.5 randomized trials was published recently. There was no evidence that overall survival with external beam radiotherapy was better than observation. Combined data from ASTEC and EN.5 in a meta-analysis of trials confirmed that there was no benefit in terms of overall survival (hazard ratio 1.04; 95% CI 0.84–1.29) and can reliably exclude an absolute benefit of external beam radiotherapy at 5 years of more than 3%. Interpretation adjuvant external beam radiotherapy cannot be recommended as part of routine treatment for women with intermediate-risk or high-risk early-stage endometrial carcinoma with the aim of improving survival [62]. Meanwhile, we should notice that adjuvant external beam radiotherapy did result in a small reduction in isolated local recurrence, but this analysis only included women who had local recurrence alone, ignoring 65% of women who had local and distant recurrence at the same time, or distant recurrence alone. The small reduction in isolated local recurrence does not translate into an effect on overall or recurrence-free survival.

ted local recurrence, but this analysis only included women who had local recurrence alone, ignoring 65% of women who had local and distant recurrence at the same time, or distant recurrence alone. The small reduction in isolated local recurrence does not translate into an effect on overall or recurrence-free survival. Up to this point, it was emerging that patients with low-risk disease do not need any adjuvant treatment and can be treated by surgery and careful follow up. Patients with intermediate-risk disease are more problematic and may still be treated with external beam radiotherapy. Although the majority of retrospective data has not demonstrated a benefit for radiation, it has been suggested that women who undergo simple hysterectomy and are found with cervical disease may benefit from radiotherapy [63]. Feltmate et al. reported excellent outcomes in a series of 65 patients with stage II endometrial carcinoma, the majority of whom were treated surgically and followed by adjuvant radiation. In their cohort, 5-year disease-specific survival was 93% with recurrences in 15% [64]. Among 203 subjects with endometrial carcinoma, Sartori et al. noted a statistically significant improvement in 5-year survival from 74% with simple hysterectomy to 94% with a radical procedure [65].

d followed by adjuvant radiation. In their cohort, 5-year disease-specific survival was 93% with recurrences in 15% [64]. Among 203 subjects with endometrial carcinoma, Sartori et al. noted a statistically significant improvement in 5-year survival from 74% with simple hysterectomy to 94% with a radical procedure [65]. Some clinical trials investigated the optimal of radiotherapy mode. It is considered that brachytherapy is a more convenient treatment than external beam radiotherapy and might be associated with less toxicity. In the PORTEC1-trial, the 5-year risk of vaginal and pelvic recurrence for high- intermediate risk patients was 19% without further treatment, compared to 5% after EBRT. Since most recurrences were located in the upper vagina, Phase II trials suggested vaginal brachytherapy (VBT) to be as effective as EBRT. PORTEC-2 is the first randomized trial comparing the efficacy of VBT and EBRT to determine which treatment provides optimal local control with best quality of life. The data suggested that vaginal brachytherapy is effective in preventing vaginal recurrence. Despite the slightly but significantly increased pelvic failure rate in the VBT arm, rates of distant metastases, OS, and RFS were similar. As indicated by the patient survey on quality of life after treatment, VBT was shown to be better than EBRT, VBT should be the treatment of choice for patients with high-intermediate risk endometrial carcinoma [66]. First results of the randomized PORTEC-2 trial are evaluation about quality-of-life (QOL) after pelvic radiotherapy or vaginal brachytherapy for endometrial carcinoma. Patients in the VBT group reported better social functioning (P < .002) and lower symptom scores for diarrhea, fecal leakage, the need to stay close to the toilet, and limitation in daily activities because of bowel symptoms (P < .001). Vaginal brachytherapy provides a better quality of life than external-beam radiotherapy for Endometrial Carcinoma, and should be the preferred treatment from a quality of life perspective [67].

fecal leakage, the need to stay close to the toilet, and limitation in daily activities because of bowel symptoms (P < .001). Vaginal brachytherapy provides a better quality of life than external-beam radiotherapy for Endometrial Carcinoma, and should be the preferred treatment from a quality of life perspective [67]. Nevertheless, the data is important and add to our understanding of the optimal management of endometrial carcinoma. Data from these data banks and the Cochrane reviews may help to address the question of which is the optimal treatment for this group. A further approach is to withhold radiation in the intermediate-risk group and offer careful surveillance and use salvage radiotherapy for relapses confined to vagina or vault. In the meantime we should consider that either immediate external beam radiotherapy or a watch and see policy with salvage radiation should be the standard approach.

to withhold radiation in the intermediate-risk group and offer careful surveillance and use salvage radiotherapy for relapses confined to vagina or vault. In the meantime we should consider that either immediate external beam radiotherapy or a watch and see policy with salvage radiation should be the standard approach. 10. Chemotherapy Chemotherapy is emerging as an important treatment modality in advanced endometrial carcinoma. The use of neoadjuvant chemotherapy resulted in a high rate (80%) of optimal interval debulking surgery for the treatment of endometrial carcinoma with transperitoneal spread [68]. GOG 122 was the first randomized study to demonstrate a survival advantage with chemotherapy in advanced stage endometrial carcinoma [69]. At 60 months, 50% of patients received doxorubicin and cisplatin chemotherapy were predicted to be alive and disease-free when adjusting for stage compared with 38% of patients who had whole-abdominal irradiation. The data from GOG 122 showed that combination chemotherapy had a survival advantage over whole abdominal radiotherapy in Stage III and IV endometrial carcinoma.

cisplatin chemotherapy were predicted to be alive and disease-free when adjusting for stage compared with 38% of patients who had whole-abdominal irradiation. The data from GOG 122 showed that combination chemotherapy had a survival advantage over whole abdominal radiotherapy in Stage III and IV endometrial carcinoma. There are several studies focused on the toxicity, tolerability, and feasibility of delivering combination chemotherapy with subsequent radiation therapy to women with advanced endometrial carcinoma, and evaluate the long-term bowel toxicity. It is notable that GOG 122 study had an extremely high toxicity rate from chemotherapy (68% Grade 4 hematologic toxicity), including 8 treatment-related deaths. It is apparent that cisplatin and/or doxorubicin-based regimens are associated with unfavorable rates of toxicity [69]. A Phase I GOG study by Soper et al. indicates that treatments comprised of whole abdomenopelvic radiation with concomitant weekly cisplatin, followed by doxorubicin and cisplatin chemotherapy, had prohibitive toxicity and did not undergo further evaluation [70]. Bruzzone et al. reported a series of 45 women who received cisplatin and cyclophosphamide followed by radiotherapy, in which 10 women (22%) completed 3 cycles or less [71]. Duska et al. reported a pilot study for advanced stage disease, which included 3 cycles of paclitaxel, doxorubicin, and carboplatin, followed by radiotherapy [72]. All patients required G-CSF support, but 50% still experienced Grade 3 or 4 acute toxicity. In RTOG 9708, in which 4 cycles of cisplatin and paclitaxel were administered after completion of radiotherapy, acute Grade 3/4 toxicity was greater than 80% [73]. In comparison, Lupe et al. used the combined modality protocol comprised of carboplatin and paclitaxel with involved field radiotherapy had a much lower acute toxicity rate, and the compliance rate was very high [74].

ere administered after completion of radiotherapy, acute Grade 3/4 toxicity was greater than 80% [73]. In comparison, Lupe et al. used the combined modality protocol comprised of carboplatin and paclitaxel with involved field radiotherapy had a much lower acute toxicity rate, and the compliance rate was very high [74]. Meanwhile, the use of chemotherapy alone has been associated with high rates of pelvic relapse, ranging from 18% to 47% [69, 75]. Recently, Takeshima et al. reported with postoperative adjuvant chemotherapy, recurrences occurred predominantly at distant sites in the absence of pelvic radiation in surgically staged grade 3 endometrial carcinoma. Estimated 5-year disease-free survival rates were 89.8% for patients with surgical stage I-II disease, 78.6% for those with surgical stage III disease, and 87.3% overall [76]. There is an emerging consensus that chemotherapy may be insufficient for reducing the risk of pelvic relapse although it appears to be an important component of treatment. Sovak et al. reported a pelvic relapse rate of 44% in patients with Stage III and IV disease who received 6 cycles of adjuvant carboplatin and paclitaxel, of whom only 5 (10%) also received adjuvant pelvic radiotherapy [77]. In RTOG 9708, the pelvic relapse rate was only 2%. It suggested that the addition of radiation to chemotherapy does appear to be associated with a lower rate of pelvic relapse [73]. However, in that study, 23% had Stage I and 16% had Stage II disease. The low rate of pelvic relapse may be partly attributed to the more favorable stage distribution. Alvarez Secord et al. published a large retrospective study of 356 Stage III and IV patients treated with radiation alone (48%), chemotherapy alone (29%), and combined modality therapy (23%) [78]. After adjusting for stage, age, grade, and debulking status, the hazard ratios (HR) for overall survival were 1.6 (95% CI 0.88–2.89) and 2.0 (95% CI 1.17–3.48) for chemotherapy and radiation alone, respectively, compared to combined modality therapy. Matsuura et al. reported most recently that combined treatment with radiotherapy/chemotherapy was associated with a better survival rate than chemotherapy alone (78% versus 62%, resp.). In Stage IIIc endometrial carcinoma, the combined use of radiotherapy and chemotherapy could reduce pelvic recurrence (33.3% and 7.1%, resp.) and was associated with a better survival rate than chemotherapy alone (78% versus 62%, resp.) [79].

with a better survival rate than chemotherapy alone (78% versus 62%, resp.). In Stage IIIc endometrial carcinoma, the combined use of radiotherapy and chemotherapy could reduce pelvic recurrence (33.3% and 7.1%, resp.) and was associated with a better survival rate than chemotherapy alone (78% versus 62%, resp.) [79]. Based on this concurrent carboplatin/paclitaxel and intravaginal radiation in surgical stage I-II serous endometrial carcinoma study, surgical staging followed by involved-field radiotherapy and carboplatin/paclitaxel is well tolerated and effective in stage I-II serous endometrial carcinoma [80]. Confirmation of these results on a larger number of patients with longer follow-up is still needed.

surgical stage I-II serous endometrial carcinoma study, surgical staging followed by involved-field radiotherapy and carboplatin/paclitaxel is well tolerated and effective in stage I-II serous endometrial carcinoma [80]. Confirmation of these results on a larger number of patients with longer follow-up is still needed. What is the optimized chemotherapy regimen is still a subject of debate. Historically, the treatments used have been a combination of a platinum and anthracycline, usually cisplatin and doxorubicin (AP), but this can be quite a toxic regime and is often poorly tolerated, therefore it is not ideal for combining with radiation therapy. Adding paclitaxel (TAP) usually needs growth factors to support the administration. Hellenic Co-operative Oncology Group (HeCOG) studied the drug regimen comprised paclitaxel, topotecan, and carboplatin in metastatic endometrial carcinoma. With G-CSF support, the drug regimen appears active with acceptable toxicity in patients with metastatic or recurrent carcinoma of the endometrium [81]. In relapsed disease, the GOG are currently evaluating TAP versus TC [82]. In addition, the optimal regimen remains to be defined as all of them (doxorubicin/cisplatin-AP, cyclophosphamide/doxorubicin/cisplatin-CAP, paclitaxel/carboplatin-TC, and paclitaxel/doxorubicin/ cisplatin-TAP) cause significant toxicity. Although randomized evidence is limited, the combination of carboplatin and paclitaxel has been commonly used in advanced endometrial carcinoma because of its manageable toxicity and excellent response rates (64–78%) [77, 83–88]. McMeekin et al. studied the maximum tolerated dose and feasibility of weekly cisplatin and paclitaxel chemotherapy administered concurrently with whole abdominal radiation therapy in women with high-risk endometrial carcinoma. A regimen of cisplatin 25 mg/m2 and paclitaxel 20 mg/m2 weekly with whole abdominal radiation therapy was determined to be feasible, but is associated with moderate acute and chronic gastrointestinal toxicity [89].

stered concurrently with whole abdominal radiation therapy in women with high-risk endometrial carcinoma. A regimen of cisplatin 25 mg/m2 and paclitaxel 20 mg/m2 weekly with whole abdominal radiation therapy was determined to be feasible, but is associated with moderate acute and chronic gastrointestinal toxicity [89]. Further investigations are required to define the subgroup of patients who benefit from postoperative adjuvant chemotherapy. Two randomized clinical trials are in progress in order to obtain available evidence which can help clinicians make wise decisions on treatment options, such as adjuvant chemotherapy of patients with high-risk stage I and II, as well as stage IIIA endometrial carcinoma. GOG 209 is an ongoing study randomizing women with Stage III or IV endometrial carcinoma to either doxorubicin, cisplatin, paclitaxelwith G-CSF, or carboplatin and paclitaxel. Additionally, PORTEC 3 is an ongoing randomized Phase III trial comparing concurrent chemoradiation and adjuvant chemotherapy (4 cycles of carboplatin and paclitaxel) versus pelvic radiation alone in high risk and advanced stage disease. This study is timely and necessary to determine whether radiotherapy or chemotherapy improves overall survival and failure-free survival, compare the rates of severe (grades 3 and 4) treatment-related toxicity, pelvic and distant recurrence, and evaluate quality of life of patients with high-risk and advanced stage endometrial carcinoma.

y and necessary to determine whether radiotherapy or chemotherapy improves overall survival and failure-free survival, compare the rates of severe (grades 3 and 4) treatment-related toxicity, pelvic and distant recurrence, and evaluate quality of life of patients with high-risk and advanced stage endometrial carcinoma. 11. Functional Preservation Although the median age of patients with endometrial carcinoma is in the early 60s, approximately 5% of patients are younger than age 40 when diagnosed. In the presence of early staged endometrial carcinoma, most have favorable outcomes, thus their quality of life after treatment is as important a consideration as a cure of carcinoma. This issue is especially imperative when endometrial carcinoma is encountered in younger or reproductive ages when the afflicted woman has not achieved her fertility function. Despite being a critical issue, there are only a few studies with definite treatment guidelines or any evidence-based recommendations concerning conservative treatment for endometrial carcinoma.

al carcinoma is encountered in younger or reproductive ages when the afflicted woman has not achieved her fertility function. Despite being a critical issue, there are only a few studies with definite treatment guidelines or any evidence-based recommendations concerning conservative treatment for endometrial carcinoma. Since early 1980s, there have been several reports on conservative treatment with progestins for early-stage endometrial carcinoma in young women. Most of them were small series and retrospective studies from single institutions. Response rates and recurrence rates varied (i.e., the response rate for endometrial carcinoma and atypical endometrial hyperplasia ranged from 57% to 76% and from 83% to 92%, respectively, and the recurrence rate ranged from 11% to 50%) [90–96]. Such variations were probably due to the differences in drugs used, dosage, and duration of treatment. Daily doses of megestrol acetate ranged between 10 and 400 mg, and that of medroxyprogesterone acetate (MPA) ranged between 200 mg and 800 mg. Nevertheless, there have been no prospective trials to investigate the optimal dosage, duration of treatment, curative rate of MPA treatment, or pregnancy rate after this therapy in young women with endometrial carcinoma and atypical endometrial hyperplasia. Therefore, Ushijima et al. conducted a multicenter, prospective phase II study on MPA treatment. Their prospective study conducted to clarify the accurate complete response (CR) rate of treatment with MPA at a fixed dose of 600 mg/d for 26 weeks, has demonstrated that the CR rate for endometrial carcinoma and atypical endometrial hyperplasia was 55% and 82%, respectively, and the recurrence rate was 57% and 38%, respectively [97]. In another prospective multicentric prospective study, Ushijima et al. evaluated the efficacy of fertility-sparing treatment by MPA for endometrial carcinoma and atypical endometrial hyperplasia. Complete response was found in 44% in endometrial carcinoma and 82% in atypical endometrial hyperplasia. 9 pregnancies and 4 normal deliveries have been recorded after MPA therapy. Twelve recurrences were found in 30 complete response patients (40%) between 7 to 22 months. Data showed that even in the complete response patients, close follow-up is required because of their high recurrence rate [98]. Recently, Signorelli et al.

gnancies and 4 normal deliveries have been recorded after MPA therapy. Twelve recurrences were found in 30 complete response patients (40%) between 7 to 22 months. Data showed that even in the complete response patients, close follow-up is required because of their high recurrence rate [98]. Recently, Signorelli et al. conducted a prospective study of conservative treatment in 21 young nulliparous women with grade G1 endometrial carcinoma stage IA or atypical complex hyperplasia. All were treated with a low-dose cyclic natural progestin therapy (200 mg/day from day 14–25) and encouraged to attempt pregnancy immediately. Overall response rate to progestin therapy was 57%. Nine women conceived spontaneously (43%) and 8 women with persistent disease or partial response to hormonal treatment. Three additional complete responses were observed after delivery [99].

(200 mg/day from day 14–25) and encouraged to attempt pregnancy immediately. Overall response rate to progestin therapy was 57%. Nine women conceived spontaneously (43%) and 8 women with persistent disease or partial response to hormonal treatment. Three additional complete responses were observed after delivery [99]. A largely unanswered question is the safety of ovarian preservation in young women with endometrial carcinoma. First, estrogen production from the ovaries may stimulate microscopic foci of residual endometrial carcinoma. Although vitro data [100] has suggested that estrogen stimulates the growth of endometrial carcinoma cells and upregulates the expression of estrogen receptors, this concern has not been observed clinically so far. Several reports examined the use of estrogen replacement therapy in postmenopausal women with endometrial carcinoma. Yet, these studies have not demonstrated any increase in the risk of recurrence or death in women receiving estrogen replacement [101–103]. The new data published by Korean Gynecologic Oncology Group (KGOG) in 2009 suggest that ovarian preservation does not adversely impact the recurrence of early stage endometrial carcinoma [103]. The most influential report was a prospective trial of estrogen replacement therapy in more than 1,200 women with endometrial carcinoma conducted by the Gynecologic Oncology Group. Although the prospective trial was closed early, the absolute recurrence rate was only 2.1% (HR 1.27; 95% CI, and 0.92 to 1.77) [101]. The findings from these studies, as well as the data from Wright group, suggest that the risk of estrogenic stimulation of residual endometrial carcinoma is quite low, particularly in women with early-stage, low-risk lesions. The second potential risk of ovarian conservation is the presence of a coexisting synchronous primary tumor within the ovaries. Synchronous primary tumors of the endometrium and ovary are reported in approximately 5% of women with endometrial carcinoma [104]. However, among young women with endometrial carcinoma, the incidence of coexisting ovarian tumors is increased and has been reported with a range from 5% to 29% [104–107].

hin the ovaries. Synchronous primary tumors of the endometrium and ovary are reported in approximately 5% of women with endometrial carcinoma [104]. However, among young women with endometrial carcinoma, the incidence of coexisting ovarian tumors is increased and has been reported with a range from 5% to 29% [104–107]. Although many studies have examined the risk of ovarian metastases in young women with endometrial carcinoma, there are no data to describe the safety of ovarian conservation. In 2009, Wright firstly reported that ovarian preservation is safe in young women with early-stage, low-grade endometrial carcinoma [108]. Their findings are notable in that ovarian preservation in premenopausal women with early-stage, low-grade endometrial carcinoma may be safe and not associated with an increased risk of carcinoma related mortality. Although the survival estimates suggest that ovarian conservation does not negatively impact outcome, it should be recognized that ovarian preservation may be associated with a two-fold or greater increase in mortality. Given the potential consequences of surgical menopause, further research to examine the safety of ovarian conservation for young women with early-stage endometrial carcinoma is clearly warranted. At present, the long-term risks and benefits of ovarian preservation should be carefully discussed with young women with endometrial carcinoma before hysterectomy.

gical menopause, further research to examine the safety of ovarian conservation for young women with early-stage endometrial carcinoma is clearly warranted. At present, the long-term risks and benefits of ovarian preservation should be carefully discussed with young women with endometrial carcinoma before hysterectomy. 12. Fertility Sparing Although there is no known fertility-sparing surgical option for women with endometrial carcinoma, selected young patients of childbearing age with apparent early endometrial carcinoma who wish to preserve fertility may consider treatment with progestin therapy rather than surgery. If such treatment is contemplated, it is recommended that a thorough hysteroscopy and curettage be performed to rule out a worse lesion prior to initiation. A review of the literature indicates 101 patients with a median age of 29 years who were treated with progestin therapy rather than definitive surgery subsequently had 56 children [91]. Additionally, Gershenson et al. provided indirect evidence to support the recent concept of using the fertility-sparing or conservative surgery or therapy for malignancies in women that the use of conservative modalities can be applied in the management of endometrial carcinomas because there are many reports showing that endometrial carcinomas can be treated with a simple diagnostic dilatation and curettage followed by some potent hormone therapy, including a progestin agent, in highly selected young women who would like to preserve their fertility potential [109].

t of endometrial carcinomas because there are many reports showing that endometrial carcinomas can be treated with a simple diagnostic dilatation and curettage followed by some potent hormone therapy, including a progestin agent, in highly selected young women who would like to preserve their fertility potential [109]. Recently, there have been a number of reports of women with uterine endometrial carcinoma who became pregnant and gave birth after the administration of medroxyprogesterone acetate (MPA) [93, 110–114]. Meanwhile subsequently assisted reproductive techniques such as transfer of embryos with intracytoplasmic sperm injection (ICSI) and preimplantation genetic diagnosis (PGD) may be valuable to achieve immediate pregnancy [115–117].

d gave birth after the administration of medroxyprogesterone acetate (MPA) [93, 110–114]. Meanwhile subsequently assisted reproductive techniques such as transfer of embryos with intracytoplasmic sperm injection (ICSI) and preimplantation genetic diagnosis (PGD) may be valuable to achieve immediate pregnancy [115–117]. Gadducci et al. reviewed the related literature and confirmed that approximately three fourths of the women achieve a histologically documented complete response, with a mean response time of 12 weeks, but about one third of these subsequently developed a recurrence after a mean time of 20 months. Following high dose progestin therapy and confirmation of the regression of carcinoma, the patient might attempt to conceive spontaneously. However, assisted reproduction techniques might increase the likelihood of pregnancy and decrease the time interval to conception. Several successful pregnancies have occurred after a fertility-sparing treatment of endometrial atypical hyperplasia or endometrial carcinoma, more frequently with assisted reproductive technologies. The implementation of in vitro fertilisation techniques not only increases the chance of conception, but it may also decrease the interval to conception [118]. However, despite the achievements of these studies on fertility-sparing treatments, there are no definite treatment guidelines or any evidence-based recommendations and many questions still remain unanswered regarding the selection of patients. Nevertheless, the optimal dose or duration and curative rate of MPA therapy in endometrial carcinoma and atypical endometrial hyperplasia in young women are still uncertain.

efinite treatment guidelines or any evidence-based recommendations and many questions still remain unanswered regarding the selection of patients. Nevertheless, the optimal dose or duration and curative rate of MPA therapy in endometrial carcinoma and atypical endometrial hyperplasia in young women are still uncertain. It is vital to choose appropriate patients with endometrial carcinoma to adopt ovarian preservation and fertility-sparing treatment. The best candidates for progestin therapy are women who have a relative hyperestrogenic state, which is thought to cause the malignancy. Indeed, some patients would not chose fertility-sparing treatment given the lack of data on oncologic safety. Fertility-sparing treatments are successfully used; however, these treatments can be offered only to a limited number of patients which meet the pathologic criteria for a conservative approach [119]. The indications for conservative treatment include the patient's desire to preserve fertility, no medical history of thrombosis, and no abnormal levels of hemostasis, a histologic diagnosis of grade 1 endometrioid adenocarcinoma by total endometrial curettage, no myometrial invasion or extrauterine spread of the disease observed by MRI, and hysteroscopy and total endometrial curettage must be repeated after 4–6 weeks of additional MPA therapy. Additionally, the expression of receptor for progesterone receptor (PR), PTEN gene, DNA mismatch repair gene MLH1 and phospho-AKT on tissue specimens may be useful for selecting patients fit for a conservative management [118, 120].

total endometrial curettage must be repeated after 4–6 weeks of additional MPA therapy. Additionally, the expression of receptor for progesterone receptor (PR), PTEN gene, DNA mismatch repair gene MLH1 and phospho-AKT on tissue specimens may be useful for selecting patients fit for a conservative management [118, 120]. The opportunity of a demolitive surgery after delivery or after childbearing being no longer required is still a debated issue. Large multicenter trials are strongly warranted to better define the selection criteria for a conservative treatment, endocrine regimen of choice, the optimal dosing, the duration of treatment and follow-up protocols. Until now, the long-term outcome of children in utero exposed to oncological treatment modalities is poorly documented. Delivery should be postponed preferably until after a gestational age of 35 weeks. Further research including international registries for gynecologic carcinoma in pregnancy is urgently needed. The gathering of both available literature and personal experience suggested models for treatment of gynecologic carcinoma in pregnancy [121]. In any case, the patient should be accurately informed about the relatively high recurrence rates after complete response to hormone treatment and expectations for pregnancy.

d. The gathering of both available literature and personal experience suggested models for treatment of gynecologic carcinoma in pregnancy [121]. In any case, the patient should be accurately informed about the relatively high recurrence rates after complete response to hormone treatment and expectations for pregnancy. 13. Biomarker and Targeted Therapy As previously stated, distinct molecular changes are associated with two subtypes, and these distinct molecular alterations also underscore prognostic differences. Therefore, active researches are enthusiastic about novel screening approaches that emerged from epigenetics, proteomics, and genomics in endometrial carcinoma. It is hopeful that the use of targeted therapies will improve the outcome for endometrial carcinoma.

ecular alterations also underscore prognostic differences. Therefore, active researches are enthusiastic about novel screening approaches that emerged from epigenetics, proteomics, and genomics in endometrial carcinoma. It is hopeful that the use of targeted therapies will improve the outcome for endometrial carcinoma. Nowadays, several novel tumor markers with increased sensitivity and specificity for endometrial carcinoma have been identified and are considered to help monitor response to therapy and to detect recurrent disease. These potential molecular biomarkers include HE4, CA125, Cyr61, p21, p53, Cathepsin-B, MMR, and ERR[alpha] progesterone receptor (PR)-A, which are estimated to contain potential value as prognostic factors for patients with endometrioid carcinoma [122–128]. Additionally, Bidus et al. reported two cell cycle checkpoint genes, CDC2, MAD2L1, and The ZIC2 zinc finger gene were associated with lymph node metastasis in endometrial carcinomas [129]. Currently, these tumor makers are utilized in this role with limited value. Further investigation in the role of biomarker for early detection of recurrent endometrial carcinoma and monitoring response to therapy is warranted. Gene expression profiling of the primary tumors in patients with endometrioid endometrial carcinomas seems promising for identifying genes associated with lymph node metastasis. Future studies should address whether the status of nodal metastasis can be determined from the expression profiles of preoperative tissue specimens.

pression profiling of the primary tumors in patients with endometrioid endometrial carcinomas seems promising for identifying genes associated with lymph node metastasis. Future studies should address whether the status of nodal metastasis can be determined from the expression profiles of preoperative tissue specimens. With the progress of advanced gene techniques, it has become possible to identify potential molecular markers of endometrial carcinoma for its diagnosis, prognosis and therapy by global gene expression profiling. It may provide a foundation for the development of new diagnostic and prognostic markers and type-specific therapies against this common female genital malignant disease. Such procedure allowed us to give shape to preliminary gene expression profile typical for neoplastic tissue and to estimate protein expression of the most significant predictors of neoplastic transformation. Comparison of obtained data with tumor grade can reveal new markers of endometrial carcinoma useful in routine diagnostic procedures [23, 130].

e shape to preliminary gene expression profile typical for neoplastic tissue and to estimate protein expression of the most significant predictors of neoplastic transformation. Comparison of obtained data with tumor grade can reveal new markers of endometrial carcinoma useful in routine diagnostic procedures [23, 130]. Genes related to the endometrial carcinoma progression and metastasis can be identified by differential gene expression profile with cDNA microarray and high-risk endometrial carcinoma may be distinguished before surgery by hierarchical cluster analysis [131]. Similarly, the dysregulation of these miRNAs appeared to be involved in the progression of endometrial carcinoma [132]. Therefore, some researchers suggested that the cDNA and miRNAs microarray techniques may be feasible to generate gene expression profiles of endometrial carcinoma. Classification based on gene expression patterns may be more accurate than histological grade and FIGO stage classification in predicting the prognosis of tumors [133]. Further extended and functional studies of these new approaches are required to confirm the potential use of them in the endometrial carcinomas.

inoma. Classification based on gene expression patterns may be more accurate than histological grade and FIGO stage classification in predicting the prognosis of tumors [133]. Further extended and functional studies of these new approaches are required to confirm the potential use of them in the endometrial carcinomas. With the applications of the target gene therapy, some valuable research had carried in advanced endometrial carcinoma. Since the year 2000, in advanced endometrial carcinoma, the GOG has conducted phase II trials with several molecular targeting agents including imatinib (Gleevec), trastuzumab (Herceptin), and gefitinib (Iressa) as single agents with negligible evidence of activity. The GOG does have active trials of chemotherapy with a molecular targeting agent such as bevacizumab (Avastin) in GOG 218, but there are no randomized molecular targeting agent trials in advanced endometrial carcinoma [134]. Some genes related with endometrial carcinoma prognosis have become a hopeful target for therapies in endometrial carcinoma, these targeting genes include mTOR inhibitors, EGFR tyrosine kinase inhibitors (erlotinib), and monoclonal antibodies to Her-2/neu (trastuzumab) [135–138]. However, acceptance of genetic consultation and testing is surprisingly low and deserves further investigation. For example, it is hypothesized that the HER-2/neu receptor could be used for targeted therapy in recurrent endometrial carcinoma. In a clinical trial, trastuzumab was of little clinical value in two cases of recurrent type II endometrial carcinoma based on the lack of response and changes in tumor biology [139]. In another trial, a multinomial design two-stage phase II study was performed to evaluate single-agent activity of erlotinib, an orally active, selective inhibitor of EGFR tyrosine kinase activity, in women who had advanced endometrial carcinoma with recurrent or metastatic disease, and were chemotherapy naive and received up to one line of prior hormonal therapy. The data showed that erlotinib is well tolerated with an overall objective response rate of 12.5% [136]. These reports underscore the importance of reassessment of targeted treatment in endometrial carcinoma. Yet, researchers still have a long way to go in order to reach the goal of applying the targeting gene therapy in clinical practice.

erlotinib is well tolerated with an overall objective response rate of 12.5% [136]. These reports underscore the importance of reassessment of targeted treatment in endometrial carcinoma. Yet, researchers still have a long way to go in order to reach the goal of applying the targeting gene therapy in clinical practice. 14. Prevention and Surveillance In the follow-up of endometrial carcinoma patients, pain was the most common complaint in patients with recurrent disease, followed by vaginal bleeding, general malaise, loss of weight and intestinal complaints. With the evidence from randomized clinical trials we can conclude that a follow-up program in the first three years after primary treatment of endometrial carcinoma is helpful in detecting recurrent disease. In 2007, van Wijk et al., evaluated their clinical data of patients with recurrent endometrial carcinoma treated in the Erasmus Medical Centre in Rotterdam over a 20-year period [140]. He reported that patients with screen-detected recurrences had a 5-year survival rate of 62%. Patients with interval screening recurrences or recurrences detected by chance had a 5-year survival rate of 47%. Evaluating the patients with an endometrioid type of tumor separately, the 5-year survival rate for patients screen-detected recurrences is 68% and for patients with interval screening recurrences is 51% [140]. The reported median intervals to local and distant recurrent disease are consistent with those reported in the literatures [141, 142].

ith an endometrioid type of tumor separately, the 5-year survival rate for patients screen-detected recurrences is 68% and for patients with interval screening recurrences is 51% [140]. The reported median intervals to local and distant recurrent disease are consistent with those reported in the literatures [141, 142]. Tjalma et al. published an overview of 11 retrospective studies (evaluating 2866 patients) on routine follow-up of endometrial carcinoma. In these studies symptomatic recurrences ranged between 41% and 81% (mean 65%) of all recurrences [143]. Retrospective data from both Agboola group and Tjalma group suggest that there is no difference in survival between symptomatic and asymptomatic recurrences, or between women with recurrences detected during routine follow up visits and those with recurrences detected during the interval between routine visits [143, 144]. Furthermore, there is no economic or clinical justification for the routine use of the Pap smear and systematic radiography in the follow-up of patients with endometrial carcinoma [144, 145]. Centers advocating surveillance should focus on the detection of potentially curable vaginal recurrences, since isolated vaginal-vault recurrence of endometrial carcinoma is curable in up to 87% of cases, in patients previously not exposed to radiation [146].

follow-up of patients with endometrial carcinoma [144, 145]. Centers advocating surveillance should focus on the detection of potentially curable vaginal recurrences, since isolated vaginal-vault recurrence of endometrial carcinoma is curable in up to 87% of cases, in patients previously not exposed to radiation [146]. Tjalma et al. pointed out that because of a difference in survival between isolated vaginal recurrence and nonvaginal recurrences, 5-year survival, respectively 50% and 6%, it is important to identify isolated vaginal recurrences early. As the sensitivity of routine follow-up schemes appears very low, tailored follow-up protocols based on high risk and low risk for recurrence are suggested [143]. Low risk patients are generally defined as patients with adenocarcinoma IA grade 1 or 2 or IB grade 1, with a recurrence rate of just under 4%, whereas high risk patients have a recurrence rate of around 23% [147]. Salvesen et al. found a low risk group, with FIGO Stage IA/IB or patient age below 60 years at primary operation was identified in multivariate recurrence-free survival analysis. No asymptomatic recurrences were found in this group. Therefore, they concluded that low risk patients should be considered for less frequent follow-up [141]. However, van Wijk et al. reported of five low risk patients with recurrent disease, only one patient, suffering from distant recurrent disease, was symptomatic. Without a follow-up program for patients with low-risk endometrial carcinoma, recurrent disease would only have been detected after symptoms had developed in four of these five patients. It was discussed that there is no reason to use different follow-up scheme for these patients, despite our low number of patients with low risk disease. Improving patient education so that early symptoms of recurrence are reported appears eminently sensible, but may serve also to heighten anxiety amongst the majority who will never develop recurrent disease [142]. For patients who have evidence of metastatic disease at time of surgery, it is nowadays generally accepted that there is a survival benefit to be gained if all gross evidence of disease can be resected or at least debulked to leave small-volume residual disease [148].

ity who will never develop recurrent disease [142]. For patients who have evidence of metastatic disease at time of surgery, it is nowadays generally accepted that there is a survival benefit to be gained if all gross evidence of disease can be resected or at least debulked to leave small-volume residual disease [148]. Most endometrial carcinomas are sporadic, but approximately 10% of cases have a hereditary basis [149–153]. Two genetic models have been suggested in the development of endometrial carcinoma: hereditary nonpolyposis colorectal carcinoma (HNPCC) syndrome, also known as Lynch II syndrome, and a predisposition for endometrial carcinoma alone. Both are autosomal dominant inherited carcinoma susceptibility syndrome caused by a germline mutation in one of the deoxyribonucleic acid (DNA) mismatch repair genes [153]. It is associated with early onset of carcinoma (age younger than 50 years) and the development of multiple carcinoma types, particularly colon and endometrial carcinoma. Women with Lynch syndrome have a 40-60% risk of endometrial carcinoma, which equals or exceeds their risk of colorectal carcinoma. In addition, they have a 12% risk of ovarian carcinoma. Despite limited information on the efficacy of surveillance in reducing endometrial and ovarian carcinoma risk in women with Lynch syndrome, the current gynecologic carcinoma screening guidelines include annual endometrial sampling and transvaginal ultrasonography beginning at age 30–35 years [154]. But the cost effectiveness of this screening has not been proven either. An alternative approach is primary prevention by using a progestogen device in utero, such as the Mirena IUCD. This merits full evaluation [155].

lude annual endometrial sampling and transvaginal ultrasonography beginning at age 30–35 years [154]. But the cost effectiveness of this screening has not been proven either. An alternative approach is primary prevention by using a progestogen device in utero, such as the Mirena IUCD. This merits full evaluation [155]. In addition, risk-reducing surgery consisting of prophylactic hysterectomy and bilateral salpingooophorectomy should be offered to women aged 35 years or older who do not wish to preserve their fertility [154]. Schmeler et al. reported a retrospective analysis of women with known germline mutations associated with Lynch syndrome. Sixty-one participants underwent prophylactic hysterectomy and were compared to over 200 matched controls with similar mutations that did not have preventive surgery. Endometrial carcinoma was eventually diagnosed in 33% of the controls with no cases in the prophylactic surgery group [156]. Pistorius et al. report detected asymptomatic muscle invasive endometrial carcinoma in two of four women who underwent prophylactic hysterectomy after requiring surgery for Lynch syndrome related colorectal carcinoma [157]. In 2006 a multiinstitutional, matched case-control study found that prophylactic hysterectomy with bilateral salpingo-oophorectomy is an effective primary preventive strategy in women with HNPCC syndrome [156]. Based on these observations, surgery as primary prevention for women at high risk due to known germline lesions or history of Lynch syndrome related malignancies may yield a meaningful reduction in progression to endometrial carcinoma.

tomy is an effective primary preventive strategy in women with HNPCC syndrome [156]. Based on these observations, surgery as primary prevention for women at high risk due to known germline lesions or history of Lynch syndrome related malignancies may yield a meaningful reduction in progression to endometrial carcinoma. 15. Summary Endometrial carcinoma is a low-grade curable malignancy and most patients who present with early disease have excellent survival rate. Endometrial carcinoma remains a management challenge, presenting with a full spectrum of disease ranging from that with excellent prognosis and high curability to aggressive disease with poor outcome. There are many debates and controversies about optimal treatment for women with different staging endometrial carcinoma. How do we summarize the current recommendations and how do we proceed? Clinicians must balance delivering adequate therapy while attempting to minimize treatment morbidity and must always be weighed carefully.

many debates and controversies about optimal treatment for women with different staging endometrial carcinoma. How do we summarize the current recommendations and how do we proceed? Clinicians must balance delivering adequate therapy while attempting to minimize treatment morbidity and must always be weighed carefully. Improved understanding of the mechanisms of carcinogenesis may help identify molecular signatures that could predict biologic behavior of individual disease presentations and discover potential molecular candidates for targeted therapies. Total hysterectomy and bilateral salphingo-oophorectomy is the primary operative procedure. Pelvic lymhadenectomy is performed in most centers on therapeutic and prognostic grounds and to individualize adjuvant treatment. Women with endometrial carcinoma can be readily segregated intraoperatively into “low-risk” and “high-risk” groups to better identify those women who will most likely benefit from thorough lymphadenectomy. Postoperative irradiation is used to reduce pelvic and vaginal recurrences in high risk cases. Treatment planning should be conservative in order to reduce patients' morbidity and overtreatment while maintaining acceptable recurrence and survival rates. Progression in diagnostic imaging, radiation delivery systems, and systemic therapies potentially can improve outcomes while minimizing morbidity. The availability of new hormonal and biological agents presents new opportunities for therapy. Novel strategies for screening and prevention also hold promise for reducing incidence and mortality of this disease. The current evidence suggests that there remain avenues to improve management and we need to continue rigorous investigation to identify and implement the best available practice. Research in the next ten years should provide valuable new strategies not only for treatment but also for prevention.

ality of this disease. The current evidence suggests that there remain avenues to improve management and we need to continue rigorous investigation to identify and implement the best available practice. Research in the next ten years should provide valuable new strategies not only for treatment but also for prevention. Table 1 WHO histological classification of endometrial carcinoma. Endometrioid adenocarcinoma Variants: with squamous differentiation Villoglandular Secretory with ciliated cells Other adenocarcinomas Mucinous carcinoma Serous carcinoma Clear-cell carcinoma Mixed carcinoma Squamous-cell carcinoma Transitional-cell carcinoma Small-cell carcinoma Undifferentiated carcinoma

1. Introduction Preimplantation genetic screening (PGS) has been used more than 10 years for selecting genetically normal embryos giving the highest potential for preimplantation genetic diagnosis (PGD). PGS usually involves the aspiration of the first polar body from oocyte before fertilization or one or two cells from a 5- to 8-cell embryo 3 days after insemination. Fluorescence in situ hybridization (FISH), is often performed, using probes for a specific number of chromosomes most commonly involved in aneuploidy. The presence or absence of a normal pair of chromosomes can be identified visually by color, so we can eliminate the abnormal embryos and select normal embryos for transfer [1]. Thus, choosing embryos selected by PGS with normal chromosomes should increase implantation rate and live-birth rate and reduce miscarriages. The indications of PGS include advanced maternal age (AMA), repeated implantation failure (RIF), repeated miscarriage (RM), and severe male-factor infertility [2].

transfer [1]. Thus, choosing embryos selected by PGS with normal chromosomes should increase implantation rate and live-birth rate and reduce miscarriages. The indications of PGS include advanced maternal age (AMA), repeated implantation failure (RIF), repeated miscarriage (RM), and severe male-factor infertility [2]. Currently, there have been a great many studies into in vitro fertilization (IVF) or intracytoplasmic sperm injection (ICSI) with and without PGS. Several of them found that selecting embryos with normal chromosomes had a significant impact on the implantation rate compared with the controls [3–6]. PGS has been advocated as a valuable tool for embryo screening. In recent years, its trends become controversial after the report published by Mastenbroek et al. [7]. They found that PGS reduced the rates of pregnancies and live births after IVF in women of AMA [7]. Actually Mastenbroek was not the only one who claimed that PGS might not be as beneficial as expected. Conclusions drawn from other studies in AMA patients after PGS showed that PGS did not significantly improve implantation rate and pregnancy rate, on the contrary it worsened the outcome [8, 9]. In this paper, we will acknowledge the importance of aneuploidy screening and review the findings of currently published studies of PGS, in order to discuss the efficacy of this technique.

Currently, there have been a great many studies into in vitro fertilization (IVF) or intracytoplasmic sperm injection (ICSI) with and without PGS. Several of them found that selecting embryos with normal chromosomes had a significant impact on the implantation rate compared with the controls [3–6]. PGS has been advocated as a valuable tool for embryo screening. In recent years, its trends become controversial after the report published by Mastenbroek et al. [7]. They found that PGS reduced the rates of pregnancies and live births after IVF in women of AMA [7]. Actually Mastenbroek was not the only one who claimed that PGS might not be as beneficial as expected. Conclusions drawn from other studies in AMA patients after PGS showed that PGS did not significantly improve implantation rate and pregnancy rate, on the contrary it worsened the outcome [8, 9]. In this paper, we will acknowledge the importance of aneuploidy screening and review the findings of currently published studies of PGS, in order to discuss the efficacy of this technique. 2. Indications of PGS in AMA, RM, RIFand Severe Male-Factor Infertility Aneuploidies, that is, deviations from the regular number of chromosomes, are predominantly the result of maldistributions of chromosomes during meiosis. Aneuploidy rates in oocytes and embryos are known to increase with maternal age [10]. In a 40-year-old woman, an estimated 50% to 70% of the mature oocytes are affected by a chromosomal abnormality [11, 12]. In a series of 6733 oocytes obtained during 1297 IVF cycles from patients of AMA (mean 38.5 years) [13], 3509 (52%) were aneuploidy, on the basis of FISH analysis using specific probes for chromosomes 13, 16, 18, 21, and 22. It is well known that the age-related increase in aneuploidy rate is correlated with a reduced implantation and a higher abortion rate. Most evidence collected so far suggests that failed implantation due to embryo aneuploidy rather than failed conception is the primary cause responsible for low human fertility [14]. To date, these patients revealed an aneuploidy rate of over 50%, suggesting the practical relevance of PGS to women of advanced reproductive age. Screening for aneuploidy in preimplantation embryos may help select the best embryo to transfer and may open the way to significant improvements in live-birth rates from IVF/ICSI, especially relevant for more effective single embryo transfer [15]. AMA patients, here defined as ≥35 years, are a good target group to assess the possible benefit of aneuploidy screening.

yos may help select the best embryo to transfer and may open the way to significant improvements in live-birth rates from IVF/ICSI, especially relevant for more effective single embryo transfer [15]. AMA patients, here defined as ≥35 years, are a good target group to assess the possible benefit of aneuploidy screening. RM is defined when two or more consecutive spontaneous abortions occur, which affects 1% of couples trying to conceive [16]. The number of miscarriages stands out as a predictor of the chromosome abnormality rate,whichis directly proportional to the number of miscarriages. A study of 108 couples with history of repeated abortions found that chromosome abnormalities were found in 5% of the couples with two abortions, in 10.3% with three abortions, and in 14.3% with four or more abortions [17]. The most common anomaly observed in abortus is aneuploidy, and reported aneuploidy rate could reach to 34–66% [18, 19]. This result suggested that aneuploidy was a common cause of RM, and led to the proposal that PGS may be beneficial in these patients.

th three abortions, and in 14.3% with four or more abortions [17]. The most common anomaly observed in abortus is aneuploidy, and reported aneuploidy rate could reach to 34–66% [18, 19]. This result suggested that aneuploidy was a common cause of RM, and led to the proposal that PGS may be beneficial in these patients. RIF can be defined as the failure of a couple to conceive after the transfer of 10 or more good-quality embryos, or after three IVF cycles [20]. Although multiple aetiologies, such as disturbed endometrial receptivity, uterine pathology, and an inadequate transfer technique, have been proposed, increased incidence of numerical chromosomal abnormalities is obviously the most common cause [21]. It has been reported that the rate of chromosome abnormalities in the embryos from RIF patients is almost twice as much as that in the controls (67.4% versus 36.3%) [22]. Significantly higher incidence of complex chromosome abnormalities (which involves three or more chromosomes) was also found in RIF [23]. The generation of aneuploidy embryos was considered as a possible causative factor in RIF [24], and it is suggested that PGS may improve the outcome in these patients.

s 36.3%) [22]. Significantly higher incidence of complex chromosome abnormalities (which involves three or more chromosomes) was also found in RIF [23]. The generation of aneuploidy embryos was considered as a possible causative factor in RIF [24], and it is suggested that PGS may improve the outcome in these patients. Infertile couples due to severe male factor can be treated with ICSI. In order to generate normally fertilized oocytes after ICSI, a spermatozoon containing a functional genome and centriole is required [25]. Current study in cases of macrocephalic spermatozoa demonstrated an increased incidence of chromosomal abnormalities, and the majority of the abnormalities were aneuploidy [26, 27]. Due to the high incidence of aneuploidy these patients might benefit from PGS owing to its effect of eliminating chromosomally abnormal embryos.

t study in cases of macrocephalic spermatozoa demonstrated an increased incidence of chromosomal abnormalities, and the majority of the abnormalities were aneuploidy [26, 27]. Due to the high incidence of aneuploidy these patients might benefit from PGS owing to its effect of eliminating chromosomally abnormal embryos. 3. Studies with Beneficial Outcome of PGS 3.1. PGS in AMA An early study published by Gianaroli et al. [3] on 157 cycles (73 for PGS group and 84 controls) with AMA using FISH in analysis of chromosomes X, Y, 13, 14, 15, 16, 18, 21, and 22 in a blastomere biopsied from day 3 embryos showed that 64% of embryos presented with chromosomal abnormalities. 22 cycles in the study group had clinical pregnancies versus 25 cycles in the control group, whereas in the study group, the mean number of embryos transferred per patient was significantly lower (2.2 ± 0.9 versus 3.2 ± 0.9), and the implantation rate was higher in comparison with the control group (25.8% versus 14.3%; P < .01). Concomitantly, the implantation rate per pregnant patient was superior in the study group compared with the controls (57.9% versus 38.5%; P < .05). More interestingly, these patients were arbitrarily divided into three classes of age: 36–37 years, 38–39 years, and ≥40 years; the pregnancy and implantation rates characterized in the control group revealed a significant decrease when patients aged ≥38 years. Conversely, in the study group, the percentages of pregnancy and implantation did not differ among the three classes of age, and the implantation rate observed in the oldest categories (≥38 years) was significantly higher after aneuploidy screening than the controls.

a significant decrease when patients aged ≥38 years. Conversely, in the study group, the percentages of pregnancy and implantation did not differ among the three classes of age, and the implantation rate observed in the oldest categories (≥38 years) was significantly higher after aneuploidy screening than the controls. Verlinsky et al. [28] performed a study of polar body diagnosis (PBD) with IVF cycles from patients of AMA. 5590 oocytes were obtained from 917 cycles and tested by polar body sampling and FISH analysis using specific probes for chromosomes 13, 16, 18, 21, and 22, this resulted in 22.2% clinical pregnancies and 140 healthy children born. It seems that polar body testing provides an approach for improving pregnancy rate in IVF patients of AMA. But no control group was presented in this report. Another study of women ageing 35 to 39 years with two or more previous IVF/ICSI treatment trials showed that a higher implantation rate was achieved in the PBD group (17.5% versus 11.8%) [29]. These results suggested that an indication-based use of PBD could certainly provide benefits in older patients.

in this report. Another study of women ageing 35 to 39 years with two or more previous IVF/ICSI treatment trials showed that a higher implantation rate was achieved in the PBD group (17.5% versus 11.8%) [29]. These results suggested that an indication-based use of PBD could certainly provide benefits in older patients. Some articles showed that aneuploidy screening in preimplantation embryos can also reduce embryo loss, increasing ongoing pregnancies and delivery rates. Munné et al. [5] designed a multicentre IVF study to compare controls and a test group that underwent aneuploidy screening, obtaining a significant improvement in the number of spontaneous abortions and ongoing pregnancies. Similar beneficial effects have been reported by other studies. Staessen et al. [8] observed a trend toward a subsequent higher ongoing implantation per transferred embryo rate in tested group (16.5% versus 10.4%; P = .06). In the recent study published in 2009 by Schoolcraft et al. [30], 62 infertile AMA couples undergoing fertility treatment were assigned to the PGS and control group. Results showed that the implantation rates, the number of oocytes, oocyte maturity, and fertilization rate were similar between the two groups. Nevertheless, the authors noted that the spontaneous abortion rate was lower for the test group (25.9% versus 32.26% in the control group), resulting in an observed increase in delivery rate for the test group (78% versus 67.74%). In addition, Hardarson et al. [9] found significantly more good morphological quality embryos (GQEs) in the PGS group on day 3 compared with those found in the control group.

test group (25.9% versus 32.26% in the control group), resulting in an observed increase in delivery rate for the test group (78% versus 67.74%). In addition, Hardarson et al. [9] found significantly more good morphological quality embryos (GQEs) in the PGS group on day 3 compared with those found in the control group. 3.2. PGS in RM, RIF, and Severe Male-Factor Infertility The randomized, prospective study including 19 couples with recurrent pregnancy loss (11 for PGS and 8 controls) by Werlin et al. suggested an improved outcome after performing PGS [31]. Pregnancy rate was 63.6% in study group and 37.5% in controls. In another study performed by Munné et al. [18], the rate of spontaneous abortions in RM subjects undergoing PGS was compared with their own a priori expectations. After PGS, miscarriage rate was reduced from previous 90% (expected 29%) to 23% in the women at age <35 years, and from 86% (expected 44.5%) to 12% in the women at age ≥35 years. Similar results were also reported by a multicenter retrospective controlled study [32], which showed that the spontaneous abortion rate in the PGS group was 14.1% for women ages 35–40 and 22.2% for over 40, compared to 19.4% (P < .03) and 40.6% in the non-PGS group (P < .001).

5%) to 12% in the women at age ≥35 years. Similar results were also reported by a multicenter retrospective controlled study [32], which showed that the spontaneous abortion rate in the PGS group was 14.1% for women ages 35–40 and 22.2% for over 40, compared to 19.4% (P < .03) and 40.6% in the non-PGS group (P < .001). Improved outcomes in RIF were achieved with the selection of chromosomally normal embryos. In a study with 57 RIF cycles by Pehlivan et al. [22], a pregnancy rate of 34.0% and an implantation rate of 19.8% was observed in the PGS group. Recent data reported that, in women with unexplained RIF [33], two consecutive PGS cycles showing euploidy embryo(s) were strongly associated with high ongoing pregnancy (40%) and implantation (18%) rates. Conversely, the patients with no euploid embryos in a PGS cycle were highly unlikely to achieve an ongoing pregnancy in subsequent cycles. Kahraman et al. compared the implantation and ongoing-pregnancy rates of PGS cycles with non-PGS cycles in cases with predominantly macrocephalic spermatozoa and absolute teratozoospermia [34]. A statistically higher implantation rate as well as a significantly reduced missed abortion rate were found in PGS group (25.0% and 14.3%) compared with non-PGS group (12.3% and 46.7%).

ancy rates of PGS cycles with non-PGS cycles in cases with predominantly macrocephalic spermatozoa and absolute teratozoospermia [34]. A statistically higher implantation rate as well as a significantly reduced missed abortion rate were found in PGS group (25.0% and 14.3%) compared with non-PGS group (12.3% and 46.7%). 4. Studies without Beneficial Outcome of PGS 4.1. PGS in AMA In the study by Staessen et al. [8] used FISH for the chromosomes X, Y, 13, 16, 18, 21, and 22 in AMA couples with a control group without PGS. In the 400 (200 for PGS and 200 controls) couples were allocated to the trial, ICSI was used to fertilize the oocytes, and two blastomeres per embryo were removed on day 3 after injection and transferred on day 5. In this study, the implantation rates were not significantly different between the two groups (17.1% in the test group versus 11.5% in the control group). But the cycles that had embryos transferred were significantly lower in test group (81 cycles versus 121; P < .001), and 38 couples in the test group had no genetically normal embryos to transfer. Less than expected success of PGS was attributed to a higher number of embryos transferred in the control group (2.8 versus 2.0) and the possible adverse effect of double-blastomere biopsy [35]. The same group, comparing single-cell versus two-cell biopsy, demonstrated a detrimental effect of two-cell biopsy; they suggested that, if one-cell biopsy had been used in their study, implantation rates may have improved.

control group (2.8 versus 2.0) and the possible adverse effect of double-blastomere biopsy [35]. The same group, comparing single-cell versus two-cell biopsy, demonstrated a detrimental effect of two-cell biopsy; they suggested that, if one-cell biopsy had been used in their study, implantation rates may have improved. Mastenbroek et al. [7] designed a multicenter, randomized, double-blind, controlled trial. 408 women of AMA underwent 836 cycles of IVF, of which 206 women with 434 cycles were assigned to PGS and 202 women with 402 cycles to the control group. The ongoing-pregnancy rate was significantly lower in the women assigned to PGS (52 of 206 women, 25%) than in those not assigned to PGS (74 of 202 women, 37%). The women assigned to PGS also had a significantly lower live-birth rate (24% versus 35%) and reduced implantation rate of (11.7% versus 14.7%) compared with those in the control group. The study was criticized mainly for inappropriate patient selection, inadequate probe selection, possible biopsy-induced embryo damage, a low average number of embryos biopsied, and a high rate of undiagnosed embryos [36, 37].

d reduced implantation rate of (11.7% versus 14.7%) compared with those in the control group. The study was criticized mainly for inappropriate patient selection, inadequate probe selection, possible biopsy-induced embryo damage, a low average number of embryos biopsied, and a high rate of undiagnosed embryos [36, 37]. In the Hardarson et al. study [9], 56 and 53 patients with age ≥38 years were randomly assigned to the PGS and control groups, respectively. Fertilization was performed by IVF or ICSI following standard techniques and FISH analyzed by probes chromosomes X, Y, 13, 16, 18, 21, and 22 in PGS group. Of the analyzed embryos (302 embryos), only 32.4% (98 of 302) had normal chromosome content and 70 of 98 normal embryos were transferred. The number of patients who had embryos transferred was 45 (80.3%) in PGS group and 53 (100%) in control group (P = .001). The clinical pregnancy rate/randomized patient in the PGS group was 8.9% compared with 24.5% in the control group (P = .039). No significant differences were found in the implantation rates (11.4% versus 18.9%) or live-birth rate (5.4% versus 18.9%) per randomized patient between the PGS group and the control group. As shown in these randomized trials, no improvement in efficacy was observed.

mpared with 24.5% in the control group (P = .039). No significant differences were found in the implantation rates (11.4% versus 18.9%) or live-birth rate (5.4% versus 18.9%) per randomized patient between the PGS group and the control group. As shown in these randomized trials, no improvement in efficacy was observed. 4.2. PGS in RM, RIF, and Severe Male-Factor Infertility Platteau et al. designed prospective cohort PGS study in women with recurrent idiopathic miscarriages [19]. The pregnancy results in the older group (≥37 years) were disappointing, with an implantation rate of 2.77% and an ongoing-pregnancy rate of 2.94%. The probable cause for this poor result was that these older women had significantly more chromosomally abnormal embryos than patients <37 years (66.95% versus 43.85%). As to the RIF, a prospectively randomized controlled trial of PGS in IVF/ICSI patients with recurrent failed implantation compared with conventional assisted reproduction treatment procedures was carried out by Blockeel et al. [38]. A total of 139 patients underwent ovarian stimulation, and PGS was performed in 72 patients. No benefit to their implantation and clinical pregnancy rates was found. The implantation rate was 21.4% in the study group and 25.3% in the control group. Moreover, the clinical pregnancy rate was much lower in the study group (25.0% versus 40.3%).

s underwent ovarian stimulation, and PGS was performed in 72 patients. No benefit to their implantation and clinical pregnancy rates was found. The implantation rate was 21.4% in the study group and 25.3% in the control group. Moreover, the clinical pregnancy rate was much lower in the study group (25.0% versus 40.3%). Although severe male-factor infertility is one of the PGS indications that have been put forward, current reports of PGS in severe male-factor infertility are rare. There is a lack of scientific evidence to prove whether PGS is effective in these patients.

s underwent ovarian stimulation, and PGS was performed in 72 patients. No benefit to their implantation and clinical pregnancy rates was found. The implantation rate was 21.4% in the study group and 25.3% in the control group. Moreover, the clinical pregnancy rate was much lower in the study group (25.0% versus 40.3%). Although severe male-factor infertility is one of the PGS indications that have been put forward, current reports of PGS in severe male-factor infertility are rare. There is a lack of scientific evidence to prove whether PGS is effective in these patients. 5. Reasons for Lack of Benefit in PGS Technical reasons for lack of benefit in PGS include both biopsy damage to the remaining embryo that reduces its developmental potential and limitations of current FISH technology that allows only a few chromosomes to be seen. As a result, it is inevitable that some other abnormal chromosomes will escape from detection [39]. Moreover, FISH could be misdiagnosed by the probability of hybridization failure and the possibility that the fluorescent signals of two chromosomes overlap each other. The testing of all chromosomes would probably further increase observed aneuploidy rates [40]. Mosaicism, a difference of the chromosomal constitution among individual cells in an embryo, is another possible reason for confusion. A single blastomere that had been biopsied might thus be classified as abnormal, whereas the remaining blastomeres in the embryo are normal. Thus, the test results from the biopsied cell may not be an accurate indication of the embryo's genetic status [41]. Besides technical limitations and mosaicism, contamination and laboratory mistakes can also result in inaccurate diagnoses. For example, DNA from sources other than the biopsied cell may be read as part of the genetic analysis, mixup, or mislabeling of a sample or embryo from clinic or laboratory mistakes in handling samples or embryos. All of these can lead to inaccurate results.

laboratory mistakes can also result in inaccurate diagnoses. For example, DNA from sources other than the biopsied cell may be read as part of the genetic analysis, mixup, or mislabeling of a sample or embryo from clinic or laboratory mistakes in handling samples or embryos. All of these can lead to inaccurate results. 6. Possible Future Trends in PGS 6.1. Comparative Genomic Hybridization (CGH) Performed on a single cell basis, CGH enables the assessment of all the chromosomes by comparing the studied DNA with a normal sample. In brief, normal DNA samples are labelled with red and test DNA with green fluorochromes, and then applied to a slide where hybridization occurs for 48–72 h [42]. The advantage of CGH over the conventional FISH is that the copy number of all chromosomes can be determined. CGH can provide a genome-wide profile without any prior information of the chromosomal aberration [40].

est DNA with green fluorochromes, and then applied to a slide where hybridization occurs for 48–72 h [42]. The advantage of CGH over the conventional FISH is that the copy number of all chromosomes can be determined. CGH can provide a genome-wide profile without any prior information of the chromosomal aberration [40]. Fragouli et al. [43] collected 270 oocytes from the 16 female patients (average age 38.4 years) and 168 embryos were fertilized on day 3 (average 12 embryos per patient, range 6–18). Of the 168 embryos, 78 (46.4%) were cultured further to the blastocyst stage and underwent trophectoderm biopsy with CGH screening. Their data displayed high implantation and pregnancy rates for the patients with RIF who have received blastocyst analysis [43]. CGH yielded results for 73 of the 78 blastocysts, leading to a diagnostic efficacy of 94%. Of these, 40 were classified as euploidy and 24 were transferred in 13 patients, leading to nine ongoing pregnancies from 13 completed cycles (69.2%) and the implantation rate was 58.3% (14/24 ETs). The limitations of CGH are that it is time-consuming and labour intensive. The long period required for hybridization (5 days) has limited the widespread clinical implementation of this technique, as it is necessary to freeze all the embryos after the biopsy. In addition, the survival rate of the thawed embryos was relatively poor (46% did not survive the thawing process). More recently, the development of highly efficient techniques has greatly reduced fears concerning the impact of cryopreservation on embryo viability. Array CGH is one of the newest technologies developed for the detection of a chromosomal imbalance; it is able to analyze the very limited amount of genetic material in a single cell and takes less time [42]. Accuracy microarray platforms also can offer the advantage of embryo fingerprinting and the potential for combined aneuploidy and single-gene disorder diagnosis [44]. The first report to show a pregnancy after PGS using array CGH technology by Hellani et al. obtained a high pregnancy rate; six out of a total eight patients had embryos for transfer with five out of those six showing positive pregnancy tests [45]. The result was encouraging and further studies on array CGH with larger sample sizes will be required before it is suitable for clinical application. However, some disadvantages need to be addressed before array-CGH is suitable for clinical services. First of all, the accuracy needs further evaluation.

ancy tests [45]. The result was encouraging and further studies on array CGH with larger sample sizes will be required before it is suitable for clinical application. However, some disadvantages need to be addressed before array-CGH is suitable for clinical services. First of all, the accuracy needs further evaluation. Array sometimes gave incorrect results for chromosomes 2, 4, 9, 11, 17 and 22 [46]. Partial aneuploidy and imbalance of chromosome segments are not currently detected. Besides, the present array CGH protocol is expensive and it doesn't seem to fit easily into all clinical PGS services. This requires us to find new ways to reduce costs and bring the advantages to more patients.

, 9, 11, 17 and 22 [46]. Partial aneuploidy and imbalance of chromosome segments are not currently detected. Besides, the present array CGH protocol is expensive and it doesn't seem to fit easily into all clinical PGS services. This requires us to find new ways to reduce costs and bring the advantages to more patients. 6.2. Blastocyst Biopsy Blastocyst biopsy or trophectoderm biopsy is an emerging technique for performing PGS. It shows several advantages over traditional day-3 biopsy [46]. One of them is more cells can be biopsied for genetic testing without damaging the inner cell mass. Biopsy at this stage has little, if any, impact on the further development of the blastocysts. The data from McArthur et al. demonstrates high blastocyst survival rates with excellent implantation rates and low rates of twinning or miscarriage [47]. Recently, a study involving 399 egg retrievals and 1879 embryo biopsies for patients undergoing PGD to avoid a serious monogenic disease or an unbalanced chromosomal translocation has been published. The implantation rates per embryo transferred were 43.4% if biopsied at the blastocyst stage and 25.6% if biopsied at the cleavage stage (P < .01), with ongoing or live-birth pregnancy rates per egg retrieval at 34.2% (average transfer number 1.1) for blastocyst biopsies and 25.5% (transfer number 1.6) for cleavage stage biopsies (P < .05). The results mean that taking the biopsy later in embryo development conferred considerable efficacy through not testing embryos whose development was compromised [48]. Nevertheless, more data is still needed to confirm these promising results.

psies and 25.5% (transfer number 1.6) for cleavage stage biopsies (P < .05). The results mean that taking the biopsy later in embryo development conferred considerable efficacy through not testing embryos whose development was compromised [48]. Nevertheless, more data is still needed to confirm these promising results. 7. Conclusions In conclusion, the efficacy of PGS is still controversial. According to the studies, there is still insufficient evidence to support a beneficial effect of PGS in AMA women. The use of PGS applied for RM, RIF and severe male factor infertility needs more scientific data from clinical trials. The routine use of PGS to avert the birth of an aneuploidy infant is still in question. Application of micro-CGH and blastocyst biopsy might be new approaches for improvement of the efficacy of PGS. Furthermore, the cost-effectiveness of PGS for the IVF patients should be considered. Acknowledgments Ning Wang and Ying-Ming Zheng are contributed equally to this paper. The research was supported by National Basic Research Program of China (2007CB948104) and Natural Science Foundation Projects of Zhejiang (Z207021).

1. Introduction Cervical cancer is the second most common malignant disease among women worldwide [1]. However, its incidence and mortality rates have been decreasing in many highly developed countries, for example, the Nordic countries, mainly due to organized Pap smear screening [2]. Members of the matrix metalloproteinase (MMP) family have the ability to degrade macromolecules of the extracellular matrix, and they are responsible for tumor invasion and infiltration. They are a family of zinc and calcium-dependent proteinases, and they are secreted in the proenzyme form [3]. Besides their role in tumor invasion and metastasis, MMPs are involved in various physiological connective tissue remodeling processes, such as ovulation and wound healing [4]. Matrix metalloproteinases form an enzyme family that takes part in virtually all events of extracellular matrix remodeling and turnover. Cancer cells capable of producing these enzymes, especially of gelatinases MMP-2 and -9, have been shown to have increased invasion potential in several malignancies [5, 6]. Their action is inhibited by their natural tissue inhibitor molecules TIMP-1 and -2. There are no previous studies on circulating metalloproteinases and squamous cervical carcinoma. The aim of this paper was to evaluate the levels of circulating metalloproteinases and their inhibitors in squamous cervical carcinoma patients and healthy women.

Matrix metalloproteinases form an enzyme family that takes part in virtually all events of extracellular matrix remodeling and turnover. Cancer cells capable of producing these enzymes, especially of gelatinases MMP-2 and -9, have been shown to have increased invasion potential in several malignancies [5, 6]. Their action is inhibited by their natural tissue inhibitor molecules TIMP-1 and -2. There are no previous studies on circulating metalloproteinases and squamous cervical carcinoma. The aim of this paper was to evaluate the levels of circulating metalloproteinases and their inhibitors in squamous cervical carcinoma patients and healthy women. 2. Materials and Methods The patient material consisted of 12 early-stage (FIGO stage IB and IIA) squamous cervical carcinoma patients who underwent surgery at Oulu University Hospital during the years from 1992 to 1995 and 27 healthy volunteer control patients. The serum samples were collected by taking venous blood samples from the patients at the time of the primary squamous cervical cancer diagnosis prior to surgery. The samples were collected in glass tubes that did not contain any artificial coagulation activators. They were allowed to coagulate, centrifuged at 3000 rpm for 10 minutes, and the native serum obtained was separated and frozen at −20° until used.

ime of the primary squamous cervical cancer diagnosis prior to surgery. The samples were collected in glass tubes that did not contain any artificial coagulation activators. They were allowed to coagulate, centrifuged at 3000 rpm for 10 minutes, and the native serum obtained was separated and frozen at −20° until used. All patients with cervical carcinoma were treated with preoperative intracavitary brachytherapy, followed by radical hysterectomy and lymph node dissection. The majority of the patients received postoperative radiation. Additional platinol-based chemotherapy was given if pelvic lymph node metastases were found in the operation. The patients' median age was 52 years, ranging from 31 to 80, at the moment of diagnosis. The follow-up time for each patient was 120 months at the minimum.

ity of the patients received postoperative radiation. Additional platinol-based chemotherapy was given if pelvic lymph node metastases were found in the operation. The patients' median age was 52 years, ranging from 31 to 80, at the moment of diagnosis. The follow-up time for each patient was 120 months at the minimum. 2.1. Assay for the Immunoreactive Proteins of MMP-2-TIMP-2 Complex, MMP-9, TIMP-1, and TIMP-2 The immunoreactive proteins for MMP-2-TIMP-2 -complex, MMP-9, TIMP-1, and TIMP-2 were assayed from the sera of the patients with squamous cervical carcinoma and from that of healthy volunteers using enzyme-linked immunoassay (ELISA). The quantification of the total protein of MMP-2-TIMP-2 complex, MMP-9, TIMP-1, or TIMP-2 was performed using standard protocols [7]. Briefly, the ELISAs were performed on EIA/RIA 8-well stripes (Corning Incorporated, Corning, New York, USA). A polyclonal antibody produced in chicken against each of the analytes was used as a secondary antibody. O-phenylenediamine dihydrochloride (OPD) (Sigma, Steinheim, Germany) was used to visualize the peroxidase label. Color formation was measured on 450 nm (Anthos 2001 microplate reader), and calculations were done using a Windows-based control and evaluation software for Rosys Anthos microplate readers (Anthos labtec instruments, Wals, Austria).

ochloride (OPD) (Sigma, Steinheim, Germany) was used to visualize the peroxidase label. Color formation was measured on 450 nm (Anthos 2001 microplate reader), and calculations were done using a Windows-based control and evaluation software for Rosys Anthos microplate readers (Anthos labtec instruments, Wals, Austria). The monoclonal antibody against MMP-2-TIMP-2 recognizes the solube lMMP-2-TIMP-2 complex. The monoclonal antibody against MMP-9 (code Ge-213) recognizes both free MMP-9 and that bound to its inhibitor TIMP-1. The monoclonal antibody against TIMP-1 (code DB-102D1) recognizes both free TIMP-1 and that complexed with MMP-9. It does not cross-react with TIMP-2. The monoclonal antibody against TIMP-2 (code T2-101) recognizes both free TIMP-2 and that complexed with MMP-2. Each sample was run in duplicate in order to minimize intraassay variation. The absorbance values for standard samples and the standard curves constructed for each assay were compared and used to minimize interassay variation. 2.2. Statistical Analysis Statistical analysis was performed with SPSS (v. 11.5) for Windows (SPSS Inc, Chicago, Ill. USA) software package. Wilcox test was used to investigate the significance of the differences between medians. A P-value <  .05 was considered statistically significant. 3. Results The mean levels of serum TIMP-2 and MMP-2-TIMP-2 complex were higher in the healthy controls compared to those with a malignant tumor (Table 1). For the MMP-9 and TIMP-1, no significant differences were found between patients' and controls' concentrations.

2.2. Statistical Analysis Statistical analysis was performed with SPSS (v. 11.5) for Windows (SPSS Inc, Chicago, Ill. USA) software package. Wilcox test was used to investigate the significance of the differences between medians. A P-value <  .05 was considered statistically significant. 3. Results The mean levels of serum TIMP-2 and MMP-2-TIMP-2 complex were higher in the healthy controls compared to those with a malignant tumor (Table 1). For the MMP-9 and TIMP-1, no significant differences were found between patients' and controls' concentrations. Serum TIMP-2 values decreased significantly from healthy controls (mean 323 μg/l, range 305–342 μg/l) to malignant (mean 136 μg/l, range 120–151 μg/l) squamous cervical carcinoma patients (P < .000). Serum proMMP2-TIMP2 complex values decreased from control patients (mean 625 μg/l, range 588–662 μg/l) to malignant (mean 514 μg/l, range 453–575 μg/l) to squamous cervical carcinoma patients (P < .006). 4. Discussion This paper shows for the first time that the levels of circulating TIMP-2 and the MMP-2-TIMP-2 complex are lower in squamous cervical cancer patients than in healthy women. Previously Ylisirniö et al. [8] found that the TIMP-2 and the MMP-2/TIMP-2 complex levels were lower in lung cancer than in the sera of the control subjects. Also in colorectal cancer, the serum levels of the MMP-2-TIMP-2 complex are found to be lower than in healthy controls [9]. Paula et al.[10] found that breast cancer patients had significantly lower TIMP-2 levels than did healthy controls.

mplex levels were lower in lung cancer than in the sera of the control subjects. Also in colorectal cancer, the serum levels of the MMP-2-TIMP-2 complex are found to be lower than in healthy controls [9]. Paula et al.[10] found that breast cancer patients had significantly lower TIMP-2 levels than did healthy controls. We found in this study that squamous cervical cancer patients had significantly lower TIMP-2 levels compared to healthy controls. In squamous cervical cancer, no evidence currently exists showing that TIMP-2 could be prognostic when measured in peripheral blood. However, the presence of the disease might alter the balance of the proteolysis. Low TIMP-2 levels in the blood of squamous cervical cancer patients could indicate more activated MMP-2 and therefore higher usage of TIMP-2, or alternatively, the lower levels could be due to lower production of TIMP-2, leading to less inhibition of MMP-2 activity. For the proMMP2-TIMP2 complex, lower concentrations are found in squamous cervical cancer patients. It has been shown that during aggregation, platelets release MMP-2 in its latent form [11] as well as other components of the proMMP2/MT1-MMP/TIMP-2 system [12]. Patients and controls did not have significant differences in their serum MMP-9 and TIMP-1 levels, although the ranges for MMP-9 and TIMP-1 concentrations were wider in cancer patients compared with healthy controls. This could indicate the presence of a disturbance, such as cancer, that could affect the levels of the analytes.

and controls did not have significant differences in their serum MMP-9 and TIMP-1 levels, although the ranges for MMP-9 and TIMP-1 concentrations were wider in cancer patients compared with healthy controls. This could indicate the presence of a disturbance, such as cancer, that could affect the levels of the analytes. In conclusion, decreased circulating levels of TIMP-2 and MMP2-TIMP2 complex are found in the serum samples of squamous cervical carcinoma patients. Future studies will determine the prognostic value of TIMP-2 and MMP2-TIMP2 in squamous cervical carcinoma. Table 1 Serum levels of MMP-2-TIMP-2 complex, MMP-9, TIMP-1, and TIMP-2 in patients with squamous cervical cancer and in healthy controls. Serum marker Cervical cancer mean (range) Healthy controls mean (range) P-value1 MMP-2-TIMP-2 (μg/l) 514 (453–575) 625 (588–662) .006 MMP-9 (μg/l) 74 (40–108) 73 (59–87) .659 TIMP-1 (μg/l) 466 (422–511) 414 (383–443) .067 TIMP-2 (μg/l) 136 (120–151) 323 (305–342) .000 The results are expressed as mean (range).1 Wilcox

1. Introduction Stillbirths and neonatal deaths remain a huge challenge in the care of pregnant women, especially in developing countries [1]. Over 3.3 million stillbirths and more than 3 million early neonatal deaths occur every year. A vast majority (98%) takes place in developing countries where stillbirths represent over half of the perinatal deaths [2]. Complications during pregnancy and child birth are known to be closely associated with high stillbirth and perinatal mortality rate [3]. Perinatal mortality and stillbirth rates are important indicators of the quality of antenatal and obstetric care in a community [4]. Despite an important indicator stillbirths are invisible in global policy and programme priorities. They are usually not captured in local data collecting systems [2, 5]. Lack of a well-defined programme agenda, coupled with the lack of data, and social invisibility, deter action and investments for stillbirth prevention and reduction [2].

indicator stillbirths are invisible in global policy and programme priorities. They are usually not captured in local data collecting systems [2, 5]. Lack of a well-defined programme agenda, coupled with the lack of data, and social invisibility, deter action and investments for stillbirth prevention and reduction [2]. Being cognizant of the distribution of stillbirths (fresh and macerated) and deaths within the immediate postpartum period may help to detect shortcomings in the quality of antenatal and obstetric care given to the pregnant woman, hence prioritize appropriate intervention programmes [6]. Data on the frequency and distribution of these adverse births outcomes are important for planning maternal and child health services in developing countries [7]. As a drive to achieve the Millennium Development Goals (MDGs), the Government of The Gambia developed a national road map to accelerate the reduction of maternal and newborn morbidity and mortality [8]. However, this strategy remains a challenge due to weak health system, gross shortage of skilled human resource, and inadequate access to emergency obstetric care. The aim of this study was to determine the frequency/rate of stillbirths and its associated socio-demographic and medical factors in two rural referral hospitals in The Gambia. To our knowledge no previous study has assessed the frequency of stillbirths in The Gambia.

and inadequate access to emergency obstetric care. The aim of this study was to determine the frequency/rate of stillbirths and its associated socio-demographic and medical factors in two rural referral hospitals in The Gambia. To our knowledge no previous study has assessed the frequency of stillbirths in The Gambia. 2. Materials and Methods 2.1. Study Setting and Design A cross-sectional retrospective study was carried out at Bansang General Hospital and Armed Forces Provisional Ruling Council (AFPRC) Hospital. These hospitals are located in the North and South bank of The Gambia, respectively, in two different health regions. Comprehensive EOC is available most of the time, mainly provided by Cuban Medical doctors. There is no Gambian medical doctor or obstetrician in any of these hospitals. The first class of medical doctors educated in The Gambia completed their education in 2007. The two hospitals serve a population of nearly 600,000 and are referral points for nearly 30 peripheral health centres and or dispensaries. Basic EOC is not available at any of these peripheral health centres. Thus, women in either the North Bank or South Bank with obstetric complications are referred to Bansang or AFPRC hospitals. Most of these women are referred during labour. The Government of The Gambia has adopted the primary health care (PHC) strategy to make health care more accessible to the rural population. Villages with more than 400 inhabitants have resident traditional birth attendants (TBA) who have eight weeks formal training in antenatal, intrapartum and postpartum care of the mother and child. These TBAs are being supervised by a community health nurse (CHN) who is in charge of a cluster of villages. Antenatal care is provided by mobile reproductive and child health clinics from the health centres and the two hospitals.

ks formal training in antenatal, intrapartum and postpartum care of the mother and child. These TBAs are being supervised by a community health nurse (CHN) who is in charge of a cluster of villages. Antenatal care is provided by mobile reproductive and child health clinics from the health centres and the two hospitals. 2.2. Study Population We used data from hospital records on all women who gave birth at or after 28 completed weeks of gestation from 1st July 2008 to 31st December 2008. Data was abstracted from maternity case notes, admission, and delivery registers. Midwives or doctors attending a birth complete a standardised form to be filled in upon admission and immediately after delivery. The form contains important information about maternal health and complications during pregnancy and the intrapartum period. It also contains information about the newborn. A precoded case abstraction questionnaire was used. Data abstraction was done by the principal investigator and assisted by research assistants, mainly midwives.

ins important information about maternal health and complications during pregnancy and the intrapartum period. It also contains information about the newborn. A precoded case abstraction questionnaire was used. Data abstraction was done by the principal investigator and assisted by research assistants, mainly midwives. 2.3. Variables The main outcome measure was stillbirth rate calculated as deaths per 1000 births. The eligibility criteria was based on the World Health Organization's (WHOs) international comparison of viability; that is birth weight of ≥1000 g and or born at ≥28 weeks of gestation. Thus, we defined stillbirth as death in the uterus of an infant at ≥28 weeks of gestation or ≥1000 g. It was classified as fresh when the baby was born with an intact skin suggesting that the death occurred during labour (less than 12 hours before delivery). A macerated stillbirth was defined when there was sign of degeneration suggesting the death having occurred more than 12–24 hours before labour. We also recorded early neonatal deaths defined as death of the newborn baby within the first twelve hours of birth. During this period all the live born babies and their mothers were still under observation in the hospital. Due to early discharge from hospital (twelve hours after delivery) and a retrospective study design, information on live-births who might have died at home within seven days of birth was not known. Due to low number and great uncertainty these results were not included in the tables.

under observation in the hospital. Due to early discharge from hospital (twelve hours after delivery) and a retrospective study design, information on live-births who might have died at home within seven days of birth was not known. Due to low number and great uncertainty these results were not included in the tables. For each birth, demographic and obstetric explanatory factors were captured. The demographic variables included maternal age in years, categorized in three groups: <20 (reference), 20–34, and ≥35. Parity coded as primiparous (0), 1–3 previous deliveries, ≥4 previous deliveries, area of residence (PHC village or non-PHC village). Obstetric factors included the following: admission status of the mother (booked or referred), antenatal care attendance for present pregnancy coded as no or yes, mode of delivery (spontaneous vaginal, assisted vaginal (breech), and caesarean section), presence of severe obstetric complication which included one or more of the following: prelabour rupture of membranes preterm, hypertensive pregnancy disorders (pre-eclampsia and eclampsia), antepartum haemorrhage comprising (placenta previa and abruption placenta), cephalopelvic disproportion (CPD), prolonged or obstructed labour, severe anaemia (haemoglobin level <9 g/dL). Foetal characteristic was birth weight <2500 g and ≥2500 g.

preterm, hypertensive pregnancy disorders (pre-eclampsia and eclampsia), antepartum haemorrhage comprising (placenta previa and abruption placenta), cephalopelvic disproportion (CPD), prolonged or obstructed labour, severe anaemia (haemoglobin level <9 g/dL). Foetal characteristic was birth weight <2500 g and ≥2500 g. A total of 1,849 maternity admissions were recorded during the six months period. We excluded 224 (12.2%) who had not delivered. Twenty-one births were further excluded due to missing information on the vital status and birth weight, and 25 who weighed less than 1000 g. We also excluded 60 deliveries that occurred before reaching the hospital. The final data set for this analysis was 1,519. The Ethics Committee of Norway and the Joint Gambia Government and Medical Research Council Review board approved the study. Permission to carry out the study was achieved from the Ministry of Health of The Gambia and the chief executive officers of Bansang and AFPRC hospitals.

nal data set for this analysis was 1,519. The Ethics Committee of Norway and the Joint Gambia Government and Medical Research Council Review board approved the study. Permission to carry out the study was achieved from the Ministry of Health of The Gambia and the chief executive officers of Bansang and AFPRC hospitals. 2.4. Statistical Analysis All 1,519 institutional births were included in the analysis. Frequency analysis and cross-tabulations were used to determine the frequency and percentage of stillbirth and early neonatal mortality. Overall, stillbirth was calculated as a proportion of all births while early neonatal mortality was presented as a proportion of live births. Fresh and macerated stillbirths were calculated as a proportion of stillbirths. Multiple births were initially excluded, but repeating the analysis including multiple births the stillbirth rate remained largely unchanged. Thus we decided to maintain multiple births in the analyses. Univariate association between covariates and stillbirths were assessed with chi-square test or Fisher's exact test as appropriate. All P-values were two-sided and a value of 0.05 was considered statistically significant. Finally all covariates were included in a multivariate logistic regression model to determine significant factor associated with stillbirth. All statistical analysis was done with Software Package for Social Sciences (SPSS) for Windows version 16.0 (SPSS Inc, Chicago, II, USA).

was considered statistically significant. Finally all covariates were included in a multivariate logistic regression model to determine significant factor associated with stillbirth. All statistical analysis was done with Software Package for Social Sciences (SPSS) for Windows version 16.0 (SPSS Inc, Chicago, II, USA). 3. Results The total number of deliveries over the six months period was 1,519. Of these, 237 were stillbirths, representing a stillbirth rate of 156 (95% CI 138–174) per 1000 births. Of the 237 recorded stillbirths, 137 (57.8%) were fresh stillbirth. We recorded 6 early hospital neonatal deaths giving a hospital neonatal mortality rate of 5 (95% CI 2–10) per 1000. More than half (54.9%) of the women were between 20–34 years old. Forty-six percent were primiparous. Nearly all (99.3%) attended antenatal care at least once. Most of the women (89%) had a spontaneous vaginal delivery, 8% delivered by caesarean section, while 3.4% of the births were assisted breech delivery. Twenty-two percent of the women were referred from peripheral health centres. Of the 1,519 recorded deliveries the partograph was not used in 958 (63%). Overall, 53 (3.5%) of the women had a breech presentation at delivery (Table 1).

elivered by caesarean section, while 3.4% of the births were assisted breech delivery. Twenty-two percent of the women were referred from peripheral health centres. Of the 1,519 recorded deliveries the partograph was not used in 958 (63%). Overall, 53 (3.5%) of the women had a breech presentation at delivery (Table 1). Table 1 shows the results of the univariate analysis. The crude analysis indicate that stillbirths were highly significantly associated with the following factors: referral from a peripheral health facility (odds ratio: 13.75, 95% CI (10.02–18.84), severe obstetric complication 11.74, 95% CI (8.58–16.06), birth weight <2500 g 6.32, 95% CI (4.67–8.55), residence in a PHC village 2.13, 95% CI (1.59–2.80), nonuse of patograph 2.48, 95% CI (1.78–3.45) and ≥4 pregnancies 2.24, 95% CI (1.61–3.10). Lack of antenatal care, assisted breech delivery, and maternal age 35 years and above were also associated with high stillbirth rate, OR = 5.5, 95% CI (1.58–19.16), 1.80, 95% CI (1.16–2.80), and 1.80, 95% CI (1.14–2.83), respectively. After adjusting for all the variables in a multivariate logistics analysis, presence of severe obstetric complication(s), birth weight <2500 g, caesarean section delivery and referral from a peripheral health facility were the most important factors highly significantly associated with higher stillbirth rates OR = 6.68 (95% CI. 3.84–11.62), 4.47 (95% CI. 3.04–6.59), 4.35 (95% CI 2.46–7.69), and 3.82 (95% CI. 2.24–6.51), respectively. However, the proportion of stillbirths was relatively lower in elective c/s groups (20.8%) than for the emergency c/s groups (23.5%), P > .05. In addition, other factors associated with high stillbirth rate were nonuse of partograph OR = 1.70 (95% CI 1.23–2.56), multiple pregnancy OR = 2.01 (95% CI 1.05–3.86), and not attending antenatal care OR = 4.46 (95% CI 0.84–23.43) (Table 1).

ective c/s groups (20.8%) than for the emergency c/s groups (23.5%), P > .05. In addition, other factors associated with high stillbirth rate were nonuse of partograph OR = 1.70 (95% CI 1.23–2.56), multiple pregnancy OR = 2.01 (95% CI 1.05–3.86), and not attending antenatal care OR = 4.46 (95% CI 0.84–23.43) (Table 1). The association between maternal demographic/obstetric factors and fresh stillbirths are presented in Table 2. On univariate analysis, complications during intrapartum period, being delivered at AFPRC hospital and birth weight <2500 g were significantly associated with fresh stillbirth, OR = 3.57 (95% CI 1.52–8.40), 2.15 (95% CI 1.26–3.66), and 2.15 (95% CI 1.27–3.63), respectively. After adjusting for the effect of all the variables in a multivariate analysis intrapartum severe obstetric complication was the only independent factor associated with high rate of fresh stillbirth; OR = 3.14, 95% CI (1.01–9.76). Of the 1339 live births registered during the study period, 11 maternal deaths were recorded representing a hospital maternal mortality rate of 822/100,000 live births (LB). Of the 11 recorded maternal deaths, 7 (1,169/100,000 LB) and 4 (541/100,000 LB) were in Bansang and AFPRC hospitals, respectively.

The association between maternal demographic/obstetric factors and fresh stillbirths are presented in Table 2. On univariate analysis, complications during intrapartum period, being delivered at AFPRC hospital and birth weight <2500 g were significantly associated with fresh stillbirth, OR = 3.57 (95% CI 1.52–8.40), 2.15 (95% CI 1.26–3.66), and 2.15 (95% CI 1.27–3.63), respectively. After adjusting for the effect of all the variables in a multivariate analysis intrapartum severe obstetric complication was the only independent factor associated with high rate of fresh stillbirth; OR = 3.14, 95% CI (1.01–9.76). Of the 1339 live births registered during the study period, 11 maternal deaths were recorded representing a hospital maternal mortality rate of 822/100,000 live births (LB). Of the 11 recorded maternal deaths, 7 (1,169/100,000 LB) and 4 (541/100,000 LB) were in Bansang and AFPRC hospitals, respectively. 4. Discussions 4.1. Main Results The stillbirth rate found in the two rural hospitals in The Gambia was unacceptably high, pegging at 156 per 1000 total births. The reported early neonatal deaths rate was 5/1000 live births. Presence of severe obstetric complication showed a close association with stillbirth, followed by low birth weight, caesarean section, and referral from a peripheral health facility. Stillbirth was also associated with nonused of the partograph, multiple pregnancy, and lack of antenatal care. In addition, obstetric complications during the intrapartum period were independently associated with fresh stillbirths.

by low birth weight, caesarean section, and referral from a peripheral health facility. Stillbirth was also associated with nonused of the partograph, multiple pregnancy, and lack of antenatal care. In addition, obstetric complications during the intrapartum period were independently associated with fresh stillbirths. 4.2. Methodological Considerations Even though hospital-based data has a limitation in the correct appraisal of the magnitude of a problem in the general population, lack of nationwide vital registration system in many developing countries including The Gambia, has made population-based studies unfeasible. Some field reports on stillbirths from Zimbabwe [7] and The Gambia [9] were established from hospital data. These data are however very vital in rendering both clinical and research priorities. Several limitations of this study should be recognized. Due to the hospital-based design of the current study, we might have overestimated the stillbirth rate. The reported numbers of early neonatal mortality are small and such deaths may also underestimate the true neonatal mortality rates since no systemic follow-up mothers or infants were undertaken. Some of the women may have experienced neonatal deaths after discharge from hospital. Due to the practice of hospital discharge within 12 hours after delivery most of the early neonatal deaths were also not captured in the maternity records. Usually hospital data will show a very high percentage of deaths due to asphyxia since complicated births are more likely to come to hospital. The very small number of observations of early neonatal deaths in our study gives estimates with large uncertainty. Thus, the findings of our study cannot be generalized to the entire country and should be interpreted with caution.

of deaths due to asphyxia since complicated births are more likely to come to hospital. The very small number of observations of early neonatal deaths in our study gives estimates with large uncertainty. Thus, the findings of our study cannot be generalized to the entire country and should be interpreted with caution. 4.3. Stillbirth Rates and Associated Factors The reported stillbirth rate in the current study is higher than in a recent hospital-based study by Cham et al., 116 per 1000 births [9]. The rate is also higher than in other previously reported findings from The Gambia [10, 11] and, in one hospital-based study conducted in Zimbabwe, 61 per 1000 births [12]. Our rate is also considerably increased compared to the reported rates within Sub-Saharan Africa; 32.2 per 1000 births [5], and the WHO model estimates of 42 per 1000 births [2]. We speculate that the higher rates registered in our study could be attributed to the referral of complicated obstetric cases from peripheral health centres. Most of the obstetric cases referred often reach the hospital when it is already late. In addition, the high stillbirth rate may be in part due to the low degree of obstetric vigilance and improper labour management. Our study demonstrated a very high stillbirth rate and a relatively very low neonatal mortality rate. This may indicate a serious delay on behalf of the baby. Thus, it would be reasonable to assume that the babies do not even live to be born “asphyxiated.” However, the small number of neonatal deaths gives insufficient power to conclude on this matter.

stillbirth rate and a relatively very low neonatal mortality rate. This may indicate a serious delay on behalf of the baby. Thus, it would be reasonable to assume that the babies do not even live to be born “asphyxiated.” However, the small number of neonatal deaths gives insufficient power to conclude on this matter. Unavailability and high cost of transportation, poor road conditions, and time to arrange for transport from remote villages may increase the time to reach a health facility [13]. Such factors could play an important role to the findings of the current study. At the time of an obstetric emergency, every moment of delay in seeking and receiving skilled care increases the risk of stillbirth, neonatal or maternal death or disability. If there were fewer delays, fewer babies would probably be stillborn, but many more would be born asphyxiated or die early. Reducing transport time to an EOC facility is challenging in rural settings where roads, public transportation, and communication infrastructures are poor and the terrain may be formidable [13]. Evidence exists that a functioning continuum of care between the home, health centre, and hospital is required to minimize potentially deadly delays and effectively link pregnant women and newborns to skilled obstetric and newborn care [13]. However, even where prompt referral was initiated, the gross shortage of trained human resources for health and inadequate facilities for emergency obstetric and neonatal care must be overcome to reduce perinatal deaths in rural hospitals in The Gambia.

ant women and newborns to skilled obstetric and newborn care [13]. However, even where prompt referral was initiated, the gross shortage of trained human resources for health and inadequate facilities for emergency obstetric and neonatal care must be overcome to reduce perinatal deaths in rural hospitals in The Gambia. Traditionally, advanced maternal age is viewed as risk factor for pregnancy complications and adverse perinatal outcomes including stillbirths [14, 15]. However, the biological mechanism underlying the increased risk for adverse perinatal outcomes with advanced maternal age is unclear [15]. Older mothers in our population had the highest percentage of stillbirths and the proportion of stillbirth increased steadily as age increases. However, the significant association was lost after adjusting for all the variables included in the study.

dverse perinatal outcomes with advanced maternal age is unclear [15]. Older mothers in our population had the highest percentage of stillbirths and the proportion of stillbirth increased steadily as age increases. However, the significant association was lost after adjusting for all the variables included in the study. Almost all the women in our study (99%) attended antenatal care at least once. However, the percentage of stillbirth was much higher among mothers who did not attend antenatal care compared with those who did. This is consistent with results of other studies carried out elsewhere [12, 16]. Better understanding of foetal growth and development and its relationship to the mother's health has resulted in increased attention to the potential of antenatal care as an intervention to improve both maternal and newborn health [17]. Antenatal care provides a critical linkage between the woman and maternity care services. Thus, if promoted in concurrence with effective EOC and delivered in skilled hands, it may become an effective instrument to improve maternal and perinatal birth outcomes particularly in developing countries [18]. However, opponents maintained that antenatal care could only detect morbidity during pregnancy and could not detect obstetric complications that would occur during labour [19].

killed hands, it may become an effective instrument to improve maternal and perinatal birth outcomes particularly in developing countries [18]. However, opponents maintained that antenatal care could only detect morbidity during pregnancy and could not detect obstetric complications that would occur during labour [19]. Noncompliance in completing the partograph is common in Gambian hospitals. A higher percentage of stillbirths were observed in situations where the partograph was not used. In a recent paper by Cham et al. on foetal outcome in severe maternal morbidity, the partograph was not used in any of the 725 identified hospital deliveries [9]. The partogram graphically represents key events during labour. It is recommended for routine monitoring of labour to provide an early warning system; thus, assists the health worker to identify slow progress in early labour, hence, initiate appropriate interventions to avert prolonged and obstructed labour [20]. Few studies have assessed partograph used versus no partograph, the impact of which would be underestimated in higher-resourced settings where all women have close surveillance by experienced clinicians. In a larger WHO prospective study in South East Asian Hospitals, partograph used was found to be associated with reduced prolonged labour and stillbirth [21]. Inadequate intrapartum foetal monitoring may result in untimely execution of life-saving intervention, particularly in complicated deliveries. Thus, in low-resource settings partograph use is recommended for monitoring all women in labour, and can serve as a guide for timely referral to Comprehensive Emergency Obstetric Care (CEmOC) facilities [22].

onitoring may result in untimely execution of life-saving intervention, particularly in complicated deliveries. Thus, in low-resource settings partograph use is recommended for monitoring all women in labour, and can serve as a guide for timely referral to Comprehensive Emergency Obstetric Care (CEmOC) facilities [22]. Complications during pregnancy and childbirth have been long known to increase the risk of perinatal death. In our study a higher percentage of women admitted with obstetric complications lost their babies either during pregnancy, labour or shortly after delivery. This is consistent with findings by Cham et al. 2009 [9]. The high rate of stillbirth in multiple pregnancies and deliveries complicated by breech presentation in our study reaffirmed the call that screening for abnormal foetal presentations and multiple pregnancies could be one key component of antenatal care particularly in developing countries. Vanneste et al. demonstrated in one community-based study in Bangladesh that measurement of the fundal height had availed midwives the opportunity to identify a large proportion of women with twin pregnancies [18]. We speculate that timely referral of breech and twin pregnancies to an EOC facility will avert some of the adverse birth outcomes. Thus, refresher training in breech and twin delivery coupled with the presence of a skilled attended at all such deliveries seems warranted.

proportion of women with twin pregnancies [18]. We speculate that timely referral of breech and twin pregnancies to an EOC facility will avert some of the adverse birth outcomes. Thus, refresher training in breech and twin delivery coupled with the presence of a skilled attended at all such deliveries seems warranted. Most of the stillbirths in our study were fresh, with a fresh to macerated stillbirths ratio of 1.3 : 1. This indicates that most of the deaths probably occurred during labour, Fresh stillbirths are often used as proxy for stillbirths due to acute intrapartum insults [23]. The ratios of fresh to macerated stillbirths may indicate the availability and quality of prenatal and obstetric care, with high ratios suggesting inadequate or poor quality of EOC [12]. Thus, our results reaffirm the need for improved and timely access to EOC services during the intrapartum period in rural Gambia. In developed countries, less than 10% of all stillbirths are intrapartum stillbirths, while in developing countries almost one-half of all stillbirths are assumed to be intrapartum related [23]. The rate of intrapartum stillbirths have been substantially reduced in developed countries. This significant decline is mainly due to increased availability, quality, and access to EOC services and better intrapartum monitoring of high risk births [24]. Thus, understanding the burden of stillbirth has important programmatic and resource implications, which are of special interest in very low-resource settings like rural Gambia [25].

is mainly due to increased availability, quality, and access to EOC services and better intrapartum monitoring of high risk births [24]. Thus, understanding the burden of stillbirth has important programmatic and resource implications, which are of special interest in very low-resource settings like rural Gambia [25]. The unmet need for obstetric care is high in developing countries where most of the intrapartum stillbirths take place [26]. However, many of these deaths could be prevented with improved obstetric care [27]. The relatively low rate of intrapartum stillbirth in developed countries was to a larger extent due to the timely purveying of caesarean section [28]. Additionally, in developing countries caesarean section availability was associated with diminution in intrapartum stillbirth rates [27, 28]. However, nonavailability of EmOC, particularly caesarean section has been implicated as a risk factor for intrapartum stillbirth especially in cases of prolonged labour [29]. While caesarean section can be a life saving interposition for mother and child, evidence showed that its use, particularly in low-resource settings could be associated with increased risk of perinatal mortality, especially when it is performed late [30]. Consistent with the above findings, caesarean section deliveries in our study showed a higher proportion of fresh stillbirths when compared with a normal vaginal delivery. In addition, a larger proportion of stillbirths are observed in emergency c/s groups when compared with elective c/s group. We therefore speculate that caesarean sections are probably applied too late in these hospitals resulting to not saving the baby's life. This emphasises the importance of performing caesarean section (C/S) with the correct timing because early c/s could save more lives. Therefore, the increased risk of stillbirths must be considered when c/s is performed in the late stage of labour. The availability of quality EOC, pendent by emergency transport services and skilled providers is pivotal for effective maternal health services in The Gambia.

ming because early c/s could save more lives. Therefore, the increased risk of stillbirths must be considered when c/s is performed in the late stage of labour. The availability of quality EOC, pendent by emergency transport services and skilled providers is pivotal for effective maternal health services in The Gambia. 5. Conclusions Our findings suggest that the stillbirth rate is unacceptably high in rural hospitals in The Gambia. The findings also reaffirmed the important contribution of severe obstetric complications on the birth outcome in these settings. We also demonstrated an association between stillbirths and nonuse of the partograph, as well as with assisted breech and caesarean section. Most of the stillbirths were fresh, suggesting that these deaths have occurred during labour or shortly before delivery, thus potentially avertable. Improved intrapartum care through safe, comprehensive essential, and emergency obstetric supported by emergency transport services and skilled personnel is warranted for improved foetal out-comes in low resource settings such as The Gambia. Competing Interest No competing interest. The authors are responsible for the content and writing of this paper.

5. Conclusions Our findings suggest that the stillbirth rate is unacceptably high in rural hospitals in The Gambia. The findings also reaffirmed the important contribution of severe obstetric complications on the birth outcome in these settings. We also demonstrated an association between stillbirths and nonuse of the partograph, as well as with assisted breech and caesarean section. Most of the stillbirths were fresh, suggesting that these deaths have occurred during labour or shortly before delivery, thus potentially avertable. Improved intrapartum care through safe, comprehensive essential, and emergency obstetric supported by emergency transport services and skilled personnel is warranted for improved foetal out-comes in low resource settings such as The Gambia. Competing Interest No competing interest. The authors are responsible for the content and writing of this paper. Acknowledgments This project was funded by the Institute of General Practice and Community Medicine, University of Oslo, Norway. The authors' sincere thanks goes to the Chief Executive Officers (CEOs)/Principal Nursing officers of Bansang and Armed Forces Provisional Ruling Council (ARPRC) Hospital, and the Regional Director, and Regional Health Team-NBD/W for providing office space and support during the data collection period. The authors also deeply appreciate the cooperation received from the data managers, nurses, and midwives in the two hospitals. The authors are deeply indebted to our research assistants and driver for making the data abstraction possible.

m-NBD/W for providing office space and support during the data collection period. The authors also deeply appreciate the cooperation received from the data managers, nurses, and midwives in the two hospitals. The authors are deeply indebted to our research assistants and driver for making the data abstraction possible. Abbreviations MDGs: Millennium development goals EOC:Emergency obstetric care AFPRC:Armed forces provisional ruling council PHC:Primary health care TBA:Traditional birth attendant ENNM:Early neonatal neonatal mortality CEmOC:Comprehensive emergency obstetric care. Table 1 Demographic/reproductive and obstetric factors associated with stillbirth. Characteristics Total birth n (%) Stillbirth n (%) Crude OR (95% CI) Adjustedaa OR (95% CI) 1519 237(15.6) Maternal age (yrs) <20 509(33.5) 61(12.0) 1 1. 20–34 827(54.4) 140(16.9) 1.20(0.80–1.81) 0.97(0.57–1.65) ≥35 183(12.0) 36(19.7) 1.80(1.14–2.83)* 1.39(0.64–2.93) Parity 0 703(46.3) 77(11.0) 1 1. 1–3 380(25.0) 66(17.4) 1.31(0.92–1.86) 0.48(0.28–0.78) ≥4 436(28.7) 94(21.6) 2.24(1.61–3.10)*** 0.35(0.20–0.63) Residence PHC Village 473(31.1) 108(22.8) 2.13(1.59–2.80)*** 1.14(0.78–1.66) Non-PHC Village 1046(68.9) 129(12.3) 1 . Admission status Referred 303(19.9) 153(50.5) 13.75(10.02–18.84)*** 3.82(2.24–6.51)*** Booked 1216(80.1) 84(6.9) 1 1. Antenatal Care No 10(0.7) 5(50.0) 5.50(1.58–19.16)** 4.45(0.84–23.43) Yes 1509(99.3) 232(15.4) 1 1. Mode of delivery Spontaneous Vaginal 1340(88.2) 191(14.3) 1 1 Assisted breech 53(3.5) 17(32.1) 1.80(1.16–2.80)** 1.64(0.74–3.60) Caesarean Section 126(8.3) 29(23.0) 0.63(0.31–1.29) 4.35(2.46–7.70)*** Used of Patograph

Referred 303(19.9) 153(50.5) 13.75(10.02–18.84)*** 3.82(2.24–6.51)*** Booked 1216(80.1) 84(6.9) 1 1. Antenatal Care No 10(0.7) 5(50.0) 5.50(1.58–19.16)** 4.45(0.84–23.43) Yes 1509(99.3) 232(15.4) 1 1. Mode of delivery Spontaneous Vaginal 1340(88.2) 191(14.3) 1 1 Assisted breech 53(3.5) 17(32.1) 1.80(1.16–2.80)** 1.64(0.74–3.60) Caesarean Section 126(8.3) 29(23.0) 0.63(0.31–1.29) 4.35(2.46–7.70)*** Used of Patograph No 958(63.1) 187(19.5) 2.48(1.78–3.45)*** 1.70(1.13–2.56)** Yes 561(36.9) 50(8.9) 1 1. Obstetric complication Yes 370 164(44.3) 11.74(8.58–16.06)*** 6.68(3.84–11.62)*** No 1149 73(6.4) 1 1. Type of birth Multiple 111(7.3) 20(18.0) 1.20(0.73–1.99) 2.01(1.05–3.86)** Singleton 1408(92.7) 217(15.4) 1 1. Birth weight <2500 g 278(18.3) 114(41.0) 6.32(4.67–8.55)*** 4.48(3.04–6.59)*** ≥2500 g 1241(81.7) 123(9.9) 1 aaAdjusted for all variables listed in the table. *P-value <.01, **P-value <.05, ***P-value <.001. Table 2 Maternal demographic and obstetric factors associated with fresh stillbirth (FSB) (%). Profile FSB (%) Crude OR (95% CI) Adjustedaa OR (95% CI) Age (years) <20 54.1 0.72(0.39–1.32) 0.76(0.27–2.15) 20–34 62.1 1.32(0.58–3.01) 1.91(0.42–8.41) ≥35 47.2 1 1 Parity 0 59.7 1.24(0.64–2.40) 1.08(0.37–3.12) 1–3 54.5 1 1 ≥4 58.5 1.05(0.57–1.94) 0.48(0.15–1.48) Recruiting hospital Bansang 49.6 1 1 AFPRC 67.9 2.15(1.26–3.66)** 1.42(0.66–3.01) Residence PHC Village 56.5 1 1 Non-PHC Village 58.9 1.11(0.66–1.85) 0.78(0.38–1.58) Antepartum admission Yes 52.8 0.65(0.39–1.10) 0.80(0.30–2.10) No 63.2 1 1 Timing of complication Antepartum 50.4 1 1 Intrapartum 78.4 3.57(1.52–8.40)** 3.14(1.01–9.76)** Partograph used

Recruiting hospital Bansang 49.6 1 1 AFPRC 67.9 2.15(1.26–3.66)** 1.42(0.66–3.01) Residence PHC Village 56.5 1 1 Non-PHC Village 58.9 1.11(0.66–1.85) 0.78(0.38–1.58) Antepartum admission Yes 52.8 0.65(0.39–1.10) 0.80(0.30–2.10) No 63.2 1 1 Timing of complication Antepartum 50.4 1 1 Intrapartum 78.4 3.57(1.52–8.40)** 3.14(1.01–9.76)** Partograph used No 58.3 1.10(0.59–2.06) 1.73(0.70–4.30) Yes 56.0 1 1 Mode of delivery Spontaneous vaginalBreech/others 55.5 1 1 Breech/others 64.7 1.43(0.40–5.18) 1.23(0.42–3.57 Caesarean Section 72.4 2.15(0.91–5.11) 1.32(0.32–5.45) Birth weight (grms) <2500 48.2 1 ≥2500 67.7 2.15(1.27–3.63)** 1.67(0.81–3.44) aaAdjusted for all variables listed in the table. **P-value <.05.

1. Introduction Women with high-grade Papanicolaou (Pap) smears have a 43%–66% risk of having moderate- to high-grade intraepithelial neoplasia (CIN 2 or CIN 3) on subsequent biopsy and a 2% risk of having invasive cancer [1]. However, when discrepancies occur between Pap smear cytology and cervical biopsy histology, this can cause a clinical dilemma. We define discrepancy as patients with High-Grade Squamous Intraepithelial Lesion (HGSIL) Pap smear followed by cervical biopsies with Cervical Intraepithelial Neoplasia (CIN) 1 histology or less. In cases of discrepancy, the American Society for Colposcopy and Cervical Pathology (ASCCP) previously favored an excisional procedure for diagnosis in nonpregnant patients when no lesion or CIN1 is identified with satisfactory colposcopy [2]. As we develop a better understanding of HPV infection, its clinical course, and prognosis, more people are considering conservative management. The most recent ASCCP guidelines in 2006 included the option of repeating Pap smear and colposcopy every 6 months for 1 year in women over 20 with discrepancy and recommended this conservative monitoring pathway in the adolescent population [3].

ourse, and prognosis, more people are considering conservative management. The most recent ASCCP guidelines in 2006 included the option of repeating Pap smear and colposcopy every 6 months for 1 year in women over 20 with discrepancy and recommended this conservative monitoring pathway in the adolescent population [3]. In nonadolescent patients, a loop electrosurgical excisional procedure (LEEP) is a reasonable management option for discrepancy, as there is a concern that the high-grade lesion found on Pap smear was missed on the colposcopically directed biopsy. However, we know that as many as 35% of women with HGSIL will regress spontaneously, making the excisional procedure unnecessary [4]. Data evaluating the time from the initial HGSIL cytology to the LEEP, as it impacts likelihood of finding significant pathology in the LEEP specimen, has not been well characterized. Our objective was to determine whether the time from initial HGSIL Pap to LEEP, when done for discrepancy, affects the pathologic grade of CIN in the LEEP specimen. We hypothesized that, as the time interval between the initial HGSIL Pap and subsequent LEEP for discrepancy increased, the likelihood of finding CIN 2-3 in the LEEP specimen would decrease.

her the time from initial HGSIL Pap to LEEP, when done for discrepancy, affects the pathologic grade of CIN in the LEEP specimen. We hypothesized that, as the time interval between the initial HGSIL Pap and subsequent LEEP for discrepancy increased, the likelihood of finding CIN 2-3 in the LEEP specimen would decrease. 2. Materials and Methods This was a retrospective case-control study of all LEEP procedures done for discrepancy at the University of Washington Medical Center and Harborview Medical Center in Seattle, Washington. We identified potential cases by performing a search in the pathology database to identify all LEEP specimens between January 1, 1997 and August 31, 2007. Of note, the Dysplasia Clinics at these two medical centers serve as referral centers for women diagnosed with abnormal cytology at local clinics. Thus, not all initial Pap smears originated from these two medical centers. In the two Dysplasia Clinics, a standardized colposcopy form is used to record impression and results of colposcopy, but the Pap smear is not routinely repeated, nor do we have a complete history of prior Pap smears. Based on colposcopic findings, biopsies were performed at the discretion of the attending physician.

s. In the two Dysplasia Clinics, a standardized colposcopy form is used to record impression and results of colposcopy, but the Pap smear is not routinely repeated, nor do we have a complete history of prior Pap smears. Based on colposcopic findings, biopsies were performed at the discretion of the attending physician. To determine which LEEPs were done for discrepancy, we linked all LEEPs with their respective preceding Pap smear, colposcopy report, and cervical biopsy result. We included all women, 18–49 years old, who had pathologic discrepancy as defined by HGSIL Pap smear and a subsequent colposcopy with cervical biopsies of CIN 1 or less. We excluded those who had an unsatisfactory colposcopy exam, previous excisional procedure, and positive endocervical curettage and those who were pregnant or HIV positive. Once discrepant cases were identified, we performed a chart review and used a standardized data collection instrument to gather general demographic information and risk factors associated with progression of cervical neoplasia (smoking, number of partners, age of coitarche, history of sexually transmitted infections, and birth control method). On this form we also documented cytologic and histologic results of the Pap smear, cervical biopsy, and LEEP procedure, and recorded the length of time between the Pap smear and the LEEP.

cal neoplasia (smoking, number of partners, age of coitarche, history of sexually transmitted infections, and birth control method). On this form we also documented cytologic and histologic results of the Pap smear, cervical biopsy, and LEEP procedure, and recorded the length of time between the Pap smear and the LEEP. A power calculation was based on the assumption that, in LEEPs done for discrepancy, the prevalence of CIN 2 or 3 found in the LEEP specimen will be higher when the LEEP was performed closer to the time of the HGSIL Pap. We divided the time intervals from Pap smear to LEEP into three intervals: less than three months, between three and five months, and greater than five months. As this was a retrospective study, we used our data of 55% prevalence of CIN 2,3 in LEEP specimens when the time interval from the Pap smear to the LEEP is less then 3 months and 16% when the interval is greater than five months. From this, the calculated sample size was 23 women to give 80% power and α = 0.05 to detect this difference in pathologic result, demonstrating that the study was appropriately powered to test our hypothesis.

val from the Pap smear to the LEEP is less then 3 months and 16% when the interval is greater than five months. From this, the calculated sample size was 23 women to give 80% power and α = 0.05 to detect this difference in pathologic result, demonstrating that the study was appropriately powered to test our hypothesis. Based on their LEEP histology, subjects were separated into two groups: CIN 1 or less and CIN 2,3. The “CIN 1 or less” group included normal, cervicitis, and CIN 1. The “CIN 2,3” group included CIN 2 and CIN 3. We chose these two groups based on clinical application, as this is the general division that determines treatment management. Statistical analysis was executed with SPSS v.16. Differences between the two groups were examined using Student's t-test and χ². Univariate and multivariate logistic regressions were used to calculate odds ratios and 95% confidence intervals for the association between time since HGSIL and histologic grade of LEEP. This study was approved by the International Review Board at the University of Washington, IRB Application Number 07-8885-E/A 01.

Based on their LEEP histology, subjects were separated into two groups: CIN 1 or less and CIN 2,3. The “CIN 1 or less” group included normal, cervicitis, and CIN 1. The “CIN 2,3” group included CIN 2 and CIN 3. We chose these two groups based on clinical application, as this is the general division that determines treatment management. Statistical analysis was executed with SPSS v.16. Differences between the two groups were examined using Student's t-test and χ². Univariate and multivariate logistic regressions were used to calculate odds ratios and 95% confidence intervals for the association between time since HGSIL and histologic grade of LEEP. This study was approved by the International Review Board at the University of Washington, IRB Application Number 07-8885-E/A 01. 3. Results Of the 1,356 patients who underwent a LEEP during this time period, 157 were performed for discrepancy. Of these, 36 women were excluded: 24 had unsatisfactory colposcopy, six had endocervical curettings positive for neoplasia, four were HIV positive, and two were older than 49. The 121 remaining patients were divided into two groups based on the pathologic grade of CIN in their LEEP specimen (CIN 1 or CIN 2,3). As demonstrated by Table 1, the two groups were similar with regard to age, ethnicity, parity, tobacco use, coitarche, number of partners, and history of sexually transmitted infections. There was a difference noted in the use of birth control method, likely due to the number of Depo-Provera users in the CIN 2,3 group.

As demonstrated by Table 1, the two groups were similar with regard to age, ethnicity, parity, tobacco use, coitarche, number of partners, and history of sexually transmitted infections. There was a difference noted in the use of birth control method, likely due to the number of Depo-Provera users in the CIN 2,3 group. Of the 121 patients who underwent a LEEP for discrepancy, 67 patients (55.4%) had CIN 2,3 on their LEEP pathology specimen and 54 (44.6%) had CIN 1 or less. We examined the time interval from the initial HGSIL Pap smear to the LEEP procedure in the CIN 2,3 group. Of the 67 who had CIN 2,3 on LEEP histology, 37 (55.2%) had their LEEP within three months of the initial Pap versus 11 (16.4%) who had their LEEP greater than five months after the Pap smear (Figure 1). Although there was no significant difference in the time interval between the groups, the Pearson correlation coefficient, a measure of linear trend, was −0.15 (P = .096). Though not statistically significant, this suggests a trend toward decreased prevalence of CIN 2,3 in the LEEP specimen as the time from HGSIL Pap smear increased. In order to explore this further, a univariate logistic regression was performed to test for trend of LEEP histology over time using CIN 2,3 as the outcome (Table 2). In this regression as time interval increased, the likelihood of CIN 2,3 decreased, though the effect did not reach statistical significance.

increased. In order to explore this further, a univariate logistic regression was performed to test for trend of LEEP histology over time using CIN 2,3 as the outcome (Table 2). In this regression as time interval increased, the likelihood of CIN 2,3 decreased, though the effect did not reach statistical significance. Although it was not our initial intent to examine the effect of birth control method on the histologic grade of the LEEP specimens, there was a significant difference in the use of Depo-Provera between women with low-grade and moderate- or high-grade CIN (3.8% in CIN 1 versus 15.2% in CIN 2,3). We therefore performed a multivariate regression using the covariates of time interval and birth control method to analyze the effect of birth control method on dysplasia (Table 3). Birth control method was divided into three groups: nonhormonal users (no method, tubal ligation, barrier methods, and Copper IUD), estrogen and progesterone combined methods (oral contraceptive pills, NuvaRing, patch), and progesterone only methods (Depo-Provera). Depo-Provera users were more likely to have CIN 2,3 in their LEEP specimen (OR 7.59, P = .02). Because of this significant relationship, we performed the multivariate regression in the CIN 2,3 group controlling for contraceptive method. After controlling for the birth control variable, we demonstrated a trend toward fewer CIN 2,3 LEEP findings when the procedure was performed more than five months from the HGSIL Pap, a trend that neared but did not achieve significance (OR 0.39, P = .07).

sion in the CIN 2,3 group controlling for contraceptive method. After controlling for the birth control variable, we demonstrated a trend toward fewer CIN 2,3 LEEP findings when the procedure was performed more than five months from the HGSIL Pap, a trend that neared but did not achieve significance (OR 0.39, P = .07). 4. Discussion This study was initiated prior to the 2006 ASCCP Consensus Guidelines, which were released in October 2007. New changes announced in the guidelines included those that direct the management of abnormal cytology in adolescents of 20 years and younger. For adolescents in the discrepant situation, it is now recommended that they be observed and monitored with Pap smears and colposcopy examinations every 6 months for 24 months [3]. Conservative management in this group has been favored, as we know that the majority of HPV infections will clear without treatment within two years and have negligible long-standing clinical significance [5]. In addition, although generally well tolerated, LEEPs are not without complications, including risks to future pregnancies such as premature delivery, premature rupture of membranes, and low birth weight [6–8]. We did not examine adolescents separately in this study as there were only 14 subjects that were 20 years or younger at the time of their LEEP. However, our results suggest that the cautious approach in treating adolescents that the ASCCP is now advising is not detrimental.

ure of membranes, and low birth weight [6–8]. We did not examine adolescents separately in this study as there were only 14 subjects that were 20 years or younger at the time of their LEEP. However, our results suggest that the cautious approach in treating adolescents that the ASCCP is now advising is not detrimental. The role of exogenous estradiol and progesterone in the development of cervical cancer has been discussed extensively. There is still insufficient data as to whether oral contraceptive pills are associated with the development of cervical cancer in HPV-infected women [9]. Thomas et al. showed an increase risk of CIN 3 but not invasive carcinoma in Depo-Provera users, a risk that increased with the length of use [10]. Many have examined the immunohistochemical presence of estrogen and progesterone receptors within the cervix and compared the makeup of receptors in normal cervix with those that contain neoplasia [11]. Konishi et al. have suggested that HPV infection may increase progesterone receptor expression that is found in neoplastic cervical squamous cells [12]. Progesterone has been hypothesized to be a modulator of the immune system [13]. In this way, the exposure to Depo-Provera may decrease the clearance of HPV infection. However, the use of hormonal birth control was not the focus of this study, and as a result this study was not powered to examine this association. Still, this is an intriguing finding that may merit further investigation.

stem [13]. In this way, the exposure to Depo-Provera may decrease the clearance of HPV infection. However, the use of hormonal birth control was not the focus of this study, and as a result this study was not powered to examine this association. Still, this is an intriguing finding that may merit further investigation. One limitation of our study is that the University of Washington and Harborview Medical Centers are referral centers for cervical neoplasia. As many patients are referred from providers outside the system, original Pap smears are not always available to us to confirm the HGSIL nor is the Pap smear routinely repeated at the time of colposcopy. Studies demonstrate that pathologists vary in their determination of cytologic grading of Pap smears [14]. If there is a tendency to err on side of overcalling the Pap smear HGSIL and the patient has a colposcopy consistent with CIN1 or less, this would lead to discrepancy. A LEEP done in this scenario would likely have lower-grade CIN than in a situation where a pathologist is more likely to read LGSIL on the original Pap smear.

re is a tendency to err on side of overcalling the Pap smear HGSIL and the patient has a colposcopy consistent with CIN1 or less, this would lead to discrepancy. A LEEP done in this scenario would likely have lower-grade CIN than in a situation where a pathologist is more likely to read LGSIL on the original Pap smear. Another limitation of this study is the inherent disadvantage of inter observer variation on colposcopy examination because, in our clinics, multiple providers perform colposcopy. We know that colposcopy is imperfect and up to 33% of CIN 2,3 can be missed on a single colposcopy examination [15]. Missing the diagnosis of CIN 2,3 on colposcopy and cervical biopsies after an HGSIL pap smear places the patient in a discrepant situation and previously obligated them to a LEEP. However, this would likely have given us a higher number of CIN 2,3 on LEEP histology, as it may have been present but missed on the preceding colposcopy examination. In addition, we used medical records of LEEP pathology to identify cases of discrepancy. This may open the potential of selection bias as some cases could be missed, or inadvertently miscoded. These LEEPs were identified, though, as a pathology database, and Table 1 shows that our two groups were similar.

Another limitation of this study is the inherent disadvantage of inter observer variation on colposcopy examination because, in our clinics, multiple providers perform colposcopy. We know that colposcopy is imperfect and up to 33% of CIN 2,3 can be missed on a single colposcopy examination [15]. Missing the diagnosis of CIN 2,3 on colposcopy and cervical biopsies after an HGSIL pap smear places the patient in a discrepant situation and previously obligated them to a LEEP. However, this would likely have given us a higher number of CIN 2,3 on LEEP histology, as it may have been present but missed on the preceding colposcopy examination. In addition, we used medical records of LEEP pathology to identify cases of discrepancy. This may open the potential of selection bias as some cases could be missed, or inadvertently miscoded. These LEEPs were identified, though, as a pathology database, and Table 1 shows that our two groups were similar. Finally, our examination of birth control method as it affected the LEEP pathology was not adequately investigated. There are many subtle effects of birth control on cervical dysplasia. We grouped non-hormonal methods together, when in fact this may represent a heterogeneous group of women. Women who are using no method may or may not be abstaining from intercourse. Those that are abstaining may have less exposure to HPV, which could lead to a higher rate of HPV regression. In the same way, we would expect condom users to have less exposure to HPV as Winer et al. have shown that condoms prevent HPV transmission [16]. A separate study would be required to elucidate the role of all the various birth control methods as they influence cervical dysplasia discrepancies over time.

PV regression. In the same way, we would expect condom users to have less exposure to HPV as Winer et al. have shown that condoms prevent HPV transmission [16]. A separate study would be required to elucidate the role of all the various birth control methods as they influence cervical dysplasia discrepancies over time. 5. Conclusions To our knowledge, this is the first study that examines the effect of time on histologic grade of CIN in the LEEP specimen when LEEP is performed for discrepancy. Research demonstrates that the neoplasia caused by HPV may regress spontaneously with time [4, 5]. Hence, we sought to determine whether the length of time between the initial HGSIL Pap smear and a LEEP performed for discrepant pathology on cervical biopsy affects the pathology results of the LEEP specimen. We demonstrated a trend suggesting that, when LEEP is performed more than 5 months from the initial HGSIL Pap smear, isolating CIN 2,3 in the LEEP specimen is less likely. This trend was more pronounced in the multivariate regression controlling for birth control method, suggesting that birth control method may impact regression of HPV. When controlling for birth control method, the trend toward decrease in CIN 2,3 with time was more clear as there were more women using Depo-Provera who had CIN 2,3 and had their LEEP greater than five months from their HGSIL Pap smear.

control method, suggesting that birth control method may impact regression of HPV. When controlling for birth control method, the trend toward decrease in CIN 2,3 with time was more clear as there were more women using Depo-Provera who had CIN 2,3 and had their LEEP greater than five months from their HGSIL Pap smear. Many have explored the option of “See-and-Treat” for HGSIL management with immediate LEEP in recent years. The 2006 ASCCP Guidelines advocate caution and observation rather than an immediate excisional procedure in women who are still considering childbearing due to the increased pregnancy complications [3]. We cannot directly extrapolate our data into the risk of overtreatment with immediate LEEP. Whenever colposcopy is performed, one must also take into account the immunological stimulation that is thought to accompany colposcopic biopsy of cervical neoplasia that hastens resolution [17]. In this study alone if all patients went to an immediate LEEP, at most 45% of women may have had CIN 1 or less on LEEP and would have been overtreated. Some studies have shown a much lower overtreatment rate when the time interval from Pap to LEEP is diminished [18, 19]. However, it would be difficult to advocate for immediate LEEP in our patient population based on the percentages of this study. Given the propensity of HPV clearance over time, a “see-and-treat” approach deserves scrutiny. This study suggests that, when in a discrepant situation, the variable of time between the initial HGSIL Pap and LEEP may be a consideration in determining whether a LEEP is appropriate.

ased on the percentages of this study. Given the propensity of HPV clearance over time, a “see-and-treat” approach deserves scrutiny. This study suggests that, when in a discrepant situation, the variable of time between the initial HGSIL Pap and LEEP may be a consideration in determining whether a LEEP is appropriate. Acknowledgment This paper was originally presented as a poster presentation at the Pacific Coast Obstetrical and Gynecological Society Annual Meeting in Victoria, BC, Canada, 10/16/2008. Figure 1 Subjects were divided into two groups by their LEEP histology (CIN 1 or CIN 2,3), and we examined the time interval from their HGSIL Pap to their LEEP. The blue bars signify those subjects who had their LEEP done less than 3 months from the HGSIL Pap. The red bars are those who had their LEEP 3–5 months from the HGSIL Pap. The green bars are those who had their LEEP after 5 months of the HGSIL Pap. Table 1 Demographic characteristics comparing women with CIN 1 or CIN 2,3 on LEEP.

Figure 1 Subjects were divided into two groups by their LEEP histology (CIN 1 or CIN 2,3), and we examined the time interval from their HGSIL Pap to their LEEP. The blue bars signify those subjects who had their LEEP done less than 3 months from the HGSIL Pap. The red bars are those who had their LEEP 3–5 months from the HGSIL Pap. The green bars are those who had their LEEP after 5 months of the HGSIL Pap. Table 1 Demographic characteristics comparing women with CIN 1 or CIN 2,3 on LEEP. Characteristic CIN 1 or less CIN 2-3 P n = 54 (%) n = 67 (%) Age: (years) ± SD 28.2 ± 8.7 27.2 ± 5.7 .46ª Ethnicity Caucasian 35 (66) 37 (55) .48b African American 5 (9) 5 (8) Hispanic 4 (8) 9 (13) Asian 6 (11) 7 (10) Other/Unknown 4 (6) 9 (13) Parity 0 36 (67) 38/65 (59) .06b 1 13 (24) 8/65 (12) 2 2 (4) 13/65 (20) >2 3 (6) 6/65 (9) Tobacco None 41 (76) 46 (69) .10b <1 ppd 12 (23) 11 (16) 1 ppd 1 (2) 8 (12) 2 ppd 0 (0) 2 (3) Coitarche: (years) ± SD 16.6 ± 2.0 16.5 ± 2.2 .81a No. of Partners <5 25/49 (51) 22/58 (38) .17b >5 24/49 (49) 36/58 (62) History of STIs None 39/51 (77) 44/65 (68) .92b CT 3/51 (6) 7/65 (11) GC 0/51 (0) 1/65 (2) Genital Herpes 2/51 (4) 3/65 (5) Genital Warts 2/51 (4) 3/65 (5) PID 1/51 (2) 1/65 (2) More than 1 4/51 (8) 6/65 (9) BCM None 8/52 (15) 1/66 (2) .01b Condoms 7/52 (14) 6/66 (9) Oral Contraceptive Pills 31/52 (60) 44/66 (67) Depo-Provera 2/52 (4) 10/66 (15) NuvaRing 1/52 (2) 1/66 (2) Patch 3/52 (6) 0 (0) BTL 0/52(0) 3/66 (5) Copper IUD 0/52 (0) 1/66 (2) Denominators may vary due to missing data. ª Student's t-test, b Chi-square.

Characteristic CIN 1 or less CIN 2-3 P n = 54 (%) n = 67 (%) Age: (years) ± SD 28.2 ± 8.7 27.2 ± 5.7 .46ª Ethnicity Caucasian 35 (66) 37 (55) .48b African American 5 (9) 5 (8) Hispanic 4 (8) 9 (13) Asian 6 (11) 7 (10) Other/Unknown 4 (6) 9 (13) Parity 0 36 (67) 38/65 (59) .06b 1 13 (24) 8/65 (12) 2 2 (4) 13/65 (20) >2 3 (6) 6/65 (9) Tobacco None 41 (76) 46 (69) .10b <1 ppd 12 (23) 11 (16) 1 ppd 1 (2) 8 (12) 2 ppd 0 (0) 2 (3) Coitarche: (years) ± SD 16.6 ± 2.0 16.5 ± 2.2 .81a No. of Partners <5 25/49 (51) 22/58 (38) .17b >5 24/49 (49) 36/58 (62) History of STIs None 39/51 (77) 44/65 (68) .92b CT 3/51 (6) 7/65 (11) GC 0/51 (0) 1/65 (2) Genital Herpes 2/51 (4) 3/65 (5) Genital Warts 2/51 (4) 3/65 (5) PID 1/51 (2) 1/65 (2) More than 1 4/51 (8) 6/65 (9) BCM None 8/52 (15) 1/66 (2) .01b Condoms 7/52 (14) 6/66 (9) Oral Contraceptive Pills 31/52 (60) 44/66 (67) Depo-Provera 2/52 (4) 10/66 (15) NuvaRing 1/52 (2) 1/66 (2) Patch 3/52 (6) 0 (0) BTL 0/52(0) 3/66 (5) Copper IUD 0/52 (0) 1/66 (2) Denominators may vary due to missing data. ª Student's t-test, b Chi-square. Table 2 Univariate logistic regression of time interval and CIN 2,3 in LEEP. Time interval (categorical) CIN 2,3 in LEEP OR (95% CI) P <3 months 1.0 (Reference) — 3–5 months 0.63 (0.27–1.44) .27 >5 months 0.47 (0.18–1.21) .12 Table 3 Univariate logistic regression with birth control method and CIN 2,3 in LEEP.

b Chi-square. Table 2 Univariate logistic regression of time interval and CIN 2,3 in LEEP. Time interval (categorical) CIN 2,3 in LEEP OR (95% CI) P <3 months 1.0 (Reference) — 3–5 months 0.63 (0.27–1.44) .27 >5 months 0.47 (0.18–1.21) .12 Table 3 Univariate logistic regression with birth control method and CIN 2,3 in LEEP. Variable CIN 2,3 in LEEP OR (95% CI) P Birth control method None/non-hormonal 1.0 (Reference) — Estrogen + Progesterone 1.83 (0.73–4.58) 0.20 Progesterone only 7.59 (1.34–43.2) 0.02 Table 4 Multivariate Regression of Time Interval and CIN 2,3 controlling for Birth Control Method. Variable CIN 2,3 in LEEP OR (95% CI) P Time interval (categorical) <3 months (Reference) — 3–5 months 0.66 (0.27–1.59) 0.35 >5 months 0.39 (0.14–1.07) 0.07

1. Introduction Trisomy 9 mosaicism is a rare chromosomal abnormality that manifests with multiple anomalies, such as facial, cardiac, osteal, genitourinary, and respiratory abnormalities. More than 50 cases have been reported, most of which were diagnosed after birth. As cases diagnosed prenatally usually culminate in induced abortions [1–4], the natural history of fetuses with trisomy 9 mosaicism remains unknown. We report three cases of trisomy 9 mosaicism diagnosed in utero with ongoing pregnancies. 2. Case Report A 36-year-old primigravida was referred to our institute at 29 weeks of gestation because of fetal growth restriction (FGR). A fetal ultrasound examination demonstrated severe asymmetric FGR (<−3.0 standard deviation [SD]) and a single umbilical artery (SUA). An amniocentesis revealed that 27 were normal 46, XX cells and 3 cells (10%) were 47, XX, +9. An intrauterine fetal demise (IUFD) was confirmed at 33 weeks of gestation. The fetus was a 915 g female with a large forehead, a bulbous nose, and micrognathia. The placental weight was 150 g. An autopsy revealed an abnormal lobulation of the right lung.

t 27 were normal 46, XX cells and 3 cells (10%) were 47, XX, +9. An intrauterine fetal demise (IUFD) was confirmed at 33 weeks of gestation. The fetus was a 915 g female with a large forehead, a bulbous nose, and micrognathia. The placental weight was 150 g. An autopsy revealed an abnormal lobulation of the right lung. The second case was that of a 36-year-old primigravida. She was referred to our institute at 31 weeks of gestation due to a left-sided congenital diaphragmatic hernia. The estimated fetal body weight by ultrasound was 1408 g (−1.7 SD). The fetal karyotype by amniocentesis indicated trisomy 9 mosaicism with 29% (6/21 cells) trisomic cells. At 37 weeks of gestation, a 1506 g male was delivered by elective cesarean section. The placenta weighed 350 g. The diaphragmatic hernia was repaired on day 2 of life, followed by a gastrostomy and bronchotomy in the 1st year. Although he had normal G-banding results on postnatal blood karyotyping, interphase FISH performed on abdominal wall muscle tissue obtained during the gastrostomy revealed a mosaic trisomy 9 karyotype. This case has been reported to highlight the cytogenetic discrepancy between amniocytes and postnatal blood [5]. He is now 4 years old and suffers from severe developmental delay.

aryotyping, interphase FISH performed on abdominal wall muscle tissue obtained during the gastrostomy revealed a mosaic trisomy 9 karyotype. This case has been reported to highlight the cytogenetic discrepancy between amniocytes and postnatal blood [5]. He is now 4 years old and suffers from severe developmental delay. The third case involved a 41-year-old primigravida. She was referred for evaluation of FGR at 26 weeks of gestation. An ultrasound examination revealed severe asymmetric FGR 530 g (−3.2 SD) and an SUA. Chromosomes from amniocentesis revealed a mosaic trisomy 9 constitution, as follows: 46, XX (18 cells)/47, XX, +9 (11 cells). An IUFD was confirmed at 38 weeks of gestation. The fetus was a 1220 g female. The placenta weighed 107 g. An autopsy revealed no major abnormalities except facial dysmorphism with a bulbous nose, micrognathia, and low-set ears. 3. Discussion We have reported three cases of trisomy 9 mosaicism diagnosed by amniocentesis with ongoing pregnancies. FGR was observed prenatally, and specific facial findings were observed after birth in all the cases. Two of three cases, which showed severe FGR and SUA, resulted in IUFDs in the third trimester. The other case with mild FGR resulted in a live birth and had severe developmental delay. Fetuses with trisomy 9 mosaicism in which the major prenatal finding is FGR rarely survive, even though they have no major anomalies.

. Two of three cases, which showed severe FGR and SUA, resulted in IUFDs in the third trimester. The other case with mild FGR resulted in a live birth and had severe developmental delay. Fetuses with trisomy 9 mosaicism in which the major prenatal finding is FGR rarely survive, even though they have no major anomalies. Trisomy 9 mosaicism manifests with various abnormalities. Facial anomalies, such as low-set ears, micrognathia, and bulbous nose, are universal. Cleft lip is expressed in about 20% of cases [6]. Orthopaedic abnormalities, including hip dislocations and arthrogryposis, are also highly expressed. Wide-ranging malformations, such as congenital heart disease, intracerebral lesions, and urogenital abnormalities, are also observed (Table 1). Our three cases had common specific facial findings, such as micrognathia and bulbous noses. Other malformations included a congenital diaphragmatic hernia and an abnormal lobulation of the lung; these visceral abnormalities are rare.

se, intracerebral lesions, and urogenital abnormalities, are also observed (Table 1). Our three cases had common specific facial findings, such as micrognathia and bulbous noses. Other malformations included a congenital diaphragmatic hernia and an abnormal lobulation of the lung; these visceral abnormalities are rare. Postpartum cytogenetic conformation of trisomy 9 mosaicism was performed in only one case. If mosaicism is detected in any one tissue, it is expected that different tissues will have different levels of mosaicism and consequently the clinical presentation of each case may vary considerably. The case which underwent cytogenetic analysis postnatally revealed cytogenetic discrepancies between the tissues [5]. There was a normal karyotype in the blood and trisomy 9 mosaicism in the abdominal wall tissue. In this case, the infant demonstrated mild FGR with a congenital diaphragmatic hernia. When a cytogenetic discrepancy between amniocytes and postnatal blood is observed, it is not rare for mosaic tissue to be confined to a specific organ [5, 7]. We failed to perform postnatal cytogenetic analysis on the other two cases which resulted in IUFDs. It is difficult to obtain usual G-banding karyotype results on specimens from stillborn infants. However, these two cases were coincident with a phenotype of trisomy 9 mosaicism, such as a specific facial appearance, and the two stillborn infants were considered to have trisomy 9 mosaicism.

ses which resulted in IUFDs. It is difficult to obtain usual G-banding karyotype results on specimens from stillborn infants. However, these two cases were coincident with a phenotype of trisomy 9 mosaicism, such as a specific facial appearance, and the two stillborn infants were considered to have trisomy 9 mosaicism. Our search revealed 20 reported prenatal cases of trisomy 9 mosaicism [3, 8–12]. Of the 20 cases, 11 resulted in induced abortions, and the pregnancies were not interrupted in the remaining 9 cases (Table 2). Of these 9 cases, 8 led to live births and 1 case resulted in an IUFD [12]. In contrast, two of three cases resulted in IUFDs in this report. In general, many fetuses with common trisomies are lost before birth. In terms of trisomy 21, which is the most common aneuploidy, the estimated fetal death rate is 30% between the 2nd trimester and term [13]. Chromosomal abnormalities were present in 38% of anomalous stillborns, and 4.6% of morphologically normal fetuses of 750 stillbirths [14]. It is not surprising, therefore, that two of three cases of trisomy 9 mosaicism resulted in IUFDs. The main feature of trisomy 9 mosaicism IUFDs is severe growth restriction. Two of the three cases we evaluated had severe FGR without major congenital abnormalities and resulted in IUFDs. According to reported 12 cases including our own cases (Table 2), three cases that resulted in IUFDs involved severe FGR. Adversely, the other 9 cases of liveborn did not reveal severe FGR. Arnold reviewed 23 cases of trisomy 9 mosaicism; most were diagnosed postnatally with an average birth weight of 2690 g, which is consistent with mild FGR [6]. In addition, reported cases incidentally detected by midtrimester routine amniocentesis associated with maternal advanced age led to live births uneventfully (Table 2). Trisomy 9 mosaicism can result in IUFDs when there is severe FGR, even though there are no structural abnormalities. However, it is still unknown whether the degree of FGR correlates with IUFDs in chromosomal abnormalities.

e amniocentesis associated with maternal advanced age led to live births uneventfully (Table 2). Trisomy 9 mosaicism can result in IUFDs when there is severe FGR, even though there are no structural abnormalities. However, it is still unknown whether the degree of FGR correlates with IUFDs in chromosomal abnormalities. Our cases contribute to clarifying the natural history of trisomy 9 mosaicism diagnosed in utero. Trisomy 9 mosaicism fetuses with severe FGR can rarely survive. These findings are helpful for genetic counseling for trisomy 9 mosaicism diagnosed. Conflict of Interest The authors declare no conflicts of interest. Table 1 Clinical findings of trisomy 9 mosaicism with high frequency. Prenatally Fetal growth restriction Ventricular septal defect Micrognathia Single umbilical artery Amniotic fluid disorders Postnatally Craniofacial abnormalities Wide fontanels Microphthalmia Bulbous nose Low-set ears Micrognathia Heart abnormalities Ventricular septal defect Patent ductus arteriosus Genitourinary malformations Hydronephrosis Microkidneys Musculoskeletal abnormalities Dislocated hips Joint limitation Overlapping fingers Abnormal ossification Table 2 Characteristics and outcomes of reported cases of prenatal trisomy 9 mosaicism.

Heart abnormalities Ventricular septal defect Patent ductus arteriosus Genitourinary malformations Hydronephrosis Microkidneys Musculoskeletal abnormalities Dislocated hips Joint limitation Overlapping fingers Abnormal ossification Table 2 Characteristics and outcomes of reported cases of prenatal trisomy 9 mosaicism. Case Reference Maternal age (years) Time of exam (weeks) Trigger of detection FGR Other fetal findings Outcomes 1 Bureau et al. [8] 24 36 CNS anomaly − Dandy-Walker variant Liveborn, death at 2 weeks 2 Saura et al. [3] 28 30 PDA, polyhydramnios − PDA Liveborn at 35 weeks 3 29 33 FGR + — Alive at 21 months; severe developmental delay 4 Greenberg et al. [9] 39 16 AAMA − — Liveborn at term 5 Hsu et al. [10] 48 NA AAMA − — Normal development at 3 years, 8 months 6 42 NA AAMA − — Liveborn at 23 weeks; death at 4 days 7 38 NA AAMA + Multiple anomalies (no details) Liveborn at term 8 Sherer et al. [11] 20 31 FGR Mild Right hydronephrosis liveborn, death at 6 weeks 9 Smoleniec et al. [12] 28 34 Severe FGR Severe — IUFD at 34 weeks 10 Case 1 35 29 Severe FGR Severe — IUFD at 33 weeks 11 Case 2 in [5] 36 31 CDH Mild Left-sided congenital diaphragmatic hernia Alive at 4 years; severe developmental delay 12 Case 3 41 24 Severe FGR Severe — IUFD at 38 weeks AAMA: amniocentesis with advanced maternal age; CNS: central nerve system; FGR: fetal growth restriction; IUFD: intrauterine fetal demise; PDA: patent ductus arteriosus; NA: not available.

1. Introduction Assisted reproduction technology (ART) is associated with epigenetic alterations [1–3] that can affect fetal growth in animals and humans and usually results from imprinting. Followup studies of ART-conceived children have shown that ART does not increase the incidence of congenital abnormalities [4–10]; however, it increases the incidence of epigenetic disorder diseases, such as Beckwith-Wiedemann Syndrome (BWS), Angelman Syndrome (AS), and Russell-Silver Syndrome (RSS) [11–17]. In BWS [MIM 130650] and RSS [MIM 180860], abnormal fetal growth is a major phenomenon, and abnormal prenatal development has been associated with the epigenetics of some imprinted genes. Reduced birth weight, which is occasionally observed in infants conceived by ART, is an important consideration as it is associated with adult diseases such as insulin insensitivity, polycystic ovary syndrome, and cardiovascular diseases [18–20]. Therefore, normal prenatal development may be very important not only for childhood health but also for long-term health. Here, we used human placental tissue to compare the imprinted gene expression of IGF2, H19, KCNQ1OT1, and CDKN1C genes known to be associated with fetal growth, in ART-conceived singletons with that in spontaneously conceived (SC) singletons.

e very important not only for childhood health but also for long-term health. Here, we used human placental tissue to compare the imprinted gene expression of IGF2, H19, KCNQ1OT1, and CDKN1C genes known to be associated with fetal growth, in ART-conceived singletons with that in spontaneously conceived (SC) singletons. 2. Materials and Methods A total of 1302 singletons delivered at our center from June 2005 to March 2007 were enrolled in this study. Of these 1302 potential subjects, 313 were excluded due to complications. A total of 860 infants had appropriate-for-date (AFD) birth weight (2500 g ≤ AFD birth weight < 3500 g), 64 cases exhibiting fetal growth retardation (FGR) had a birth weight of <2500 g, and 65 cases had a birth weight of ≥3500 g. Thus, 989 subjects (ART n = 65; SC n = 924) were assessed with 3 idiopathic FGR cases in the ART group and 61 in the SC group (Table 1). For the gene expression study, placental tissue was collected from 297 cases after receiving informed consent under the IRB protocol of our center for genetic analysis (Table 2). Total RNA was extracted from the fetal placenta, and reverse transcription was performed. Gene expressions of IGF2, H19, KCNQ1OT1, and CDKN1C were analyzed by real-time PCR with GAPDH serving as the endogenous control.

after receiving informed consent under the IRB protocol of our center for genetic analysis (Table 2). Total RNA was extracted from the fetal placenta, and reverse transcription was performed. Gene expressions of IGF2, H19, KCNQ1OT1, and CDKN1C were analyzed by real-time PCR with GAPDH serving as the endogenous control. 3. Results and Discussion The mean birth weight was significantly lower (P < .001) in the ART group (2905.1 ± 459.0 g) than in the SC group (3607.9 ± 589.9 g). The mean placental weight, however, showed no significant difference (ART = 689.3 ± 152.6 g; SC = 613.0 ± 142.5 g) (Table 3). Gene expression patterns in the AFD birth weight cases were similar in both the ART and SC groups (Figure 1). H19 expression was reduced in FGR cases both in the ART and SC groups compared with the AFD cases (P < .02 and P < .05, resp.) (Figure 2). Conversely, H19 expression was significantly enhanced in SC cases with a birth weight of ≥3500 g (P < .01) (Figure 3). On the other hand, CDKN1C expression was reduced in ART cases with FGR (P < .01), and KCNQ1OT1 appeared to be hyperexpressed in SC cases with FGR (P <  .05) (Figure 2). The expression of other genes examined showed no difference from the control.

enhanced in SC cases with a birth weight of ≥3500 g (P < .01) (Figure 3). On the other hand, CDKN1C expression was reduced in ART cases with FGR (P < .01), and KCNQ1OT1 appeared to be hyperexpressed in SC cases with FGR (P <  .05) (Figure 2). The expression of other genes examined showed no difference from the control. The results demonstrated that birth weight was significantly lower in the ART group than in the SC group, which is in agreement with the results of other studies [21–23]. Some followup studies of ART-conceived children suggest that low birth weight is due to multiple pregnancies. However, even in singleton cases, low birth weight has been observed in infants conceived by ART. For cases conceived using fresh embryo replacement, birth weight was comparably lower than that for cases conceived using cryopreserved embryos [24, 25]. Although we did not separate cases conceived with fresh embryos and cryopreserved embryos, many cases in this study were conceived by fresh embryo replacement. On the other hand, placental weight showed no significant difference between the ART and SC groups. In other studies, however, placental thickness was significantly larger in ART cases than in SC cases, but there were no differences in morphological or histopathological features of the placenta between both groups [26]. There were no differences in the gene expression patterns in the AFD cases between the ART and SC groups. However, the expression of H19, a paternally methylated imprinted gene, was reduced in FGR cases in both the ART and SC groups. As maternally expressed genes such as H19 enhance fetal development, the hypoexpression of H19 affects fetal development. Here, we established the relationship between the hypoexpression of H19 and reduced fetal weight. Additionally, CDKN1C, another maternally expressed gene, exhibited reduced expression in FGR cases conceived by ART. In contrast, the expression of KCNQ1OT1, a paternally expressed gene with a complementary relationship to CDKN1C, was enhanced in FGR cases conceived by natural conception. In this study, we confirmed differences in the expression of imprinted genes in the placental tissue of infants conceived by ART. However, even in the SC cases, epigenetic alteration has been observed. The loss of imprinting on genes located on chromosome 11 is identified as a cause of poor fetal growth in humans [27], which is also reflected in our study.

s in the expression of imprinted genes in the placental tissue of infants conceived by ART. However, even in the SC cases, epigenetic alteration has been observed. The loss of imprinting on genes located on chromosome 11 is identified as a cause of poor fetal growth in humans [27], which is also reflected in our study. We postulate that ART could affect the epigenetic characteristics of male and female gametes or it can have an impact on early embryogenesis. Additionally, ART could be associated with an increased risk of genomic imprinting abnormalities as epigenetic reprogramming occurs during gametogenesis or immediately following fertilization [28–32]. 4. Conclusions Imprinted gene expression patterns of placental tissue in FGR cases were altered compared with cases of normal fetal growth. However, imprinted gene expression patterns of placental tissue in ART cases were different from those of SC cases. In cases with a birth weight of ≥3500 g, gene expression differed from cases with standard fetal growth. While we recognize the possibility of changes in epigenetic status in any pregnancy, we speculate that epigenetic status is altered by ART. Although ART has been widely accepted and safety performed, epigenetics should remain an important factor for evaluating the safe development of reproductive medicine, as well as for considering the health of the next generation.

epigenetic status in any pregnancy, we speculate that epigenetic status is altered by ART. Although ART has been widely accepted and safety performed, epigenetics should remain an important factor for evaluating the safe development of reproductive medicine, as well as for considering the health of the next generation. Figure 1 Gene expression of placental tissue. ART versus SC in AFD birth weight cases. ART: assisted reproductive technology. SC: spontaneous conception. AFD: appropriate-for-date. Results of gene expression analysis compared with the endogenous control GAPDH. In AFD birth weight cases, gene expression patterns were similar in both the ART and SC groups. Figure 2 Gene expression of placental tissue. ART versus SC in FGR cases. There were no differences in the gene expression of IGF2; however, H19 expression was significantly reduced in FGR cases both in the ART and SC groups compared with the AFD birth weight cases (P < .02 and P < .05, resp.). Conversely, KCNQ1OT1 was hyperexpressed in FGR cases in the SC group (P < .05), while CDKN1C expression was reduced in FGR cases in the ART group (P < .01). Figure 3 Gene expression in placental tissue. FGR and birth weight ≥3500 g cases in the SC group. H19 expression was significantly reduced in FGR cases, but significantly enhanced in cases with a birth weight of ≥3500 g (P < .01). Table 1 Subject characteristics. ART (n) SC (n) Total (n) AFD (2500 g ≤, <3500 g) 62 798 860 FGR (≤ 2500 g) 3 61 64 OG (≥3500 g) — 65 65 Total 65 924 989

Figure 3 Gene expression in placental tissue. FGR and birth weight ≥3500 g cases in the SC group. H19 expression was significantly reduced in FGR cases, but significantly enhanced in cases with a birth weight of ≥3500 g (P < .01). Table 1 Subject characteristics. ART (n) SC (n) Total (n) AFD (2500 g ≤, <3500 g) 62 798 860 FGR (≤ 2500 g) 3 61 64 OG (≥3500 g) — 65 65 Total 65 924 989 n: number of cases, AFD: appropriate-for-date, FGR: fetal growth retardation, OG: over growth, ART: assisted reproductive technology, and SC: spontaneous conception. Table 2 Imprinted gene expression analysis in placental tissue samples. ART (n) SC (n) Total (n) AFD (≥2500 g, <3500 g) 45 173 218 FGR (≤2500 g) 3 51 54 OG (≥3500 g) — 25 25 Total 48 249 297 n: number of cases, AFD: appropriate-for-date, FGR: fetal growth retardation, OG: over growth, ART: assisted reproductive technology, and SC: spontaneous conception. Table 3 Birth weight and placenta weight. Weight (g) n Neonate Placenta ART 65 2905.1 ± 459.0* 589.3 ± 152.6 SC 924 3607.9 ± 589.9* 613.0 ± 142.5 *P < .001. n: number of cases, ART: assisted reproductive technology, and SC: spontaneous conception.

1. Introduction Adolescence constitutes a period of rapid growth, both physical and emotional. It encompasses young people aged 10 to 19 years. Adolescent pregnancy is considered as high risk and is associated with high rates of obstetric and psychological complications [1–5]. One out of four pregnancy-related deaths occurs in adolescents [6]. One of the objectives of safe motherhood programs, of which family planning is an essential component, is to reduce the contribution of this age group to childbirth [7]. Contraceptive prevalence in Cameroon is low among married women and shows uneven distribution varying from 2.6% in North Cameroon to 43.9% in Yaoundé [8]. This low-contraceptive use prevalence among married couples is reported at 2.6%, 3.3%, and 17.6% in North, Far North, and Adamawa regions. The low prevalence use among married couples suggests that once a woman is married she is exposed to pregnancy irrespective of her age. In Cameroon, the illiteracy rate varies from 44.3% in Far North to 3.4% in Yaoundé, with a rate of 33%, 39%, and 44%, respectively, in Adamaoua, Far North and North. Illiteracy predisposes to early marriage and early childbearing among adolescents [9].

1. Introduction Adolescence constitutes a period of rapid growth, both physical and emotional. It encompasses young people aged 10 to 19 years. Adolescent pregnancy is considered as high risk and is associated with high rates of obstetric and psychological complications [1–5]. One out of four pregnancy-related deaths occurs in adolescents [6]. One of the objectives of safe motherhood programs, of which family planning is an essential component, is to reduce the contribution of this age group to childbirth [7]. Contraceptive prevalence in Cameroon is low among married women and shows uneven distribution varying from 2.6% in North Cameroon to 43.9% in Yaoundé [8]. This low-contraceptive use prevalence among married couples is reported at 2.6%, 3.3%, and 17.6% in North, Far North, and Adamawa regions. The low prevalence use among married couples suggests that once a woman is married she is exposed to pregnancy irrespective of her age. In Cameroon, the illiteracy rate varies from 44.3% in Far North to 3.4% in Yaoundé, with a rate of 33%, 39%, and 44%, respectively, in Adamaoua, Far North and North. Illiteracy predisposes to early marriage and early childbearing among adolescents [9]. Existing data on Cameroon indicate that adolescents represent 21% of the total population and contribute for 13.8% of deliveries [9, 10]. Two recent studies revealed that adolescents' contribution to deliveries is 6.69% at the University Teaching Hospital in Yaoundé (capital city of Cameroon) and 26.54% in Maroua Regional Hospital in the Far North Region [11, 12]. These findings suggest that there is a great disparity in the geographic distribution of adolescent deliveries in Cameroon. Knowledge of this geographic distribution can provide useful information for updating and strengthening adolescent reproductive health strategies in Cameroon.

tal in the Far North Region [11, 12]. These findings suggest that there is a great disparity in the geographic distribution of adolescent deliveries in Cameroon. Knowledge of this geographic distribution can provide useful information for updating and strengthening adolescent reproductive health strategies in Cameroon. Objective To determine the frequency and the trend of adolescents (10–19 years) in childbirth within a period of 3 years in referral maternity units in Cameroon. 2. Methods 2.1. Study Design and Setting It is a retrospective study carried out over a period of 3 years to assess adolescents' contribution to deliveries in referral maternity units in Cameroon. We included referral maternity units supervised by Obstetrician-Gynecologist in the 10 regions of Cameroon. We identified referral maternity units by region: Centre (General Hospital, Gynaeco-Obstetric and Pediatric Hospital, University Hospital Centre, Yaoundé Central Hospital), Littoral (General Hospital, Laquintinie Hospital), South (Ebolowa Provincial Hospital, Nsangmelima District Hospital), West (Bafoussam Provincial Hospital), North West (Bamenda Provincial Hospital), South West (Limbe Provincial Hospital, Assimilated Provincial Hospital of Buea), East (Bertoua Provincial Hospital), Adamaoua (Ngaoundere Provincial Hospital), North (Garoua Provincial Hospital), and Far North (Maroua Provincial Hospital). We excluded Buea and Bertoua Provincial Hospitals for insufficient data during the period of study. We therefore analyzed data of 14 referral hospitals from ten regions of the Country (Figure 1).

, Adamaoua (Ngaoundere Provincial Hospital), North (Garoua Provincial Hospital), and Far North (Maroua Provincial Hospital). We excluded Buea and Bertoua Provincial Hospitals for insufficient data during the period of study. We therefore analyzed data of 14 referral hospitals from ten regions of the Country (Figure 1). 2.2. Population The study population consisted of girls and women who were received in the delivery room of the maternity units at the referral hospitals in Cameroon during a three-year period (2003 to 2005) irrespective of where they had antenatal care. 2.3. Variables We collected data retrospectively from delivery registers. We contacted at least one Obstetrician-Gynecologist in each selected maternity unit, either by telephone or by email. We explained the objectives of the study and gave them the data collection instrument. The variables specified in the instruments were name of the hospital, year of delivery (2003, 2004, and 2005), number of deliveries in the year separated by classes (the overall, less than 17 years, 17–19 years, with age nonspecified). We did not include women whose pregnancy was terminated before 28 weeks (7 months) and commonly considered as abortions. The sociodemographic variables, management of labor, and pregnancy outcome were not assessed because the objective of this study was mainly to evaluate the proportion of deliveries by adolescents.

). We did not include women whose pregnancy was terminated before 28 weeks (7 months) and commonly considered as abortions. The sociodemographic variables, management of labor, and pregnancy outcome were not assessed because the objective of this study was mainly to evaluate the proportion of deliveries by adolescents. 2.4. Statistical Analysis We calculated the crude rate and annual contribution of adolescents to deliveries. The maternity unit of Laquintinie Hospital reported cumulative data over a period of three years and the national annual rates reported in this study did not include this health facility. Data were analyzed with Epi Info 3.5. The Chi2 square test for trend was used to evaluate the contribution of adolescents to deliveries over time. The trend was considered statistically significant if P < .05. We grouped the towns into zones ranging from I to V according to the contribution of adolescents to deliveries. The zones were defined as follows: zone I (<10%), zone II (10–14%), zone III (15–19%), zone IV (20–25%), and zone V (≥25%).

deliveries over time. The trend was considered statistically significant if P < .05. We grouped the towns into zones ranging from I to V according to the contribution of adolescents to deliveries. The zones were defined as follows: zone I (<10%), zone II (10–14%), zone III (15–19%), zone IV (20–25%), and zone V (≥25%). 3. Results The 14 hospitals included in this study fall into several categories of the health facilities in Cameroon. Category 1 (University Teaching Hospital, Gynecology-Obstetric and Pediatric Hospital, General Hospitals of Yaoundé and Douala), Category 2 (Yaoundé Central Hospital and Laquintinie Hospital in Douala), Category 3 (Provincial Hospitals of Maroua, Garoua, Ngaoundere, Bertoua, Bamenda, Limbe and Bafoussam), and Category 4 (Nsangmelima District Hospital) (Table 1). During the study period, we had a total of 57787 births where maternal age was available. We excluded of the analysis 165 births (1.97%) where maternal age was not specified. Overall, we found that 8222 of the births were by teenage mothers. We found that during the study period, adolescents contributed for 12.83% to 15.08% of deliveries and this percentage decreased significantly over time (P for trend <.001) (Table 2). The contribution of adolescents to deliveries varies from region to region ranging from 6.87% in the Bafoussam (West) to 26.51% in Maroua (Far North) with an overall national frequency of 14.23%. The contribution of adolescents aged 16 or less varied from 0.81% in the West to 6.15% in Far North.

aggressive clinical course and poorer prognosis than Type I carcinomas. Common genetic changes include mutations of TP53 and CDH1 (E-cadherin) genes [4]. Despite the recent advances in molecular diagnostics, the most important factors in predicting patient prognosis remain to be tumor grade, stage, and subtypes [5, 6]. Sox proteins are transcription factors related by a 79-amino acid high-mobility-group (HMG) DNA-binding domain that was first identified in the mammalian Sry protein [7]. They take up various roles in neural development, including neural stem cell maintenance, glial specification, and lineage-specific terminal differentiation [8]. More than 20 members of the SOX gene family have been identified in mammals [9]. Among them, SOX2 was first found crucial for maintaining the stemness of neural stem cells and then of embryonic stem cells. In conjunction with OCT3/4 and NANOG, SOX2 is considered a master regulator of mammalian embryogenesis and part of a complex network of transcription factors that affects both pluripotency and differentiation in embryonic stem cells [10]. In fact, forced expression of OCT3/4, SOX2, c-MYC, and KLF4 was sufficient to induce stem cell-like pluripotency in adult fibroblast [11] and CD34+ blood cell [12].

trend <.001) (Table 2). The contribution of adolescents to deliveries varies from region to region ranging from 6.87% in the Bafoussam (West) to 26.51% in Maroua (Far North) with an overall national frequency of 14.23%. The contribution of adolescents aged 16 or less varied from 0.81% in the West to 6.15% in Far North. We observed that Bafoussam (West) and Yaoundé (Centre) had the lowest number of adolescent deliveries as a percentage of all births, with 6.87% and 8.68%, respectively. At the central level, we found that General Hospitals had lower rates of adolescent deliveries. This is highlighted by the difference of adolescents' contribution to deliveries in Douala General Hospital (1.06%) and Laquintinie Hospital (8.36%), two hospitals in the same city. We found that 4 out of 10 towns included in this study were of zones IV and V with more than 20% of deliveries among adolescents and those four towns were Nsangmelima, Ngaoundere, Garoua, and Maroua. 4. Discussion Several studies have reported deliveries among adolescents as a percentage of all births and this percentage varies from 6% to 26% [11–16]. In our study, we found that adolescents contributed to 14.23% of deliveries in referral hospitals in Cameroon. This percentage is lower than that reported in South Africa and in Maroua-Cameroon, where the authors found that one out of every four mothers was aged less than 20 years [12, 14].

ries from 6% to 26% [11–16]. In our study, we found that adolescents contributed to 14.23% of deliveries in referral hospitals in Cameroon. This percentage is lower than that reported in South Africa and in Maroua-Cameroon, where the authors found that one out of every four mothers was aged less than 20 years [12, 14]. Data from Far North in Cameroon revealed that 97% of adolescents are married at the time of delivery [17]. Early marriage can be considered as an indicator for early childbearing, since very few married teenagers (3.3%) take any modern contraception as shown by the 2004 DHS [8]. Another study on 8174 deliveries (which included women with all parities) in the University Hospital Centre in Yaoundé showed that adolescents contributed to 6.69% of deliveries [11]. In the present study, we found that adolescents contributed to 9.20% of deliveries in the same hospital (University Hospital Centre in Yaoundé) suggesting a rising rate of adolescent childbirth in this setting. One study in the USA revealed that 6.21% of women giving birth were aged 15 to 19 [18]. Creatsas and Elsheikh recently reported that 7.53% of deliveries in Greece occur among adolescents [16], a rate comparable to that of Yaoundé Cameroon [11] and USA [18]. The national rate reported in our study is higher than that reported in the USA.

SA revealed that 6.21% of women giving birth were aged 15 to 19 [18]. Creatsas and Elsheikh recently reported that 7.53% of deliveries in Greece occur among adolescents [16], a rate comparable to that of Yaoundé Cameroon [11] and USA [18]. The national rate reported in our study is higher than that reported in the USA. Our study revealed that adolescents aged 16 or less contributed to 2.82% of deliveries. This rate is comparable to the 1.21% reported previously in the University Hospital in Yaoundé, but lower than the 7.6% recently reported in Maroua-Cameroon [12, 19]. We found geographic differences in the contribution of adolescents to deliveries, with hospitals in zone IV located in the Northern regions and in Nsangmelima. The cultural and social values in northern Cameroon contribute to high fertility rate and the desire for many children. This region is dominated by Muslims and early marriage among adolescents is promoted by cultural values.

ts to deliveries, with hospitals in zone IV located in the Northern regions and in Nsangmelima. The cultural and social values in northern Cameroon contribute to high fertility rate and the desire for many children. This region is dominated by Muslims and early marriage among adolescents is promoted by cultural values. We found that the contribution of adolescents to deliveries is decreasing over time. This observation corroborates with that of the trend in decrease of total fertility rate of the Country from 5.8 in 1991 to 5.0 in 2004 [10]. Moreover, the national modern contraceptive prevalence used among couples increased from 4% in 1991 to 14% in 2004 [8]. This suggests that multisectorial and multidisciplinary efforts to discourage early sexual debut have been effective. However, these national figures can hide isolated circumstances with increasing number of adolescents as a percentage of all births in some localities. In fact, an increasing proportion of childbirth among adolescents has previously been reported in the University Hospital Centre and was confirmed in our study. To the best of our knowledge, this is the third among studies reported in English or French (following two previous studies in University Hospital Centre Yaoundé, and Maroua Provincial Hospital) to have analyzed the contribution of adolescents to deliveries and changes over time [11, 12].

We found that the contribution of adolescents to deliveries is decreasing over time. This observation corroborates with that of the trend in decrease of total fertility rate of the Country from 5.8 in 1991 to 5.0 in 2004 [10]. Moreover, the national modern contraceptive prevalence used among couples increased from 4% in 1991 to 14% in 2004 [8]. This suggests that multisectorial and multidisciplinary efforts to discourage early sexual debut have been effective. However, these national figures can hide isolated circumstances with increasing number of adolescents as a percentage of all births in some localities. In fact, an increasing proportion of childbirth among adolescents has previously been reported in the University Hospital Centre and was confirmed in our study. To the best of our knowledge, this is the third among studies reported in English or French (following two previous studies in University Hospital Centre Yaoundé, and Maroua Provincial Hospital) to have analyzed the contribution of adolescents to deliveries and changes over time [11, 12]. The contribution of adolescents to deliveries is an indicator of early unprotected sexual activity. This does not only have demographic implications, but also affect adolescent reproductive health such as exposure to various sexually transmitted infections, HIV infection, cancer of the cervix, obstetric fistula, and death [20, 21].

ion of adolescents to deliveries is an indicator of early unprotected sexual activity. This does not only have demographic implications, but also affect adolescent reproductive health such as exposure to various sexually transmitted infections, HIV infection, cancer of the cervix, obstetric fistula, and death [20, 21]. In a recent study in the University Hospital Centre in Yaoundé, the authors reported that the first delivery among adolescents was associated with prematurity, perinea tear, fetal distress, episiotomy, use of oxytocin, delivery by cesarean section, and stillbirth [13]. Different studies have shown that adolescents, and in particular those less than 17 years, have higher rates of caesarean delivery and stillbirth [22–24].

very among adolescents was associated with prematurity, perinea tear, fetal distress, episiotomy, use of oxytocin, delivery by cesarean section, and stillbirth [13]. Different studies have shown that adolescents, and in particular those less than 17 years, have higher rates of caesarean delivery and stillbirth [22–24]. Adolescents contribute significantly to deliveries in Cameroon and more especially in the Northern regions. Nowadays women have more responsibilities for themselves and for their children; they want to continue their education and have a job before childbearing. However, all women do not have access to information and sufficient education to achieve their objectives. The priority of health services and all organizations involved in the provision of adolescent reproductive health should be to reduce the contribution of adolescents to deliveries. The different risk factors for adolescent pregnancy dictate which preventive measures to take. Studies have been conducted on knowledge, attitudes, and practice among adolescents, on sexually transmitted infections and unwanted pregnancy, and it seems that the best interventions are when adolescents themselves participate in the development of the strategies [25].

nancy dictate which preventive measures to take. Studies have been conducted on knowledge, attitudes, and practice among adolescents, on sexually transmitted infections and unwanted pregnancy, and it seems that the best interventions are when adolescents themselves participate in the development of the strategies [25]. Various interventions based on the introduction of modules in school curricula have been assessed with the hope to reduce unwanted pregnancy and its consequences on adolescents [26]. These interventions include school programs based on abstinence only, abstinence with information on contraception, encouragement of community activities, and teaching on competences to overcome peer pressure. Considering the fact that the contribution of adolescents to deliveries is higher in northern than southern Cameroon, it is essential to rethink of the strategies implemented so far in Cameroon and to undertake a study that identify specific strategies for each zone. We recommend early integration of education and reproductive health in school curriculum, and education of parents to improve parental guide of adolescents. We acknowledge that this study has some limitations. Data collection was not easy in maternity units due to absence of registers for some years (Buea and Ebolowa) and the presence of registers with no data (Bertoua). In Cameroon, 40% of deliveries occur at home and this study concern only the women who delivered in the health facilities [27].

study has some limitations. Data collection was not easy in maternity units due to absence of registers for some years (Buea and Ebolowa) and the presence of registers with no data (Bertoua). In Cameroon, 40% of deliveries occur at home and this study concern only the women who delivered in the health facilities [27]. Additionally, some sociodemographic conditions like, educational level, marital status, illiteracy, occupation, and emotion are potential risk factors for early pregnancy; unfortunately, these factors were not assessed in the present study. 5. Conclusion Adolescents contribute significantly to deliveries in Cameroon. This study highlights the need to conduct a study to determine the impact of very young age on pregnancy. Our study underscores the importance of political will and Public Health programs in strengthening adolescent health services. In particular, there is a need to readapt education strategies in order to improve health among adolescents. Despite the willingness of the state and international institutions, for any intervention to be effective, competent authorities, experts, and interested parties must come together to develop a multidisiplinary and concerted approach to reduce adolescent pregnancies.

tegies in order to improve health among adolescents. Despite the willingness of the state and international institutions, for any intervention to be effective, competent authorities, experts, and interested parties must come together to develop a multidisiplinary and concerted approach to reduce adolescent pregnancies. Acknowledgments The authors wish to thank the Gynecologist-Obstetricians and all staff of the Obstetrics and Gynecology Department of the University Hospital Centre in Cameroon for their kind collaboration, participation in this study, and proper handling and documentation of files and maternity registers. Dr. Kemfang Jean Dupont (Yaounde General Hospital), Dr. Nana Njamen Théophile (Yaounde General Hospital), Dr. Séraphin Fokoua ( Yaounde Gynaeco-Obstetric and Pediatric Hospital), Pr luc Kouam (University Hospital Centre), Dr. Philippe Nana ( Yaounde Central Hospital), Dr. Halle Ekane Gregory (General Hospital), Dr. Nguento (Laquintinie Hospital), Dr. Roger Dogmo (Ebolowa Provincial Hospital), Dr. Bruno kenfack ( Nsangmelima District Hospital), Dr. Paul Doumbé (Bafoussam Provincial Hospital), Dr. Isaac Sandjong (Bamenda Provincial Hospital), Dr. Tchounzou Robert (Limbe Provincial Hospital), Dr. Dupleix Fogaing (Assimilated Provincial Hospital of Buea), Dr. Adamo Bongoe (Bertoua Provincial Hospital); Pr James Lenga (Ngaoundere Provincial Hospital), and Dr. Thomas Zra (Garoua Provincial Hospital). Figure 1 Regional presentation of Cameroon. Table 1 Characteristics of health facilities studied.

Acknowledgments The authors wish to thank the Gynecologist-Obstetricians and all staff of the Obstetrics and Gynecology Department of the University Hospital Centre in Cameroon for their kind collaboration, participation in this study, and proper handling and documentation of files and maternity registers. Dr. Kemfang Jean Dupont (Yaounde General Hospital), Dr. Nana Njamen Théophile (Yaounde General Hospital), Dr. Séraphin Fokoua ( Yaounde Gynaeco-Obstetric and Pediatric Hospital), Pr luc Kouam (University Hospital Centre), Dr. Philippe Nana ( Yaounde Central Hospital), Dr. Halle Ekane Gregory (General Hospital), Dr. Nguento (Laquintinie Hospital), Dr. Roger Dogmo (Ebolowa Provincial Hospital), Dr. Bruno kenfack ( Nsangmelima District Hospital), Dr. Paul Doumbé (Bafoussam Provincial Hospital), Dr. Isaac Sandjong (Bamenda Provincial Hospital), Dr. Tchounzou Robert (Limbe Provincial Hospital), Dr. Dupleix Fogaing (Assimilated Provincial Hospital of Buea), Dr. Adamo Bongoe (Bertoua Provincial Hospital); Pr James Lenga (Ngaoundere Provincial Hospital), and Dr. Thomas Zra (Garoua Provincial Hospital). Figure 1 Regional presentation of Cameroon. Table 1 Characteristics of health facilities studied. Characteristics N(%) N = 16 Level Central 6 (37.5) Intermediate 9 (56.3) Peripheral 1 (6,3) Category Category 1 4 25.0 Category 2 2 (12.5) Category 3 9 (56.3) Category 4 1 (6.3) Number of health facilities by Province Centre 4 (25.0) Littoral 2 (12.5) West 1 (6.3) North West 1 (6.3) South West 2 (12.5) South 2 (12.5) East 1 (6.3) Adamaoua 1 (6.3) North 1 (6.3) Far North 1 (6.3)

Characteristics N(%) N = 16 Level Central 6 (37.5) Intermediate 9 (56.3) Peripheral 1 (6,3) Category Category 1 4 25.0 Category 2 2 (12.5) Category 3 9 (56.3) Category 4 1 (6.3) Number of health facilities by Province Centre 4 (25.0) Littoral 2 (12.5) West 1 (6.3) North West 1 (6.3) South West 2 (12.5) South 2 (12.5) East 1 (6.3) Adamaoua 1 (6.3) North 1 (6.3) Far North 1 (6.3) Number of deliveries in 3 years 10,000 and more 1 (6.3) 9000–9999 1 (6.3) 6000–6999 3 (18.8) 5000–5999 2 (12.5) 4000–4999 1 (6.3) 3000–3999 1 (6.3) 2000–2999 4 (25.0) Less than 2000 3 (18.8) N: Number of women who gave birth; %: Percentage. Table 2 Distribution of deliveries by year during the period of the study. Year Deliveries per year 10–16 year 17–19 year Total N (%) N (%) N (%) 2003 18554 544 (2.93) 2254 (12.15) 2798 (15.08) 2004 19687 612 (3.11) 2239 (11.37) 2851 (14.48) 2005 19546 473 (2.42) 2100 (10.74) 2508 (12.83) 2003–2005 57787 1629 (2.82) 6593 (11.41) 8222 (14.23) Chi2 for trend: 30.2 and P = .0000; N  : Number of women who gave birth; %: Percentage.s

1. Introduction DNA methylation of cytosine located 5′ to a guanosine is one of the most important modifications of genomic DNA in eukaryotic cells. Methylation of cytosine at CpG dinucleotides is described as an epigenetic regulation mechanism of genomic function that plays an important role in different biological processes including embryogenesis [1], genomic imprinting [2], X-chromosome inactivation, and cancer [3, 4]. Aberrant methylation pattern has been found to be a common event in many cancers [5–7]. Global hypomethylation is considered to play a role in carcinogenesis; however, local hypermethylation changes gene expression [8]. This hypermethylation alteration resulted in transcriptional inactivation followed by silencing of promoter at nearby tumor suppressor genes, contributing to development of cancer. The hypermethylation was thought to be an early event in carcinogenesis [9–12]. A large number of studies in cancers including breast cancer have focused on the use of CpG island hypermethylation profiling as cancer biomarkers in tissue and circulating cell-free DNA of patients, with the aim of improving cancer treatment via accurate early diagnosis, noninvasive diagnosis, prognosis, and prognosis therapy selection [7, 13–18].

ancers including breast cancer have focused on the use of CpG island hypermethylation profiling as cancer biomarkers in tissue and circulating cell-free DNA of patients, with the aim of improving cancer treatment via accurate early diagnosis, noninvasive diagnosis, prognosis, and prognosis therapy selection [7, 13–18]. Recent technology development has provided the analysis of DNA methylation in a genome-wide scale [19, 20] which may not be easily accessible for many institutions. Thereby, in most of the research centers methylation assays can be only determined on gene-by-gene-based methods that use bisulfite conversion. The bisulfite reaction was first described in early 1970s [21, 22]. Since the first description of bisulfite reaction in the application of studying CpG sites, many methods based on the same principle have been developed and categorized according to primer designing strategies. Based on primer designing strategies two different DNA methylation assays are described, methylation-independent-specific PCR (MIP) primers and methylation-specific PCR (MSP) primers [23].

g CpG sites, many methods based on the same principle have been developed and categorized according to primer designing strategies. Based on primer designing strategies two different DNA methylation assays are described, methylation-independent-specific PCR (MIP) primers and methylation-specific PCR (MSP) primers [23]. Primer and probe design for methylation assays based on bisulfite conversion is challenging because of the DNA composition after bisulfite modification. One of the most critical steps for methylation study is designing primers and probes for the modified DNA and it needs special constrains on primers or probe and their location on the DNA. A large number of studies have been focused on identification of biomarkers; however, the clinical use of these biomarkers is still very limited because of lack of specificity and sensitivity for diagnostic test. This highlights the critical need for specific primer and probe design to avoid false-positive detection of methylation. We review a brief guideline of CpG island prediction, designing primers and probes for MIP and MSP assays that are used for methylation studies based on bisulfite conversion. Some important web-tools for methylation studies are introduced as well.

Primer and probe design for methylation assays based on bisulfite conversion is challenging because of the DNA composition after bisulfite modification. One of the most critical steps for methylation study is designing primers and probes for the modified DNA and it needs special constrains on primers or probe and their location on the DNA. A large number of studies have been focused on identification of biomarkers; however, the clinical use of these biomarkers is still very limited because of lack of specificity and sensitivity for diagnostic test. This highlights the critical need for specific primer and probe design to avoid false-positive detection of methylation. We review a brief guideline of CpG island prediction, designing primers and probes for MIP and MSP assays that are used for methylation studies based on bisulfite conversion. Some important web-tools for methylation studies are introduced as well. 2. CpG Island Prediction Methylation at the cytosine bases of CpG dinucleotide-rich region mostly within 0.5–4 kb are known as CpG islands [24, 25]. Although analysis of the methylation status of some critical CpG sites as biomarkers are better than others, it is essential to find CpG islands at the promoter region of candidate genes which are in close proximity to the transcription start site. In order to predict CpG islands as target region, the following rules should be applied. If CpG island prediction is used for primer design and more than one island is found, any of the predicted islands can be a target region for primer selection.

2. CpG Island Prediction Methylation at the cytosine bases of CpG dinucleotide-rich region mostly within 0.5–4 kb are known as CpG islands [24, 25]. Although analysis of the methylation status of some critical CpG sites as biomarkers are better than others, it is essential to find CpG islands at the promoter region of candidate genes which are in close proximity to the transcription start site. In order to predict CpG islands as target region, the following rules should be applied. If CpG island prediction is used for primer design and more than one island is found, any of the predicted islands can be a target region for primer selection. If a CpG island size is smaller than the minimum product size, the primer pair should span the whole island. If a CpG island size is greater than the maximum product size, the primer pair should be within the island. If a CpG island size is between the minimum and maximum product size, at least two thirds of the island region should be amplified.

If a CpG island size is smaller than the minimum product size, the primer pair should span the whole island. If a CpG island size is greater than the maximum product size, the primer pair should be within the island. If a CpG island size is between the minimum and maximum product size, at least two thirds of the island region should be amplified. 3. Methylation-Independent-Specific PCR (MIP) Primers MIP primers are used in different PCR-based methylation analysis methods including bisulfite-sequencing PCR (BSP) (in 1992, [26]), pyrosequencing [27, 28], combined bisulfite restriction analysis (COBRA) [29], methylation-sensitive single-nucleotide primer extension (MS-SnuPE) [30–32], methylation-sensitive melting curve analysis (MS-MCA) [33], methylation-sensitive high-resolution melting (MS-HRM) [34], matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry with base-specific cleavage and primer extension [35, 36], heavy methyl [37], and microarray DNA methylation profiling technique based on bisulfite conversion, that is, methylation-specific oligonucleotide microarray (MSO) [38].

], matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry with base-specific cleavage and primer extension [35, 36], heavy methyl [37], and microarray DNA methylation profiling technique based on bisulfite conversion, that is, methylation-specific oligonucleotide microarray (MSO) [38]. Incomplete bisulfite modification of DNA is sometimes a concern [39] and results in high representation of methylation levels in studied samples. Successful application of MIP methods depends on whether PCR primer could be designed to amplify the complete modified fragment of interest. To reduce bias of bisulfite-modified DNA against unmodified or incompletely modified DNA or even unsuccessful experimental PCR optimization, primer pair should be picked from a region that have adequate number of cytosines “C”s (no-CpG) in the original sequence [40]. Primer pairs with more “C”s will be preferred by receiving higher weighing scores and increasing the annealing temperature (Table 1). Besides general consideration for designing primer pair, the following constraints are enforced for MIP primer design. Primers should not contain any CpG sites within their sequence to avoid discrimination against methylated or unmethylated DNA (Figure 1). Primers should have an adequate number of “C”s (no-CpG) in their sequence to amplify only bisulfite modified DNA. Primers with more “C”s will be preferred (at least 30%) [40] (Figure 1). A good primer pair should span a maximal number of CpG sites in the selected amplicon to map as many CpG sites as possible.

Primers should not contain any CpG sites within their sequence to avoid discrimination against methylated or unmethylated DNA (Figure 1). Primers should have an adequate number of “C”s (no-CpG) in their sequence to amplify only bisulfite modified DNA. Primers with more “C”s will be preferred (at least 30%) [40] (Figure 1). A good primer pair should span a maximal number of CpG sites in the selected amplicon to map as many CpG sites as possible. If CpG island prediction is not used for primer selection, selected amplicons must span at least 5 CpG sites as a default. Long length primer (25–30 mer) is preferred to ensure uniqueness of the primer [39]. Primer sets should not amplify more than 500 bp because DNA degradation occurs by bisulfite modification.

A good primer pair should span a maximal number of CpG sites in the selected amplicon to map as many CpG sites as possible. If CpG island prediction is not used for primer selection, selected amplicons must span at least 5 CpG sites as a default. Long length primer (25–30 mer) is preferred to ensure uniqueness of the primer [39]. Primer sets should not amplify more than 500 bp because DNA degradation occurs by bisulfite modification. 4. Methylation-Specific PCR (MSP) Primers Methods based on MSP primers are considered to have the highest analytical sensitivity and are designed to specifically amplify either methylated or unmethylated DNA by using primers that distinguish the methylated sequence from the unmethylated sequence [23, 40]. The precision and sensitivity of MSP depends on appropriate primer or probe design not prone to false-positive results [23]. MSP primers-based methods include methylation-specific PCR (MSP) [40], methylight [41, 42], SYBER green-based quantitative MSP [43, 44], sensitive melting analysis after real-time MSP (SMART-MSP) [45], and methylation-specific fluorescent amplicon generation (MS-FLAG) [46]. The specificity of methylation-based PCR methods is achieved by appropriate primer pair or probes design (Table 1). The following constraints are recommended to reduce false-priming events for amplification of methylated DNA.

MSP (SMART-MSP) [45], and methylation-specific fluorescent amplicon generation (MS-FLAG) [46]. The specificity of methylation-based PCR methods is achieved by appropriate primer pair or probes design (Table 1). The following constraints are recommended to reduce false-priming events for amplification of methylated DNA. To discriminate between a methylated and unmethylated DNA fragment, primers have to contain as much CpG sites as possible (at least one CpG) preferably at the very 3′-end. At least one of the last three bases at 3′-end of the primer has to be a CpG “C” (Figure 1). A part from CpG site(s) at the very 3′-end, additional CpG sites in a primer sequence is preferred (Figure 1). Primers for methylated DNA and unmethylated DNA should contain the same CpG sites in their sequence. For example, a forward primer for methylated pair has this sequence: ATAAGTATTCGTTAATGGTTCGA, the forward primer in the unmethylated pair must also contain the two CpG sites, for example, ATAAGTATTTGTTAATGGTTTGA. But they may differ in length and start position [3]. The two sets of primers for methylated and unmethylated DNA should have similar Tm values (max Tm difference 5°C). Elimination of secondary structure formation and primer-dimer pairs by increasing primer length. Primer sets should not amplify more than 500 bp because DNA degradation occurs by bisulfite modification. False-priming event can be prevented by designing appropriate primers and increasing annealing temperature. Having an appropriate negative control in the experiment might help to find out false-priming events.

Elimination of secondary structure formation and primer-dimer pairs by increasing primer length. Primer sets should not amplify more than 500 bp because DNA degradation occurs by bisulfite modification. False-priming event can be prevented by designing appropriate primers and increasing annealing temperature. Having an appropriate negative control in the experiment might help to find out false-priming events. 5. Guidelines for Probe Designing In methylation studies, the discrimination between methylated and unmethylated DNA is achieved by three ways: design of primers that contain or does not contain CpG sites, design of fluorescent labeled probe (for instance MSO and bead array), and design of the both primer and probe, that is, methylight technology [41]. MIP and MSP methods are associated with false positive results. By using fluorescent probes, for instance methylight methodology or applying heavy methyl probe-based methodology, the false positives can be limited. Using probe as a detection method increases the specificity to discriminate between methylated and unmethylated DNA by designing probes that contain additional CpG sites [40]. The selection of new primer pairs for methylation-specific PCR and suitable hybridization probes for real-time PCR-based assays require the identification of the CpG sites that are methylated (Table 1). Moreover, using probe provides possibility to detect more than one target with multiplex reaction by different reporter dyes [38, 47].

new primer pairs for methylation-specific PCR and suitable hybridization probes for real-time PCR-based assays require the identification of the CpG sites that are methylated (Table 1). Moreover, using probe provides possibility to detect more than one target with multiplex reaction by different reporter dyes [38, 47]. In addition probe-based assays can provide quantitative information; further advantages are the speed and high throughput of the 96-well-based, real-time PCR system and the omission of all postamplification steps, which has less labour and the risk of contamination. Also, the efficiency of individual reactions is accessible from the slope of the amplification plot in the logarithmic phase. This allows for the direct quality control of every amplification reaction and the identification of samples containing impurities or poor template that interfered with optimal amplification and thereby with the quantification [48]. A general guideline for probe designing is described as follows: The probe sequences should include 3 to 5 potential methylation sites to maximize specificity and reduce false-priming event. The probe binding sites should include several cytosines in the original sequence to ensure specificity for converted DNA and overcome false positives due to incomplete bisulfite conversion. Long repetitive stretches should be avoided. Probe Tm value should be 10°C higher than primers. G + C content should be 30%–80%. No G should be at the 5′ end. Probes should have 15–30 mer in length. No more than two G + C should be at the 3′ end.