Browse the corpus

1. Introduction Staphylococcus aureus (S. aureus) is an etiologic agent for a wide range of illnesses from localized skin infections such as impetigo, furuncles, and cellulitis to life-threatening diseases such as pneumonia, meningitis, osteomyelitis, necrotizing fasciitis, endocarditis, toxic shock syndrome, and bacteremia [1]. Hospital-associated methicillin-resistant S. aureus (HA-MRSA) first appeared in 1961 shortly after the introduction of methicillin under the selective pressure of antibiotics [2, 3]. Since then, numerous MRSA clones have emerged and become a major cause of nosocomial infections throughout the world [4–6]. MRSA produces penicillin-binding protein 2′ (PBP2′) with a low affinity for β-lactam antibiotics. PBP2′ is encoded by the mecA gene, which is located on a large mobile genetic DNA element, the staphylococcal cassette chromosome mec (SCCmec) [7, 8]. Early MRSA clones were HA-MRSA isolated from patients with compromised immune systems under medical procedures. However, in 1993, community-associated MRSA (CA-MRSA) emerged in patients without risk factors in Western Australia [9]. CA-MRSA clones, which are distinct from HA-MRSA clones, have evolved differently in separate geographical areas and have spread in communities [10]. The incidence of infections due to CA-MRSA, including infections in children with no identifiable risk factors, has increased worldwide as a community pathogen [11, 12]. There are several types of CA-MRSA clones including ST1 (USA400) in Asia, Europe, and the US, ST8 (USA 300) in Europe and the US, ST30 in Australia, Europe, and South America, ST59 in Asia and the US, and ST80 (European clone) in Asia, Europe, and the Middle East. However, recent studies have shown that CA-MRSA strains are spreading into hospitals and are replacing traditional HA-MRSA strains [13, 14]. Therefore, distinction between the traditional HA-MRSA and CA-MRSA strains based on epidemiologic definitions has become difficult, and distinction based on the molecular typing is unclear [4, 15].

udies have shown that CA-MRSA strains are spreading into hospitals and are replacing traditional HA-MRSA strains [13, 14]. Therefore, distinction between the traditional HA-MRSA and CA-MRSA strains based on epidemiologic definitions has become difficult, and distinction based on the molecular typing is unclear [4, 15]. S. aureus is both a commensal organism and a pathogen. S. aureus colonized in the anterior nares, skin, and gastrointestinal tract plays an important role in transmission among individuals and in development of infection [16]. Hisata et al. and Ozaki et al. reported that the prevalences of nasal carriage of S. aureus in healthy Japanese children were 28.2% and 40.4%, respectively, and that the ratios of MRSA strains among the S. aureus isolates were 15.1% and 9.1%, respectively [17, 18]. Impetigo is a highly contagious infection of the superficial epidermis and is commonly caused by exfoliative toxins (ETs) of S. aureus [19]. The ratio of MRSA among S. aureus isolates from patients with impetigo between 1994 and 2000 was below 20% in Japan [20]. However, the ratio of MRSA has recently increased to 20 ~ 50% in Japan [21–24].

ontagious infection of the superficial epidermis and is commonly caused by exfoliative toxins (ETs) of S. aureus [19]. The ratio of MRSA among S. aureus isolates from patients with impetigo between 1994 and 2000 was below 20% in Japan [20]. However, the ratio of MRSA has recently increased to 20 ~ 50% in Japan [21–24]. The characteristics of CA-MRSA strains in Japan are different from those reported in other countries. Although highly virulent CA-MRSA strains carrying Panton-Valentine leukocidin (PVL) genes have been spreading widely in the world, PVL-negative CA-MRSA has been disseminated in Japan [15, 17, 18, 23, 25]. The aim of this study was to investigate the molecular epidemiology by multilocus sequence typing (MLST) and staphylococcal cassette chromosome mec (SCCmec) typing and the antimicrobial susceptibilities of S. aureus isolates from children with impetigo. 2. Materials and Methods 2.1. Bacterial Strains Samples were isolated from skin swabs taken from pediatric outpatients with impetigo. A total of 136 S. aureus isolates were collected from 136 different patients (69 males and 67 females; mean age: 3.8 years; SD: 2.3 years) with impetigo at private offices of 25 practicing pediatricians and 10 hospitals between June 2009 and June 2010 in Hokkaido, the northernmost island of Japan. All samples were collected after obtaining informed consent from the children's parents. MRSA strains were identified by polymerase chain reaction (PCR) for the mecA gene and by the detection of penicillin-binding protein 2′ (Denka Seiken, Co., Tokyo, Japan).

and June 2010 in Hokkaido, the northernmost island of Japan. All samples were collected after obtaining informed consent from the children's parents. MRSA strains were identified by polymerase chain reaction (PCR) for the mecA gene and by the detection of penicillin-binding protein 2′ (Denka Seiken, Co., Tokyo, Japan). 2.2. Molecular Typing and Virulence Gene Analysis PCR primers used for molecular typing and virulence gene analysis in this study are listed in Table 1. The presence of genes encoding PVL (lukS-PV-lukF-PV), toxic shock syndrome toxin 1 (TSST-1) (tst), and 2 ETs (ETA, ETB) (eta, etb) was examined by the PCR method using previously reported primers [26]. MRSA isolates were characterized by MLST as specific sequence types (STs) and clonal complexes (CCs) based on the DNA sequences at 7 defined loci. MLST was performed as described by Enright et al. [27]. MLST is based on sequence analysis of PCR products from seven S. aureus housekeeping genes, that is, arcC, aroE, glpF, gmk, pta, tpi, and yqiL. Each sequence was submitted to the MLST database website (http://www.mlst.net/) for the assignment of the allelic profile and ST. Different sequences are assigned distinct alleles of each housekeeping gene, and each isolate is defined by alleles of the seven genes. The ST data were further analyzed using eBURST (http://eburst.mlst.net/) to determine the CC. SCCmec typing was performed by the multiplex PCR method described by Kondo et al. SCCmec types (I to VI) were determined on the basis of the class (classes A, B, and C) of mec gene complex and the type (types 1, 2, 3, 4, and 5) of ccr gene complex [26]. Subtyping of SCCmec types II and IV was performed on the basis of different J1 regions by PCR using published primers [28].

escribed by Kondo et al. SCCmec types (I to VI) were determined on the basis of the class (classes A, B, and C) of mec gene complex and the type (types 1, 2, 3, 4, and 5) of ccr gene complex [26]. Subtyping of SCCmec types II and IV was performed on the basis of different J1 regions by PCR using published primers [28]. 2.3. Antimicrobial Susceptibility Testing For in vitro antimicrobial susceptibility testing, minimal inhibitory concentrations (MICs) of a panel of 12 antimicrobial drugs, that is, oxacillin (MPIPC), ampicillin (ABPC), cefaclor (CCL), cefditoren (CDTR), clarithromycin (CAM), clindamycin (CLDM), fosfomycin (FOM), minocycline (MINO), gentamicin (GM), vancomycin (VCM), nadifloxacin (NDFX), and fucidic acid (FA), were determined using a frozen plate (Eiken Chemical Co., Tokyo, Japan). This plate is based on the broth microdilution method recommended by the Japan Society for Chemotherapy (JSC) [29, 30]. The breakpoints for these antimicrobial agents were determined according to the interpretation criteria of Clinical and Laboratory Standards Institute (CLSI) [31, 32] and JSC [29, 30]. Susceptibilities to antimicrobial drugs for which breakpoints are not determined by CLSI and JSC were defined in this study. A nitrocefin method was used for β-lactamase detection [33].

e determined according to the interpretation criteria of Clinical and Laboratory Standards Institute (CLSI) [31, 32] and JSC [29, 30]. Susceptibilities to antimicrobial drugs for which breakpoints are not determined by CLSI and JSC were defined in this study. A nitrocefin method was used for β-lactamase detection [33]. 3. Results In the bacteriological analysis of the 155 patients with impetigo, S. aureus and Streptococcus pyogenes were isolated from 136 and 6, respectively. Two cases involved two pathogens. No pathogenic organism was isolated from 15 patients. Of the 136 S. aureus cases, 122 (89.7%) were MSSA and 14 (10.3%) were MRSA. The 14 MRSA patients except for two newborns did not have a history of hospitalization during the previous year. Age distribution in the 136 patients was shown in Table 2. The molecular characteristics of the 14 MRSA strains isolated from the patients with impetigo were examined and are summarized in Table 3. Among the 14 strains that carried the mec gene complex, 11 were determined the SCCmec element type. By SCCmec typing, the 14 MRSA strains were defined as SCCmec type II (IIb: 3 strains; IINT: 4 strains), type IVa (3 strains), type V (1 strain), and unknown type (3 strains). Types of ccr gene complex in 3 strains could not be determined.

ns that carried the mec gene complex, 11 were determined the SCCmec element type. By SCCmec typing, the 14 MRSA strains were defined as SCCmec type II (IIb: 3 strains; IINT: 4 strains), type IVa (3 strains), type V (1 strain), and unknown type (3 strains). Types of ccr gene complex in 3 strains could not be determined. The 14 MRSA strains were comprised of 4 CCs. The four CCs were CC89 (11/14), CC5 (1/14), CC8 (1/14), and CC121 (1/14). ST89, ST91, and ST2117 strains belonged to the same CC89. No PVL-positive strains were identified in the 14 MRSA strains. The MRSA strains belonging to ST5 and ST8 carried TSST-1. Of the 14 MRSA strains, 11 (78.6%) were etb gene positive, and one was eta gene positive. There was no strain carrying both eta and etb genes. The MRSA strains belonging to ST5 and ST8 were both eta gene- and etb gene-negative. Antimicrobial susceptibilities of the MRSA and MSSA strains to 12 antimicrobial agents are summarized in Table 4. The MICs of 12 antimicrobial agents for 14 individual MRSA strains are shown in Table 5. β-Lactamase was produced by 99 (72.8%) of the 136 S. aureus isolates. Of the 122 MSSA isolates, 85 (69.7%) were β-lactamase producers. All MRSA isolates were β-lactamase-positive. MRSA strains were highly resistant to CCL, CDTR, and CLDM. GM- and CAM-resistant strains were frequently found in both MRSA and MSSA. MINO, NDFX, FA and VCM had significant antimicrobial activity against the S. aureus isolates. The ST5 MRSA strain exhibited resistance to multiple drugs, including MINO and NDFX.

sitive. MRSA strains were highly resistant to CCL, CDTR, and CLDM. GM- and CAM-resistant strains were frequently found in both MRSA and MSSA. MINO, NDFX, FA and VCM had significant antimicrobial activity against the S. aureus isolates. The ST5 MRSA strain exhibited resistance to multiple drugs, including MINO and NDFX. 4. Discussion The rate of CA-MRSA in patients with impetigo has been increasing in Japan. In this study, only 10.3% of the S. aureus isolates from patients with impetigo in Hokkaido, the northernmost island of Japan, were MRSA. The ratio of MRSA strains in Hokkaido is lower than those in other regions in recent studies [21–24]. This ratio of CA-MRSA in patients with impetigo in Hokkaido is similar to that in healthy children previously reported in Japan [17, 18]. However, further continuous surveillance is needed to clarify whether the low ratio of CA-MRSA in Hokkaido is true or due to biases.

those in other regions in recent studies [21–24]. This ratio of CA-MRSA in patients with impetigo in Hokkaido is similar to that in healthy children previously reported in Japan [17, 18]. However, further continuous surveillance is needed to clarify whether the low ratio of CA-MRSA in Hokkaido is true or due to biases. HA-MRSA strains usually carry large SCCmec type I, II, or III and are usually multidrug resistant. On the other hand, CA-MRSA strains usually carry smaller SCCmec type IV or V and are usually more susceptible to non-β-lactam antibiotics [11, 34]. In previous studies, SCCmec type IV MRSA strains have been predominantly isolated from patients with impetigo [22, 35, 36]. Noguchi et al. reported that 98.7% of the MRSA strains isolated from patients with impetigo and staphylococcal scalded skin syndrome between 1999 and 2004 in Japan had type IV SCCmec [35]. Takizawa et al. reported that all of the characterized MRSA strains isolated from patients with impetigo in 2003 and 2004 also had type IV SCCmec [22]. Nakaminami et al. examined MRSA strains isolated from patients with impetigo in Japan in 2006 and reported that all isolates with SCCmec were classified into type IV (92.8%) and V (7.2%) [36]. MRSA strains isolated from children with impetigo in China from 2003 to 2007 were classified into type IV (54.5%), type V (18.2%), and VI (9.1%) [37].

ned MRSA strains isolated from patients with impetigo in Japan in 2006 and reported that all isolates with SCCmec were classified into type IV (92.8%) and V (7.2%) [36]. MRSA strains isolated from children with impetigo in China from 2003 to 2007 were classified into type IV (54.5%), type V (18.2%), and VI (9.1%) [37]. However, in very recent studies, MRSA strains isolated from patients with impetigo had diverse SCCmec types, and SCCmec type IIb has been increasing [23, 24]. Type IIb SCCmec was first identified from MRSA strains isolated from healthy Japanese children [17]. Hisata et al. characterized 17 MRSA strains isolated from patients with impetigo at a Japanese hospital from June through August 2002. They found that the 17 MRSA strains carried diverse types of SCCmec: type II (IIb: 4 strains; unknown subtype: 2 strains), type IV (7 strains), and type V (4 strains) [23]. Shi et al. reported that among 26 MRSA strains isolated from patients with impetigo, 57.7% carried type IIa SCCmec element and 42.3% carried type IV SCCmec element [24]. In the present study, the MRSA strains isolated from patients with impetigo also had diverse SCCmec types, and SCCmec type II, including type IIb and IINT, was the predominant type among the CA-MRSA strains.

rom patients with impetigo, 57.7% carried type IIa SCCmec element and 42.3% carried type IV SCCmec element [24]. In the present study, the MRSA strains isolated from patients with impetigo also had diverse SCCmec types, and SCCmec type II, including type IIb and IINT, was the predominant type among the CA-MRSA strains. The role of PVL in the pathogenesis of staphylococcal infections is controversial [38]. However, PVL is responsible at least in part for the increased virulence of CA-MRSA. PVL-positive S. aureus strains are more frequently associated with cellulitis and abscesses than with impetigo [39–41]. In this study, none of the MRSA isolates carried the PVL gene. Although highly virulent CA-MRSA strains carrying PVL genes are known to prevail in the world, the prevalence of PVL-positive strains in Japan is not high [15, 17, 18, 23, 25]. There are three serological forms of ET, ETA, ETB, and ETD, which are linked to human impetigo [19]. The eta gene, encoding ETA, is located on a chromosome, whereas the etb gene, encoding ETB, is present on a plasmid. The eta and etb genes are generally found in S. aureus isolated from patients with impetigo. The etb gene is predominantly found in MRSA strains with mecA. In this context, our data support previous data [21, 23, 24, 35, 36].

ETA, is located on a chromosome, whereas the etb gene, encoding ETB, is present on a plasmid. The eta and etb genes are generally found in S. aureus isolated from patients with impetigo. The etb gene is predominantly found in MRSA strains with mecA. In this context, our data support previous data [21, 23, 24, 35, 36]. MLST is an excellent tool for investigating clonal evolution of MRSA. Six STs (ST5, ST8, ST89, ST91, ST121, and ST2117) were identified, and they were classified into four CCs (CC5, CC8, CC89, and CC121) in the present study. PVL-negative CA-MRSA strains, which belong to ST89 and ST91, have been associated with impetigo in Japanese children [22–24]. In this study, the most common ST in MRSA strains was also ST89 (57.1%), followed by ST91 (14.3%). ST89 and ST91 both belonged to the same CC89. We found that 11 of the 14 MRSA strains belonged to CC89 and carried diverse SCCmec types, IIb, IINT, Iva, and NT. Although one strain belonging to ST5, which has been commonly identified in HA-MRSA with type II SCCmec, was found in our MRSA isolates, it carried a nontypeable SCCmec element. The class of mec gene complex in the strain was class A, indicating that its SCCmec type should be a type other than type I, IV, V, VI, or VII [42]. In a previous study by Hisata et al. ST5 MRSA with SCCmec type IIa, which was indistinguishable from HA-MRSA strains (New York/Japan clone [5]) in Japan, accounted for 22.7% of MRSA strains isolated from healthy children [17]. Our ST5 MRSA strain was resistant to multiple drugs, including MINO and NDFX. Although ST8 MRSA strain with SCCmec type IVa [22, 43] and ST121 MRSA strain with SCCmec type V [44, 45] have been identified in CA-MRSA associated with impetigo, our MRSA strains with the same phenotypes did not carry PVL.

thy children [17]. Our ST5 MRSA strain was resistant to multiple drugs, including MINO and NDFX. Although ST8 MRSA strain with SCCmec type IVa [22, 43] and ST121 MRSA strain with SCCmec type V [44, 45] have been identified in CA-MRSA associated with impetigo, our MRSA strains with the same phenotypes did not carry PVL. It is generally accepted that bacterial resistance to antimicrobial agents parallels the frequency of use of the agents. Unlike HA-MRSA, CA-MRSA is generally susceptible to non-β-lactam antibiotics, such as aminoglycosides, tetracyclines, and fluoroquinolones. However, S. aureus isolates from patients with impetigo were highly resistant to GM and CAM, reflecting frequent usage of GM ointment for the treatment of skin infections including impetigo and frequent usage of CAM for respiratory infection in Japan [20, 21, 35, 36]. 5. Conclusions This study and a previous study indicated that PVL-negative CC89 strain with a marked divergence in SCCmec types predominated in Japanese communities, suggesting that it might be a Japanese domestic CA-MRSA clone. The existence of multiple types of SCCmec elements suggested that MRSA strains might be emerging locally in the community under the selective pressure of antibiotics. Furthermore, SCCmec II CA-MRSA may at least partially be beginning to replace SCCmec IVa CA-MRSA. Continuous monitoring of MRSA epidemiology using MLST and SCCmec typing and drug resistance trends would show change in the clonal distribution of MRSA, which may be crucial for the effective management of impetigo.

otics. Furthermore, SCCmec II CA-MRSA may at least partially be beginning to replace SCCmec IVa CA-MRSA. Continuous monitoring of MRSA epidemiology using MLST and SCCmec typing and drug resistance trends would show change in the clonal distribution of MRSA, which may be crucial for the effective management of impetigo. Acknowledgments The following clinicians participated in this study: Dr. Yasushi Akutsu, Dr. Hideki Arioka, Dr. Yasushi Deguchi, Dr. Shihoko Fujisaki, Dr. Tomoko Hayashi, Dr. Wataru Ikemoto, Dr. Akira Inagawa, Dr. Fumie Inyaku, Dr. Keisaku Imamura, Dr. Syousuke Imai, Dr. Noriyuki Ishii, Dr. Akihito Ishizaka, Dr. Shouichi Kasahara, Dr. Hideaki Kikuta, Dr. Mutsuko Konno, Dr. Kimikazu Kojima, Dr. Shuji Nakata, Dr. Naoko Nagano, Dr. Hitoshi Ochi, Dr. Yasuhisa Odagawa, Dr. Yachio Ohta, Dr. Miho Oshima, Dr. Hideto Otoi, Dr. Masahiko Mizumoto, Dr. Youko Sakata, Dr. Toshiya Satou, Dr. Mutsuo Shibata, Dr. Yoshitaka Shibuya, Dr. Kazunori Shinojima, Dr. Yuichi Taguchi, Dr. Chie Tobise, Dr. Hideaki Tsukazaki, Dr. Susumu Ukae, Dr. Michio Yamamoto, and Dr. Hideatsu Yoshimura. The authors thank Meiji Seika Pharma Co. Ltd. for financial support and molecular analysis and Mr. Stewart Chisholm for proofreading the manuscript. In addition, they thank Mr. Suguru Konishi for performing the statistical analysis. Table 1 List of primers. Gene Primer name Size of product (bp) Primer sequence (5′→3′) Reference (1) Detection of virulence gene [26] lukS-PV-lukF-PV luk-PV-1 ATCATTAGGTAAAATGTCTGGACATGATCCA luk-PV-2 433 GCATCAASTGTATTGGATAGCAAAAGC tst TST-1 TTCACTATTTGTAAAAGTGTCAGACCCACT

Acknowledgments The following clinicians participated in this study: Dr. Yasushi Akutsu, Dr. Hideki Arioka, Dr. Yasushi Deguchi, Dr. Shihoko Fujisaki, Dr. Tomoko Hayashi, Dr. Wataru Ikemoto, Dr. Akira Inagawa, Dr. Fumie Inyaku, Dr. Keisaku Imamura, Dr. Syousuke Imai, Dr. Noriyuki Ishii, Dr. Akihito Ishizaka, Dr. Shouichi Kasahara, Dr. Hideaki Kikuta, Dr. Mutsuko Konno, Dr. Kimikazu Kojima, Dr. Shuji Nakata, Dr. Naoko Nagano, Dr. Hitoshi Ochi, Dr. Yasuhisa Odagawa, Dr. Yachio Ohta, Dr. Miho Oshima, Dr. Hideto Otoi, Dr. Masahiko Mizumoto, Dr. Youko Sakata, Dr. Toshiya Satou, Dr. Mutsuo Shibata, Dr. Yoshitaka Shibuya, Dr. Kazunori Shinojima, Dr. Yuichi Taguchi, Dr. Chie Tobise, Dr. Hideaki Tsukazaki, Dr. Susumu Ukae, Dr. Michio Yamamoto, and Dr. Hideatsu Yoshimura. The authors thank Meiji Seika Pharma Co. Ltd. for financial support and molecular analysis and Mr. Stewart Chisholm for proofreading the manuscript. In addition, they thank Mr. Suguru Konishi for performing the statistical analysis. Table 1 List of primers. Gene Primer name Size of product (bp) Primer sequence (5′→3′) Reference (1) Detection of virulence gene [26] lukS-PV-lukF-PV luk-PV-1 ATCATTAGGTAAAATGTCTGGACATGATCCA luk-PV-2 433 GCATCAASTGTATTGGATAGCAAAAGC tst TST-1 TTCACTATTTGTAAAAGTGTCAGACCCACT TST-2 180 TACTAATGAATTTTTTTATCGTAAGCCCTT eta mpETA-1 ACTGTAGGAGCTAGTGCATTTGT mpETA-3 190 TGGATACTTTTGTCTATCTTTTTCATCAAC etb mpETB-1 CAGATAAAGAGCTTTATACACACATTAC mpETB-2 621 AGTGAACTTATCTTTCTATTGAAAAACACTC (2) SCCmec typing [28] (i) ccr gene complex type with mecA mecA mA1 TGCTATCCACCCTCAAACAGG mA2 286 AACGTTGTAACCACCCCAAGA

tst TST-1 TTCACTATTTGTAAAAGTGTCAGACCCACT TST-2 180 TACTAATGAATTTTTTTATCGTAAGCCCTT eta mpETA-1 ACTGTAGGAGCTAGTGCATTTGT mpETA-3 190 TGGATACTTTTGTCTATCTTTTTCATCAAC etb mpETB-1 CAGATAAAGAGCTTTATACACACATTAC mpETB-2 621 AGTGAACTTATCTTTCTATTGAAAAACACTC (2) SCCmec typing [28] (i) ccr gene complex type with mecA mecA mA1 TGCTATCCACCCTCAAACAGG mA2 286 AACGTTGTAACCACCCCAAGA Type 1 α1 AACCTATATCATCAATCAGTACGT βc 695 ATTGCCTTGATAATAGCCITCT Type 2 α2 TAAAGGCATCAATGCACAAACACT βc 937 ATTGCCTTGATAATAGCCITCT Type 3 α3 AGCTCAAAAGCAAGCAATAGAAT βc 1,791 ATTGCCTTGATAATAGCCITCT Type 4 α4.2 GTATCAATGCACCAGAACTT β4.2 1,287 TTGCGACTCTCTTGGCGTTT type 5 γF CGTCTATTACAAGATGTTAAGGATAAT γR 518 CCTTTATAGACTGGATTATTCAAAATAT (ii) mec gene complex class mecA ml6 CATAACTTCCCATTCTGCAGATG mA7 1,963 ATATACCAAACCCGACAACTACA mecB IS7 ATGCTTAATGATAGCATCCGAATG mA7 2,827 ATATACCAAACCCGACAACTACA mecC (C2) mA7 ATATACCAAACCCGACAACTACA IS2(iS-2) 804 TGAGGTTATTCAGATATTTCGATGT (iii) SCCmec type II subtyping IIa (II.1) kdpB2 TAAACTGTGTCACACGATCCAT kdpB1 287 GATTACTTCAGAACCAGGTCAT IIb (II.2) 2b3 GCTCTAAAAGTTGGATATGCG 2b4 1,518 TGGATTGAATCGACTAGAATCG IIE (II.3) 4b3 AACCAACAGTGGTTACAGCTT 4b4 726 CGGATTTTAGACTCATCACCAT II.4 II4-3 GTACCGCTGAATATTGATAGTGAT II4-1 2,003 ACTCTAATCCTAATCACCGAAC (iv) SCCmec type IV subtyping IVa (IV.1) 4al TTTGAATGCCCTCCATGAATAAAAT 4a3 458 AGAAAAGATAGAAGTTCGAAAGA IVb (IV.2) 4b3 AACCAACAGTGGTTACAGCTT 4b4 726 CGGATTTTAGACTCATCACCAT IVc (IV.3) 4c5 ATCCATTTCTCAGGAGTTAG 4c4 259 AGGAAATCGATGTCATTATAA IVd (IV.4) 4d3 AATTCACCCGTACCTGAGAA 4d4 1,242 AGAATGTGGTTATAAGATAGCTA (3) Multilocus sequence typing (MLST) [27] carbamate kinase (arcC) arcC-Up TTGATTCACCAGCGCGTATTGTC arcC-Dn 456 AGGTATCTGCTTCAATCAGCG

4a3 458 AGAAAAGATAGAAGTTCGAAAGA IVb (IV.2) 4b3 AACCAACAGTGGTTACAGCTT 4b4 726 CGGATTTTAGACTCATCACCAT IVc (IV.3) 4c5 ATCCATTTCTCAGGAGTTAG 4c4 259 AGGAAATCGATGTCATTATAA IVd (IV.4) 4d3 AATTCACCCGTACCTGAGAA 4d4 1,242 AGAATGTGGTTATAAGATAGCTA (3) Multilocus sequence typing (MLST) [27] carbamate kinase (arcC) arcC-Up TTGATTCACCAGCGCGTATTGTC arcC-Dn 456 AGGTATCTGCTTCAATCAGCG Shikimate dehydrogenase (aroE) aroE-Up ATCGGAAATCCTATTTCACATTC aroE-Dn 456 GGTGTTGTATTAATAACGATATC Glycerol kinase (glpF) glpF-Up CTAGGAACTGCAATCTTAATCC glpF-Dn 465 TGGTAAAATCGCATGTCCAATTC Guanylate kinase (gmk) gmk-Up ATCGTTTTATCGGGACCATC gmk-Dn 429 TCATTAACTACAACGTAATCGTA Phosphate acetyltransferase (pta) pta-Up GTTAAAATCGTATTACCTGAAGG pta-Dn 474 GACCCTTTTGTTGAAAAGCTTAA Triosephosphate isomerase (tpi) tpi-Up TCGTTCATTCTGAACGTCGTGAA tpi-Dn 402 TTTGCACCTTCTAACAATTGTAC Acetyl coenzyme A acetyltransferase (yqiL) yqiL-Up CAGCATACAGGACACCTATTGGC yqiL-Dn 516 CGTTGAGGAATCGATACTGGAAC Table 2 Age distribution in the 136 patients. Age (years) MSSA MRSA Total ＜1 13 2 15 1–<2 19 2 21 2–<3 16 5 21 3–<4 16 2 18 4–<5 15 1 16 5–<6 17 0 17 6–<7 15 2 17 7–<12 11 0 11 Total 122 14 136 Table 3 Molecular characterization of 14 MRSA strains isolated from patients with impetigo. Strain Age SCCmec typing MLST Presence of exotoxin genes SCCmec type ccr mec ST CC Allelic profile lukS-PV-lukF-PV tst eta etb

＜1 13 2 15 1–<2 19 2 21 2–<3 16 5 21 3–<4 16 2 18 4–<5 15 1 16 5–<6 17 0 17 6–<7 15 2 17 7–<12 11 0 11 Total 122 14 136 Table 3 Molecular characterization of 14 MRSA strains isolated from patients with impetigo. Strain Age SCCmec typing MLST Presence of exotoxin genes SCCmec type ccr mec ST CC Allelic profile lukS-PV-lukF-PV tst eta etb T-11 1 yr IIb 2 A 2117 89 1-310-28-18-18-33-50 − − − + T-33 6 yrs 4 mos IVa 2 B 91 89 1-26-28-18-18-54-50 − − − + T-55 2 yrs 6 mos NT NT A 89 89 1-26-28-18-18-33-50 − − − + T-64 2 yra 6 mos IINT 2 A 89 89 1-26-28-18-18-33-50 − − − + T-71 6 yrs 6 mos IIb 2 A 89 89 1-26-28-18-18-33-50 − − − + T-79 3 yrs 10 mos IVa 2 B 91 89 1-26-28-18-18-54-50 − − − + T-83 3 yrs 11 mos IINT 2 A 89 89 1-26-28-18-18-33-50 − − − + T-114 2 yrs 3 mos NT NT A 89 89 1-26-28-18-18-33-50 − − − + T-119 2 yrs 2 mos NT NT A 5 5 1-4-1-4-12-1-10 − + − − T-125 4 yrs V 5 C2 121 121 6-5-6-2-7-14-5 − − + − T-132 19 ds IINT 2 A 89 89 1-26-28-18-18-33-50 − − − + T-137 1 yr 1 mo IIb 2 A 89 89 1-26-28-18-18-33-50 − − − + T-139 10 ds IINT 2 A 89 89 1-26-28-18-18-33-50 − − − + T-144 2 yrs 1 mo IVa 2 B 8 8 3-3-1-1-4-4-3 − + − − NT: nontypeable. Table 4 Comparison of antimicrobial susceptibilities of MRSA and MSSA.

T-11 1 yr IIb 2 A 2117 89 1-310-28-18-18-33-50 − − − + T-33 6 yrs 4 mos IVa 2 B 91 89 1-26-28-18-18-54-50 − − − + T-55 2 yrs 6 mos NT NT A 89 89 1-26-28-18-18-33-50 − − − + T-64 2 yra 6 mos IINT 2 A 89 89 1-26-28-18-18-33-50 − − − + T-71 6 yrs 6 mos IIb 2 A 89 89 1-26-28-18-18-33-50 − − − + T-79 3 yrs 10 mos IVa 2 B 91 89 1-26-28-18-18-54-50 − − − + T-83 3 yrs 11 mos IINT 2 A 89 89 1-26-28-18-18-33-50 − − − + T-114 2 yrs 3 mos NT NT A 89 89 1-26-28-18-18-33-50 − − − + T-119 2 yrs 2 mos NT NT A 5 5 1-4-1-4-12-1-10 − + − − T-125 4 yrs V 5 C2 121 121 6-5-6-2-7-14-5 − − + − T-132 19 ds IINT 2 A 89 89 1-26-28-18-18-33-50 − − − + T-137 1 yr 1 mo IIb 2 A 89 89 1-26-28-18-18-33-50 − − − + T-139 10 ds IINT 2 A 89 89 1-26-28-18-18-33-50 − − − + T-144 2 yrs 1 mo IVa 2 B 8 8 3-3-1-1-4-4-3 − + − − NT: nontypeable. Table 4 Comparison of antimicrobial susceptibilities of MRSA and MSSA. Antimicrobial agent Resistance breakpoint (μg/mL) MSSA (n = 122) MRSA (n = 14) Range (μg/mL) MIC90 (μg/mL) Resistance (%) Range (μg/mL) MIC90 (μg/mL) Resistance (%) MPIPC 4 ≦2 ≦2 0 4–>8 >8 100 ABPC 0.5 ≦0.12  5–>16 16 71.3 4–>16 >16 100 CCL 16 1–16 4 1.6 4–>16 >16 92.9 CDTR 2* 0.5 –2 1 0.8 2–>16 >16 100 CAM 4 ≦0.125–>16 >16 67.2 >16 >16 100 CLDM 1 ≦0.125–0.25 0.25 0 ≦0.125–>16 >16 71.4 FOM 8** 0.25–>16 16 20.5 1–>16 8 14.3 MINO 8 ≦0.125–0.25 ≦0.125 0 ≦0.125–16 0.25 7.1 GM 8 0.25–>16 >16 54.1 1–>16 >16 92.9 VCM 4 ≦0.5–1 1 0 1-2 1 0 NDFX 4** ≦0.125–2 ≦0.125 0 ≦0.125–>16 ≦0.125 7.1 FA 2** ≦0.125–>16 0.5 0.82 ≦0.125–1 0.5 0 The breakpoints for these antimicrobial agents were determined according to the interpretation criteria of Clinical and Laboratory Standards Institute (CLSI). Susceptibility to antimicrobial drugs for which breakpoints are not determined by CLSI was according to the interpretation criteria of the Japan Society for Chemotherapy (*) or defined in this study (**).

nts were determined according to the interpretation criteria of Clinical and Laboratory Standards Institute (CLSI). Susceptibility to antimicrobial drugs for which breakpoints are not determined by CLSI was according to the interpretation criteria of the Japan Society for Chemotherapy (*) or defined in this study (**). Table 5 Antimicrobial susceptibilities of 14 MRSA strains. Strain MIC (μg/mL) of antimicrobial agents MPIPC ABPC CCL CDTR CAM CLDM FOM MINO GM VCM NDFX FA T-11 >8 >16 >16 >16 >16 >16 1 ≦0.125 >16 2 ≦0.125 0.25 T-33 >8 16 >16 4 >16 >16 2 ≦0.125 >16 1 ≦0.125 0.25 T-55 >8 16 >16 8 >16 >16 4 0.25 >16 1 ≦0.125 0.25 T-64 >8 >16 >16 >16 >16 >16 2 ≦0.125 >16 1 ≦0.125 0.25 T-71 >8 16 16 4 >16 >16 2 ≦0.125 >16 1 ≦0.125 ≦0.125 T-79 >8 16 >16 4 >16 ≦0.125 4 ≦0.125 >16 1 ≦0.125 0.25 T-83 >8 16 >16 4 >16 >16 2 0.25 >16 1 ≦0.125 ≦0.125 T-114 >8 16 >16 16 >16 >16 1 ≦0.125 >16 1 ≦0.125 0.25 T-119 >8 >16 >16 >16 >16 ≦0.125 >16 16 >16 1 >16 0.5 T-125 4 4 4 2 >16 0.25 4 ≦0.125 >16 1 ≦0.125 0.25 T-132 >8 16 >16 16 >16 >16 2 ≦0.125 >16 1 ≦0.125 ≦0.125 T-137 >8 >16 >16 >16 >16 >16 2 ≦0.125 >16 1 ≦0.125 1 T-139 >8 16 >16 16 >16 >16 2 ≦0.125 >16 1 ≦0.125 0.25 T-144 >8 16 >16 16 >16 ≦0.125 8 ≦0.125 1 1 ≦0.125 0.25

1. Introduction Of the various respiratory diseases, influenza is a major cause of mortality and morbidity among patients, particularly the very young and the elderly [1, 2]. Two options are available for moderating the effect of the influenza virus: vaccines, which although effective, are underutilized and not completely protective because of frequent antigenic shifts in the viral surface proteins and antiviral drugs [2, 3]. Antiviral drugs have emerged as attractive options in the battle against influenza. Amantadine, oseltamivir, zanamivir, laninamivir, and peramivir are the five antiviral drugs currently used to treat influenza in Japan [2]. However, firm guidelines for prescribing these drugs remain to be established. Amantadine is limited in effectiveness because of its lack of activity against influenza B virus [4] and the rapid emergence of resistant viral strains. Hemagglutinin and neuraminidase, two glycoproteins present on the viral surface, have antiviral targets [5]. Recently, oseltamivir and zanamivir, two influenza neuraminidase inhibitors, have commonly been prescribed for influenza A and B [6–12]. Laninamivir is a long-acting neuraminidase inhibitor for the treatment of influenza. A single inhalation of laninamivir is effective for the treatment of influenza, including that caused by the oseltamivir-resistant viruses, in adults [13, 14]. However, seriously ill and pediatric patients need a parenteral formulation because the injectable drug is much easier to administer in such cases than oral oseltamivir, inhaled zanamivir, or laninamivir.

ive for the treatment of influenza, including that caused by the oseltamivir-resistant viruses, in adults [13, 14]. However, seriously ill and pediatric patients need a parenteral formulation because the injectable drug is much easier to administer in such cases than oral oseltamivir, inhaled zanamivir, or laninamivir. In Japan, peramivir has recently been approved for use not only in adults but also in children over 1 month of age [15]. In this study, we compared the efficacy of intravenous peramivir with that of other neuraminidase inhibitors for treating influenza infections in pediatric patients. 2. Material and Methods The present study included 223 patients under the age of 18 years diagnosed with influenza at the Hikita Pediatric Clinic between February and April 2011. The patients presented with a complaint of fever lasting for less than 48 h, and they were clinically diagnosed with rapid diagnostic tests. Specimens from nasal swabs or nasal aspirates were subjected to antigen detection. Commercial antigen detection kits based on immunochromatography (The Quick Chaser Flu A, B rapid antigen test [Mizuho Medy Co., Ltd. Saga, Japan]) was used for the diagnosis of influenza A or B. Subsequently, after obtaining informed consent from the parents, 35 patients diagnosed with influenza A by the rapid antigen A, B test underwent a 2009 influenza A H1N1 virus infection test using the Quick Chaser Flu AH1pdm (Mizuho Medy) to differentiate patients with 2009 H1N1 influenza from those with seasonal influenza.

equently, after obtaining informed consent from the parents, 35 patients diagnosed with influenza A by the rapid antigen A, B test underwent a 2009 influenza A H1N1 virus infection test using the Quick Chaser Flu AH1pdm (Mizuho Medy) to differentiate patients with 2009 H1N1 influenza from those with seasonal influenza. The efficacy, potential adverse effects, and convenience of administration of the five antiviral drugs were explained to the patients and/or their families prior to study initiation. The choice of antivirals for influenza treatment was then discussed, and after obtaining informed consent from patients and/or their families, all patients underwent antiviral therapy. Laninamivir was administered as a single-inhalation dose of 40 mg for patients aged ≥10 years or 20 mg for patients aged <10 years. Peramivir was administered intravenously as a single dose of 10 mg/kg/dose (maximum 300 mg/dose) over a period of 15 min. Oral oseltamivir was prescribed twice per day in divided dosages for 5 days (4 mg/kg/day, maximum of 150 mg/day). Zanamivir was administered twice per day at an inhalation dosage of 20 mg/day for 5 days.

ivir was administered intravenously as a single dose of 10 mg/kg/dose (maximum 300 mg/dose) over a period of 15 min. Oral oseltamivir was prescribed twice per day in divided dosages for 5 days (4 mg/kg/day, maximum of 150 mg/day). Zanamivir was administered twice per day at an inhalation dosage of 20 mg/day for 5 days. In 2007, the Ministry of Health, Labor, and Welfare, Japan, issued emergency instructions suspending the use of oseltamivir in patients aged 10–19 years [16]. Accordingly, oseltamivir was not prescribed for the teenage patients with influenza in our study. In addition, inhaled drugs are difficult to use in infants. Therefore, laninamivir and zanamivir were prescribed for patients aged ≥5 years. We created 2 age groups for statistical analysis: patients aged <10 years and those aged between 5 and 18 years. We compared the duration of fever after initiating antiviral therapy. Fever was considered positive if the patient's body temperature was ≥37.5°C. When a patient's body temperature dropped and remained at <37.5°C for 48 continuous h, the fever was considered as resolved. Data were analyzed using JMP software version 8.0.2 (SAS Institute, Cary, NC, USA). The Wilcoxon rank-sum test and/or the Kruskal-Wallis rank-sum test were used to compare the ages of influenza patients treated with laninamivir, oseltamivir, peramivir, or zanamivir. We used Kaplan-Meier analysis and the log-rank test to compare the duration of fever after antiviral therapy initiation. A P value of <0.05 was considered statistically significant.

l-Wallis rank-sum test were used to compare the ages of influenza patients treated with laninamivir, oseltamivir, peramivir, or zanamivir. We used Kaplan-Meier analysis and the log-rank test to compare the duration of fever after antiviral therapy initiation. A P value of <0.05 was considered statistically significant. 3. Results We treated 131 (70 males) and 92 (53 males) patients with influenza A and B, respectively. Nine of 35 (25.7%) patients tested positive in the 2009 influenza A H1N1 test. Patients were administered the following antiviral drugs: laninamivir (n = 14), oseltamivir (n = 125), peramivir (n = 45), and zanamivir (n = 39). No patient was treated with amantadine in this study. The median ages of the patients treated with laninamivir, oseltamivir, peramivir, and zanamivir were 132.5 months (range, 80–164), 56.5 months (range, 3–135), 76 months (range, 2–219), and 124.5 months (range, 59–198), respectively. No adverse effects were observed with any of the antiviral drugs used in this study.

y. The median ages of the patients treated with laninamivir, oseltamivir, peramivir, and zanamivir were 132.5 months (range, 80–164), 56.5 months (range, 3–135), 76 months (range, 2–219), and 124.5 months (range, 59–198), respectively. No adverse effects were observed with any of the antiviral drugs used in this study. We compared <10-year-old influenza patients who were administered either oseltamivir or peramivir (Table 1). The age of patients treated with oseltamivir was not significantly different from that of patients treated with peramivir. We also compared 5–18-year-old influenza patients treated with peramivir, zanamivir, or laninamivir. The age of patients did not significantly differ between treatment groups. The median duration of fever after zanamivir treatment in 5–18-year-old patients with influenza A was 2 days (range, 0–3 days), whereas that after peramivir treatment was 1 day (range, 1-2 days); this difference was statistically significant (P = 0.0283). The median duration of fever after laninamivir treatment in 5–18-year-old patients with influenza B was 3 days (range, 1–5 days), whereas that after peramivir treatment was 1 day (range, 1–4 days); this difference was also statistically significant (P = 0.0097). No other significant differences were observed for any of the other drug/disease type/age group combinations.

ment in 5–18-year-old patients with influenza B was 3 days (range, 1–5 days), whereas that after peramivir treatment was 1 day (range, 1–4 days); this difference was also statistically significant (P = 0.0097). No other significant differences were observed for any of the other drug/disease type/age group combinations. 4. Discussion Hernandez et al. reported their clinical experience with children (n = 11) hospitalized for 2009 influenza A (H1N1) and treated with peramivir [17]. In their study, all patients had rapidly progressing, radiographically confirmed viral pneumonia with respiratory failure. In our study, peramivir was administered to patients with influenza, including 2009 influenza A (H1N1), seasonal influenza H3N2, and influenza B. None of the patients had severe disease and all were treated as outpatients. Severe adult influenza infection has been reportedly treated with peramivir [18]. Although oseltamivir and zanamivir are used for severe influenza infections such as encephalopathy [19], peramivir is more suitable for such infections because it is much easier to administer (intravenously) to a severe case when compared with oral oseltamivir or inhaled zanamivir. Currently, most patients in Japan with an influenza-like illness are tested using rapid diagnostic tests and treated with an appropriate choice of antiviral drugs if results prove positive [20]. Particularly, after the 2009 H1N1 influenza pandemic virus was isolated, antiviral therapy was recommended for all pediatric influenza patients.

ients in Japan with an influenza-like illness are tested using rapid diagnostic tests and treated with an appropriate choice of antiviral drugs if results prove positive [20]. Particularly, after the 2009 H1N1 influenza pandemic virus was isolated, antiviral therapy was recommended for all pediatric influenza patients. A limitation of this study was the small study sample and the lack of a randomized open label study; therefore, more data are needed before the clinical implications of this study become clear. We compared only fever duration and did not analyze any other symptoms because influenza patients in Japan can only return to school 48 hours after fever resolution. However, we were able to confirm the usefulness of peramivir in children. Randomized case-control studies with sufficiently large populations will be conducted during the next influenza season following approval by an ethics committee. Recent developments in antigen detection tests have made it possible to differentiate between influenza A, 2009 influenza A (H1N1), and influenza B. Therefore, on determination of the variant involved, physicians, patients, and their families can make informed decisions regarding choice of antiviral drug. The inhalation drugs laninamivir and zanamivir are difficult to use in infants. Vaccines are not completely protective because of frequent antigenic shifts in the viral surface proteins, as observed in the last pandemic. The results of this study suggest that peramivir is an important and a feasible therapeutic option for pediatric influenza patients.

Recent developments in antigen detection tests have made it possible to differentiate between influenza A, 2009 influenza A (H1N1), and influenza B. Therefore, on determination of the variant involved, physicians, patients, and their families can make informed decisions regarding choice of antiviral drug. The inhalation drugs laninamivir and zanamivir are difficult to use in infants. Vaccines are not completely protective because of frequent antigenic shifts in the viral surface proteins, as observed in the last pandemic. The results of this study suggest that peramivir is an important and a feasible therapeutic option for pediatric influenza patients. Conflict of Interests The authors declare that they have no conflict of interests. Table 1 Comparison of the effectiveness of oseltamivir, zanamivir, laninamivir, and peramivir against influenza virus infection. Groups Therapy Number of patients Median age in months (range) Duration of fever before treatment, median day (range) Duration of fever after treatment, median day (range) P value Influenza type Ages (years) Influenza A 0–9 Oseltamivir 83 51 (3–118) 1 (0–2) 2 (0–6) 0.4499 Peramivir 22 41.5 (2–106) 1 (0–2) 1.5 (1–3) Influenza B 0–9 Oseltamivir 41 81 (25–118) 1 (0–2) 2 (0–4) 0.6435 Peramivir 13 75 (10–118) 0 (0–2) 2 (1–4) Influenza A 5–18 Laninamivir 1 80 (80-80) 0 (0-0) 1 (1-1) Not performed Peramivir 15 94 (72–219) 1 (0–2) 1 (1-2) Zanamivir 18 34.5 (69–198) 1 (0–2) 2 (0–3) 0.0242*

Influenza A 0–9 Oseltamivir 83 51 (3–118) 1 (0–2) 2 (0–6) 0.4499 Peramivir 22 41.5 (2–106) 1 (0–2) 1.5 (1–3) Influenza B 0–9 Oseltamivir 41 81 (25–118) 1 (0–2) 2 (0–4) 0.6435 Peramivir 13 75 (10–118) 0 (0–2) 2 (1–4) Influenza A 5–18 Laninamivir 1 80 (80-80) 0 (0-0) 1 (1-1) Not performed Peramivir 15 94 (72–219) 1 (0–2) 1 (1-2) Zanamivir 18 34.5 (69–198) 1 (0–2) 2 (0–3) 0.0242* Influenza B 5–18 Laninamivir 13 134 (91–164) 0 (0–2) 3 (1–5) 0.0097* Peramivir 13 98 (63–167) 1 (0-1) 1 (1–4) Zanamivir 20 120 (87–179) 0 (0–2) 2 (0–4) 0.2979 “Not performed” indicates that statistical analysis was not performed because the number of subjects was too small. *indicates a significant difference between peramivir and the other administered drug. (Both the 0–9 years group and the 5–18 years group included 5–9-year-old children treated with peramivir).

1. Introduction Both the palmar grasp reflex and the plantar grasp reflex are very primitive in the sense that they can be elicited in all normal preterm infants at as early as 25 weeks of postconceptional age (PCA) [1]. During routine ultrasound examination, fetal palmar reflex grasping of the umbilical cord has been repeatedly observed, which first appears at 16 weeks' gestation [2–4]. These grasp reflexes are easy to elicit but have been proved to be of distinctive clinical significance for the early detection of infants with neurodevelopmental abnormalities [5–8]. The Moro reflex was first described by Ernst Moro in 1918 [9]. This reflex is also primitive, being seen in some preterm infants at 25 weeks of PCA and in the majority by 30 weeks of PCA [1]. Since his report, many authors have devised a variety of methods for eliciting the reflex, and the underlying neural mechanism including afferent pathways of the reflex has been a subject of considerable discussion [10–13]. The phylogenetic meaning of this reflex also remains unclear [9, 10].

ty by 30 weeks of PCA [1]. Since his report, many authors have devised a variety of methods for eliciting the reflex, and the underlying neural mechanism including afferent pathways of the reflex has been a subject of considerable discussion [10–13]. The phylogenetic meaning of this reflex also remains unclear [9, 10]. It is interesting that, in spite of the great difference in the motor behavior, there is a close interrelationship between these primitive reflexes in the responses: the palmar grasp reflex inhibits the Moro reflex [12, 14–17]. This paper mainly concerns the clinical significance and neural mechanism of the grasp reflex and the Moro reflex and also attempts to discuss the meaning of these reflexes based on comparison of the responses in human infants with those in monkey infants and the hierarchical interrelation of the responses of the primitive reflexes.

nly concerns the clinical significance and neural mechanism of the grasp reflex and the Moro reflex and also attempts to discuss the meaning of these reflexes based on comparison of the responses in human infants with those in monkey infants and the hierarchical interrelation of the responses of the primitive reflexes. 2. The Grasp Reflex 2.1. Elicitation To elicit the palmar grasp reflex, the examiner inserts his or her index finger into the palm of the infant from the ulnar side and applies light pressure to the palm, with the infant lying on a flat surface in the symmetrical supine position while awake [18–20]. Tactile without pressure and nociceptive stimulation of the palm are both inadequate. The response of the reflex comprises flexion of all fingers around the examiner's finger, which is composed of two phases: finger closure and clinging. The latter occurs as a reaction to the proprioceptive stimulation of the tendons of the finger muscles due to slight traction subsequent to the application of pressure to the palm [21, 22].

eflex comprises flexion of all fingers around the examiner's finger, which is composed of two phases: finger closure and clinging. The latter occurs as a reaction to the proprioceptive stimulation of the tendons of the finger muscles due to slight traction subsequent to the application of pressure to the palm [21, 22]. The plantar grasp reflex is elicited by pressing a thumb against the sole of a foot just behind the toes [6, 18, 23]. The state and position are the same as for eliciting the palmar grasp reflex. The response of the reflex consists of flexion and adduction of all the toes. Milani-Comparetti and Gidoni [24] devised another method for eliciting the plantar grasp reflex. They tested for the presence of the reflex by placing the infant in a supported standing position, stimulating the soles of the feet by floor contact, looking for plantar flexion of the toes. Determination of whether the response of the hands and feet is a true reflex or a voluntary grasping movement can be difficult in older infants. An examiner should test several times with an appropriate interval between the tests, carefully observing the infant's behavior [8, 25].

oking for plantar flexion of the toes. Determination of whether the response of the hands and feet is a true reflex or a voluntary grasping movement can be difficult in older infants. An examiner should test several times with an appropriate interval between the tests, carefully observing the infant's behavior [8, 25]. 2.2. Clinical Significance 2.2.1. Responses in Normal Infants The grasp reflexes of the hands and feet in normal term infants have been studied by several authors. Their results were fairly consistent regarding the times of the appearance and disappearance of the reflexes. The palmar grasp reflex and the plantar grasp reflex can be elicited in all infants during the first 3 and 6 months of age, respectively. Thereafter they decrease along with the intensity of the responses, usually disappearing by 6 and 12 months of age, respectively [6, 7, 25–27]. The disappearance of the reflexes is significantly related to the commencement of the voluntary use of hands or standing [28, 29].

he first 3 and 6 months of age, respectively. Thereafter they decrease along with the intensity of the responses, usually disappearing by 6 and 12 months of age, respectively [6, 7, 25–27]. The disappearance of the reflexes is significantly related to the commencement of the voluntary use of hands or standing [28, 29]. In contrast to the studies involving term infants, those involving preterm infants have been few. Allen and Capute [1] examined 47 infants and concluded that the intensity of the primitive reflexes including the grasp reflexes of the hands and feet in the preterm infant at term (40 weeks of PCA) was similar to that in full-term newborns reported in the literature. We analyzed the data for 834 infants (332 term and 502 preterm) who were later confirmed to be normal on follow-up examination between ages 3 to 9 years [6]. The primitive reflex responses of preterm infants were compared with those of term infants, according to corrected age as to expected birth date. No difference was evident in the changes in responses including that of the plantar grasp reflex between term and preterm infant groups throughout the first year of life.

6]. The primitive reflex responses of preterm infants were compared with those of term infants, according to corrected age as to expected birth date. No difference was evident in the changes in responses including that of the plantar grasp reflex between term and preterm infant groups throughout the first year of life. 2.2.2. Abnormal Responses In general, a primitive reflex in infants is regarded as abnormal when it is absent or diminished during the period it should be actively elicitable or lasts beyond the normal age limit for its disappearance. An exaggerated reflex can also be abnormal. The response of the palmar grasp reflex may be less intense during the first and second days after birth [18]. The absence of this reflex usually reflects peripheral (i.e., root, plexus, or nerve) or spinal cord involvement, especially regarding asymmetrical responses [18, 30, 31]. However, lesions of the upper brain structures also can affect the response. The response may be increased and retained longer, compared with that in normal infants, on the affected side(s) of the upper limb(s) in infants with spastic hemiplegia or quadriplegia, whereas it is very weak in infants with cerebral palsy (CP) of the athetoid type [26, 27, 32].

tructures also can affect the response. The response may be increased and retained longer, compared with that in normal infants, on the affected side(s) of the upper limb(s) in infants with spastic hemiplegia or quadriplegia, whereas it is very weak in infants with cerebral palsy (CP) of the athetoid type [26, 27, 32]. The clinical value of the plantar grasp reflex in infants has been investigated in more detail than that of the palmar grasp reflex. In 1932, Brain and Curran [33] reported two cases showing no response at the age of 6 months who both suffered from bilateral marked spasticity and one case showing a vigorous response at the age of 2.5 years who suffered from congenital choreoathetosis. They also investigated the reflex in 59 patients with Down's syndrome, ranging in age from 1 to 45 years. They discovered that up to age 20 years, the response was present in 25 of 42 patients and, after age 20 years, in 3 of 17. We analyzed 617 infants (291 term and 326 preterm) whose plantar grasp reflex was examined before 1 year of age [5]. Their diagnoses were confirmed by follow-up examinations and comprised normality in 458, CP in 78, and mental retardation (MR) in 81. We obtained several results: the reflex reactivity in infants with CP of the spastic type was significantly reduced; infants with CP of the athetoid type exhibited an extremely strong retention of the reflex; infants with MR also exhibited a tendency for prolonged retention of the reflex; the reflex profile in infants with CP of the athetoid type with spasticity, or of the ataxic type, was not different from that in normal control subjects. Other authors also reported the characteristics of the plantar grasp reflex in children with neurodevelopmental abnormalities of various types, and their results were well consistent with ours [26, 27, 34].

the athetoid type with spasticity, or of the ataxic type, was not different from that in normal control subjects. Other authors also reported the characteristics of the plantar grasp reflex in children with neurodevelopmental abnormalities of various types, and their results were well consistent with ours [26, 27, 34]. A reduced or negative plantar grasp reflex during early infancy can be a sensitive indicator of later development of spasticity. Of 2267 infants whose plantar grasp reflex had been examined before 1 year of age, we analyzed the neurodevelopmental outcome in 47 infants exhibiting a negative response during the first 6 months of age and in 46 infants exhibiting a significantly reduced response at ages 1 to 4 months [6]. The diagnoses comprised CP in 75, MR in 7, borderline intelligence in 2, motor delay in 1, and normality in 8. Of the 75 cases with CP, 69 had the spastic type and 4 the athetoid type with spasticity. The total number of patients with CP of the spastic type or the mixed type with spasticity among the entire 2267 infants in this series was 107, and 73 of these 107 patients with CP with spasticity (68.2%) exhibited a negative or reduced response during early infancy. In the remaining 34 infants with CP with spasticity who exhibited a normal response during the corresponding period, the extent of motor disturbance was significantly milder [35].

series was 107, and 73 of these 107 patients with CP with spasticity (68.2%) exhibited a negative or reduced response during early infancy. In the remaining 34 infants with CP with spasticity who exhibited a normal response during the corresponding period, the extent of motor disturbance was significantly milder [35]. A high concordance between the side of an abnormal plantar grasp reflex during infancy and the laterality of the disturbance of motor function in children with CP of the spastic type was demonstrated [36]. The side affected, or more affected, in motor function was in accordance with the side that exhibited a diminished or more diminished response in most subjects with spastic hemiplegia, diplegia, and quadriplegia. 2.3. Neural Mechanism 2.3.1. Spinal Reflex Center and Higher Brain Mechanism Because anencephalic infants demonstrate a positive grasp reflex in both the hands and feet, the cerebral hemispheres are apparently not necessary for the reflexes [37–39]. Shahani et al. [40] reported that the latency of the palmar grasp reflex was 40 msec on direct electrical stimulation of the afferent fibers in the median and ulnar nerves at the level of the wrist in an adult patient who showed the palmar grasp reflex following a vascular lesion in the cerebral hemisphere. This short latency excludes the long cerebral reflex arcs and puts this grasp reflex in the category of segmental reflexes mediated at the level of the spinal cord.

n and ulnar nerves at the level of the wrist in an adult patient who showed the palmar grasp reflex following a vascular lesion in the cerebral hemisphere. This short latency excludes the long cerebral reflex arcs and puts this grasp reflex in the category of segmental reflexes mediated at the level of the spinal cord. The spinal reflex center, however, is controlled by a higher brain mechanism. The grasp reflexes can be elicited in neonates and early infants as a result of insufficient control of the spinal mechanism by the immature brain, but the reflexes gradually disappear with age, due to the increased inhibition accompanying brain maturation [23, 41]. Adult patients with lesions in the frontal lobes sometimes exhibit a grasp reflex of the hands and feet [23, 33, 37, 41–45]. Such reappearance of these reflexes in adults is attributed to the release of the spinal reflex center from the disturbed higher brain mechanism, suggesting that these reflexes are only inhibited, and not lost, after the infantile period.

lobes sometimes exhibit a grasp reflex of the hands and feet [23, 33, 37, 41–45]. Such reappearance of these reflexes in adults is attributed to the release of the spinal reflex center from the disturbed higher brain mechanism, suggesting that these reflexes are only inhibited, and not lost, after the infantile period. Many attempts have been made to determine the pathological site of the grasp reflexes by means of clinical observation or animal experiments [37, 46–48]. In 1927, Adie and Critchley [37] analyzed 13 patients with a tumor or vascular lesion in a frontal lobe who exhibited a palmar grasp reflex on the side opposite to that of the diseased frontal lobe. They confirmed that the palmar grasp reflex was most evident when pyramidal signs were absent, and thus they concluded that the lesion responsible for the reflex was extrapyramidal. In 1938, Goldstein [41] described that a lesion involving the medial aspect of a frontal lobe seemed to be especially prone to produce the plantar grasp reflex on the opposite side. More recent studies have implicated lesions of the medial or lateral frontal cortex anterior to the primary motor area, that is, the supplementary motor area (SMA), premotor cortex, and cingulate motor cortex, as the etiology of the palmar grasp reflex [23, 45, 48, 49]. These findings indicate that nonprimary motor areas comprise substantial portions of the brain that exert inhibitory control over the spinal reflex mechanism underlying the grasp reflexes and that the destruction of these structures will release the inhibitory control and thus lead to the reappearance of the reflexes.

These findings indicate that nonprimary motor areas comprise substantial portions of the brain that exert inhibitory control over the spinal reflex mechanism underlying the grasp reflexes and that the destruction of these structures will release the inhibitory control and thus lead to the reappearance of the reflexes. 2.3.2. Descending Control of the Spinal Circuit by the Nonprimary Motor Cortex Nonprimary motor areas play important roles in the planning, preparation, initiation, and execution of motor behavior [50–55]. These areas contain corticospinal neurons and thereby a somatotopically arranged map of the body [54, 56–58]. However, the cortical projections from these areas to the spinal cord do not necessarily have a direct influence on spinal motor neurons. The majority of the neurons terminate in the intermediate zone in the spinal cord and connect with interneurons [53, 56, 59, 60]. The essential roles of nonprimary motor areas in the spinal cord are thought to be in the preparation and modulation of the spinal circuitry through interneurons rather than the generation of movements that require a direct command to the spinal motor neurons [52, 53, 60, 61]. On the other hand, because interneurons are substantially related to the modulation of spinal reflexes [52, 57], the descending inhibitory control of the grasp reflexes by nonprimary motor areas may also occur through spinal interneurons. As brain maturation proceeds, increasing control of the spinal circuit by the nonprimary motor areas will replace the simple reflex grasping during early infancy, with more regulatory and adaptive voluntary grasping in older infants.

e grasp reflexes by nonprimary motor areas may also occur through spinal interneurons. As brain maturation proceeds, increasing control of the spinal circuit by the nonprimary motor areas will replace the simple reflex grasping during early infancy, with more regulatory and adaptive voluntary grasping in older infants. However, the proportion of patients with a lesion of a nonprimary motor area in whom a grasp reflex can be elicited is not necessarily high. In the series of De Renzi and Barbieri [44], the palmar grasp reflex was elicited in 21 of 32 (66%) patients with a medial frontal lesion and in 8 of 30 (26%) with a lateral frontal lesion. On the other hand, a minority of patients with a deep lesion including the basal ganglia without frontal cortical damage were reported to exhibit a positive palmar grasp reflex [44], and the extension of an SMA lesion into more lateral regions of area 6 may increase the strength of the grasp reflex [48]. These observations seem to indicate that the inhibitory stimuli from upper brain structures travel via multiple routes [44], and the extent of the release of control depends not only on the specificity of the location but also on the extent and severity of the lesions.

ay increase the strength of the grasp reflex [48]. These observations seem to indicate that the inhibitory stimuli from upper brain structures travel via multiple routes [44], and the extent of the release of control depends not only on the specificity of the location but also on the extent and severity of the lesions. Clinical studies have revealed that some patients with a frontal lesion exhibit a grasp reflex of the hands or feet, or both [33, 41, 43]. This finding suggests that different sites are specific to each of the grasp reflexes of the hands and feet within the nonprimary motor cortex. As demonstrated, each nonprimary motor area retains some degree of a somatotopically arranged map of the body that represents peripheral innervation, which is dominated by descending pathways from the area through connections to different levels of the spinal cord [54, 56–58]. We speculate that the grasp reflexes of the hands and feet may be elicited according to the specific location of each lesion corresponding to the fingers or toes on the map in nonprimary motor areas.

s dominated by descending pathways from the area through connections to different levels of the spinal cord [54, 56–58]. We speculate that the grasp reflexes of the hands and feet may be elicited according to the specific location of each lesion corresponding to the fingers or toes on the map in nonprimary motor areas. 2.3.3. Factors Affecting Responses in Infants In normal circumstances, higher brain centers control the spinal mechanism that regulates the coordination among agonists, antagonists, and synergists in an adaptable manner by means of reciprocal innervation. In children with spastic type CP, however, deviation in terms of reciprocal innervation caused by damage to the pyramidal tract leads to excess cocontraction at proximal joints, whereas deviation leads to excess reciprocal inhibition through spastic antagonists at distal joints, inducing weakness of the agonists [62]. A diminished or negative plantar grasp reflex may be caused by weak toe flexors, as induced by excessive reciprocal inhibition of the flexors through the predominance of extensors over flexors in tonus in a lower limb in these children. They also exhibit the predominance of finger flexors over extensors in their upper limbs, which may cause a facilitated palmar grasp reflex on the affected side(s) in spastic hemiplegia and quadriplegia.

bition of the flexors through the predominance of extensors over flexors in tonus in a lower limb in these children. They also exhibit the predominance of finger flexors over extensors in their upper limbs, which may cause a facilitated palmar grasp reflex on the affected side(s) in spastic hemiplegia and quadriplegia. On the other hand, in children with CP of the athetoid type, the deviation of reciprocal innervation always leads to excess reciprocal inhibition. Any attempt at movement produces excessive relaxation of the antagonists, inducing an extreme range of movements [62]. Although the released control of the spinal reflex mechanism because of a lesion in the basal ganglia seems to be the most likely cause of facilitation of the plantar grasp reflex in these children, the excessive relaxation of toe extensors, in response to the toe flexion at the moment of elicitation of the reflex, may be another factor causing an exaggerated response. Children with athetosis generally exhibit decreased tonus of finger muscles, especially of flexors, which causes a weak grasp and the release of objects too easily [62]. The relative predominance of finger extensors over flexors, in addition to the specific weakness of the finger muscles, may be a factor causing the diminished palmar grasp reflex in these children.

ased tonus of finger muscles, especially of flexors, which causes a weak grasp and the release of objects too easily [62]. The relative predominance of finger extensors over flexors, in addition to the specific weakness of the finger muscles, may be a factor causing the diminished palmar grasp reflex in these children. In infants, the maturation of cortical connections overrides the generators of primitive reflexes in the spinal cord and brain stem with age and eventually leads to the disappearance of the primitive reflexes and the emergence of righting and equilibrium reactions [34, 63]. The disappearance and emergence of these reflex mechanisms were demonstrated to be chronologically related to the attainment of motor milestones in normal children [24]. In children with MR, however, the process of evolution of these reflex mechanisms is prolonged, in accordance with the retardation of maturation in brain function, resulting in retention of primitive reflexes and the delayed attainment of motor milestones [24, 29, 34, 64].

t of motor milestones in normal children [24]. In children with MR, however, the process of evolution of these reflex mechanisms is prolonged, in accordance with the retardation of maturation in brain function, resulting in retention of primitive reflexes and the delayed attainment of motor milestones [24, 29, 34, 64]. 3. The Moro Reflex 3.1. Elicitation In his original method, Moro [9] elicited the reflex by hitting the pillow on either side of an infant's head with the hands. Later a variety of methods for eliciting the reflex were devised including hitting on the table surface, warm or cold application to the chest or stomach, and a tap on the abdomen [10, 11]. Nowadays, however, the head drop method is the most common, because a slight drop of the infant's head relative to the body axis in the supine position is generally accepted as the most effective technique for eliciting the reflex [10, 11, 19]. For elicitation in this method, the infant is held suspended in a symmetrical supine position with one of the examiner's hands behind the chest and the other supporting the head, and the head being held in a midline position and then dropped back a few cm. It is important to ensure that both the subject's hands are open at the moment of elicitation of the reflex so as not to provoke an asymmetrical response [14]. Infants should be tested while they are awake, but not crying [18].

ting the head, and the head being held in a midline position and then dropped back a few cm. It is important to ensure that both the subject's hands are open at the moment of elicitation of the reflex so as not to provoke an asymmetrical response [14]. Infants should be tested while they are awake, but not crying [18]. The drop of the baby method is an alternative one for eliciting the reflex: the infant is suspended horizontally, as in the head drop method, and then the examiner lowers his or her hands rapidly about 10 to 20 cm and brings them to an abrupt halt. There is no dorsiflexion of the neck with this technique [18]. On the other hand, Lesný [65] reported that a nociceptive stimulus applied to the infant's skin and subcutaneous tissue of the epigastrium by pinching was effective for eliciting the reflex.

hands rapidly about 10 to 20 cm and brings them to an abrupt halt. There is no dorsiflexion of the neck with this technique [18]. On the other hand, Lesný [65] reported that a nociceptive stimulus applied to the infant's skin and subcutaneous tissue of the epigastrium by pinching was effective for eliciting the reflex. The initial phase of the response comprises abduction of the upper limbs at the shoulders and extension of the forearms at the elbows, with slight extension of the spine and retraction of the head. The forearms are supinated and the digits extended, except for the semiflexed index fingers and thumbs, forming the shape of a “C”. There is sometimes a slight tremor or clonus-like rhythmic movements of the limbs. Subsequently the arms adduct at the shoulders and the forearms flex at the elbows: the upper limbs describe an arc-like movement, bringing the hands in front of the body, which finally return to the original position [10, 11, 18]. The responses of the lower limbs are usually eliminated from the evaluation, because they show wide variability among the normal population [11, 19, 20]. With this reflex, habituation develops only on an experimental basis with intensively repetitive trials, that is, not in the clinical setting [17, 66]. No significant difference in the response has been reported at birth or through the first 5 months of age between the term cephalic-presenting and breech-presenting infant groups [67].

re noted to have associated extracranial cysticercosis. Thirty-nine children had “multiple lesion” NCC; 30 (76.9%) of them had raised ICT, and 4 (10.3%) had focal neurological deficits. All the 18 cases with intraventricular NCC had more than 5 NCC lesions with 10 cases having 5–10 lesions and 8 cases with >10 lesions. Serum Taenia solium antibodies were tested in 43 children and a positivity rate of 65.1% was observed (Table 3). CSF antibodies against T solium was done in 20 children only but it revealed a very high positivity (90%) (Table 4).

, habituation develops only on an experimental basis with intensively repetitive trials, that is, not in the clinical setting [17, 66]. No significant difference in the response has been reported at birth or through the first 5 months of age between the term cephalic-presenting and breech-presenting infant groups [67]. 3.2. Clinical Significance 3.2.1. Responses in Normal Infants The study of the Moro reflex in normal term infants has been undertaken by many authors. The results obtained with the head drop method are well consistent. The reflex can be elicited in all infants during the first 12 weeks of age. After the neonatal period, however, the response becomes increasingly less typical with age, eventually consisting only of abduction and extension of the upper limbs. Beyond 12 weeks of age, the proportion of infants exhibiting a negative response rapidly increases, reaching about 80% at 20 weeks of age [25]. The reflex usually disappears by 6 months of age [11, 31, 68–70]. Several authors compared the Moro reflex in preterm infants tested at 40 weeks PCA or at 4 months of corrected age with that in term infants. Although none of the studies confirmed the outcome in the subjects, there is a general agreement as to the similarity in the response between the two groups, especially when only infants with no or low perinatal risk factors are compared [1, 71–75].

0 weeks PCA or at 4 months of corrected age with that in term infants. Although none of the studies confirmed the outcome in the subjects, there is a general agreement as to the similarity in the response between the two groups, especially when only infants with no or low perinatal risk factors are compared [1, 71–75]. 3.2.2. Abnormal Responses Based on the findings in normal infants, the absence or diminution of the Moro reflex within 2 to 3 months of age and the persistence of the response beyond 6 months of age can be regarded as abnormal. The absence of the response during the neonatal period and early infancy is of especial clinical significance and may indicate a compromised condition or disorder including birth injury, severe birth asphyxia, intracranial hemorrhage, infection, brain malformation, general muscular weakness of any cause, and CP of the spastic type [10, 31, 69, 76, 77]. On the other hand, a hyperactive response of the reflex is a common feature of neonatal withdrawal from maternal drug abuse including volatile substances, heroin, and opioids [78–80]. An exaggerated response may also be detected in infants with a severe bilateral intrauterine disturbance such as hydranencephaly [31].

. On the other hand, a hyperactive response of the reflex is a common feature of neonatal withdrawal from maternal drug abuse including volatile substances, heroin, and opioids [78–80]. An exaggerated response may also be detected in infants with a severe bilateral intrauterine disturbance such as hydranencephaly [31]. Asymmetry of the response is usually a sign of local injury. Damage to a peripheral nerve or cervical cord or a fracture of the clavicle may inhibit the reflex on the affected side. However, it should be noted that Dubowitz [14] demonstrated an asymmetrical response in normal infants. Their responses appeared to be related to the clenching of one fist during the procedure, and he considered that this might be caused by inhibition of the response on one side due to contraction of the finger flexors. Because Reiners et al. [81] found, in a prospective study, that none of 22 infants with a clavicular fracture exhibited an asymmetrical Moro response, the diagnostic value of the reflex for the detection of such fractures should not be overestimated. Retention of the reflex is common in children with MR without motor disturbance including Down's syndrome and in children with CP of the athetoid type [10, 69]. It is also sometimes observed in children with a severe brain malformation or with CP of the spastic type [31, 69].

such fractures should not be overestimated. Retention of the reflex is common in children with MR without motor disturbance including Down's syndrome and in children with CP of the athetoid type [10, 69]. It is also sometimes observed in children with a severe brain malformation or with CP of the spastic type [31, 69]. 3.3. Neural Mechanism 3.3.1. The Reflex Center Katona [17] reported that the Moro reflex could be elicited in anencephalic newborns with a nervous system that had developed only to the rostral level of the pons. In a study involving analysis of the relationship between the morphological structures of the rudimentary brain and primitive reflexes in six anencephalic newborns, Hanabusa [39] found that the Moro reflex could be elicited only when the vestibular nuclei were preserved. This finding indicates that the reflex is principally mediated by the vestibular nuclei. Rönnqvist et al. [82] reported that the average latency of the Moro reflex in 15 term neonates in the quiet awaking state detected with an optoelectronic device was 117.0 ms on the right arm and 129.2 ms on the left. These latencies are much longer than those of the spinal reflexes, clearly indicating that the reflex is mediated in the brain stem, not at the level of the spinal cord [83, 84]. Thus, the center of the Moro reflex seems to be in the lower region of the pons to the medulla.

ms on the right arm and 129.2 ms on the left. These latencies are much longer than those of the spinal reflexes, clearly indicating that the reflex is mediated in the brain stem, not at the level of the spinal cord [83, 84]. Thus, the center of the Moro reflex seems to be in the lower region of the pons to the medulla. 3.3.2. Afferent and Efferent Pathways The origin of afferent pathways for the Moro reflex, whether it is primarily vestibular, proprioceptive, or exteroceptive, has been a main subject of discussion. The head drop, the most common way of eliciting the reflex, stimulates both the vestibular system and the proprioceptive receptors in the neck. Rönnqvist [13] investigated the reflex by tilting the table without extension of the infants' neck to eliminate the proprioceptive inputs from the cervical vertebrae and neck muscles. The response could be elicited in 225/250 trials (90%), and twenty-one of the 25 negative trials were made while the infants were sleeping or crying. Prechtl [12] also demonstrated that sudden raising or dropping of the infants, whose head, neck, and trunk were fixed in a plaster cast, could elicit the response. Bloomfield et al. [85] reported an infant with CHARGE syndrome who exhibited a persistent complete absence of the Moro reflex with preservation of other primitive reflexes. Moro's original method cannot yield a satisfactory response if the table is too stable to produce a change in position on striking and necessitates jolting movement of the table to elicit a response [10]. These findings support the view that this reflex is principally mediated by the vestibular system. In contrast to the grasp reflex, the Moro reflex has not been observed in a fetus, which is also in agreement with its vestibular origin, because fetuses are protected from acceleration or shaking in intrauterine life [68, 86].

findings support the view that this reflex is principally mediated by the vestibular system. In contrast to the grasp reflex, the Moro reflex has not been observed in a fetus, which is also in agreement with its vestibular origin, because fetuses are protected from acceleration or shaking in intrauterine life [68, 86]. On the other hand, Parmelee Jr. [11] found that vestibular stimulation was not sufficient for a good Moro response when neck movement was prevented. Prechtl [12] described an infant with bilateral absence of the inner ears who had exhibited a normal Moro response to a head drop, which was reported by Karlsson in an address to a study group in Oxford. These observations suggest that the proprioceptive inputs from the neck also contribute to elicitation of the reflex. There are direct and indirect, via the cervical cord, ascending pathways that originate in the proprioceptive receptors in the neck and connect with vestibular nuclei. These pathways originally send signals to the brain stem to regulate the neck righting [57]. The signals generated by the head drop may travel via these routes to reach and activate the reflex mechanism in the brain stem. The head drop method is most effective, because it produces a large number of ascending signals to the target through the two pathways and thereby induces a high level of neural excitation that acts on the reflex center.

head drop may travel via these routes to reach and activate the reflex mechanism in the brain stem. The head drop method is most effective, because it produces a large number of ascending signals to the target through the two pathways and thereby induces a high level of neural excitation that acts on the reflex center. Although pinching of an infant's epigastrium has been reported to be effective for eliciting the reflex [65], a nociceptive stimulus is generally ineffective [10, 11]. Besides the primary somatosensory afferents, there is a path taken by nociceptor axons that reaches the pontine reticular formation, which has close interconnections with vestibular nuclei [87–89]. The nociceptive signals may travel via this pathway toward the reflex center in the brain stem. However, the level of neural excitation generated by nociceptive stimulation appears to be usually low, and the response can be elicited only when it infrequently exceeds the threshold of the reflex.

stibular nuclei [87–89]. The nociceptive signals may travel via this pathway toward the reflex center in the brain stem. However, the level of neural excitation generated by nociceptive stimulation appears to be usually low, and the response can be elicited only when it infrequently exceeds the threshold of the reflex. The reflex center probably contains a number of interneurons, because of the relatively long latency. The routes of afferent pathways can be multiple, and the efferent pathways of the response seem to originate in the vestibulospinal and/or reticulospinal neurons, because the response can even be obtained in anencephalic newborns devoid of both corticospinal and rubrospinal neurons [17, 39]. Thus, the reflex movement is generated by the subcortical structures without cortical participation, which explains why focal cerebral injury does not cause distinct disturbance of the Moro reflex [31]. It is also noteworthy that no asymmetrical response is sometimes detected in neonates and young infants who later develop spastic hemiplegia. The primary motor cortex and nonprimary motor areas project a lot of neurons to different motor centers in the brain stem including vestibulospinal and reticulospinal neurons. The brain stem also receives inputs from the basal ganglia and cerebellum. The Moro reflex in infants disappears with age, due to the increased inhibition of these upper brain structures.

or areas project a lot of neurons to different motor centers in the brain stem including vestibulospinal and reticulospinal neurons. The brain stem also receives inputs from the basal ganglia and cerebellum. The Moro reflex in infants disappears with age, due to the increased inhibition of these upper brain structures. 3.3.3. The Moro Reflex and Startle Reaction Although there has been much confusion regarding the Moro response and the startle reaction in the past, most authors agree today that they are different entities [10–12, 17]. The startle reaction, the response to a sudden stimulus, is one of the defensive reactions and consists essentially of flexion movements. It differs considerably from the Moro response primarily characterized by extension. Detailed observations with video recording [90] or film [91] and an electrophysiological study with surface electrodes [12] demonstrated the differences in motor behavior between them. Katona [17] found that the startle reaction induced by an auditory stimulus showed clear habituation in premature infants, whereas the Moro reflex did not, and that the startle reaction could not be elicited in anencephalic newborns, while the Moro reflex was always elicited in these infants. Pucher et al. [68] investigated the Moro reflex using a tilt table with simultaneous monitoring of autonomic parameters including respiration, heart rate, and transcutaneous pO2 and pCO2 and concluded that the reflex was not the result of a startle reaction, because of no significant alteration in these parameters.

ts. Pucher et al. [68] investigated the Moro reflex using a tilt table with simultaneous monitoring of autonomic parameters including respiration, heart rate, and transcutaneous pO2 and pCO2 and concluded that the reflex was not the result of a startle reaction, because of no significant alteration in these parameters. 4. Phylogenetic Meaning and Hierarchy of Responses 4.1. Phylogenetic Meaning 4.1.1. The Grasp Reflex In 1891, Robinson [92] described that a newborn human infant was able to support its own weight when it was holding on to a horizontal rod. He tested more than 60 infants under one month old and found that they were able to hang by their hands for at least 10 seconds, and one infant succeeded in hanging for 2 minutes and 35 seconds. Richter [46] had the opportunity of testing the palmar grasp reflex in five monkey infants. They showed a grasp reflex that was much stronger than that of human newborn infants. The time of hanging ranged from 7 to 33 minutes with only one hand. Brain and Curran [33] examined the plantar grasp reflex in nine adult anthropoid apes and monkeys and found that none of the monkeys showed the grasp reflex and that the prehensile function of the adult monkey's feet was highly specialized, responding to a variety of stimuli. They also observed a number of infant monkeys, some within a few days of birth, slung beneath the mother's belly or clinging to her flanks only with their hands and feet, receiving no support from her. Although only a day or two old, they held on while she jumped from bough to bough and did not leave her until they were weaned.

observed a number of infant monkeys, some within a few days of birth, slung beneath the mother's belly or clinging to her flanks only with their hands and feet, receiving no support from her. Although only a day or two old, they held on while she jumped from bough to bough and did not leave her until they were weaned. Based on these findings, the grasp reflex of the hands and feet in human infants could be regarded as a rudiment of phylogenetic functions that were once essential for monkey infants in arboreal life and that have lost their usefulness in the human species [28, 33, 62]. As McGraw [93] stated, modern infants, as well as their fairly recent human antecedents, do not need to hang on with their hands and feet from the moment of birth.

f phylogenetic functions that were once essential for monkey infants in arboreal life and that have lost their usefulness in the human species [28, 33, 62]. As McGraw [93] stated, modern infants, as well as their fairly recent human antecedents, do not need to hang on with their hands and feet from the moment of birth. 4.1.2. The Moro Reflex In his original paper, Moro [9] emphasized the clasping aspect of the reflex and claimed that the reflex is a primitive movement analogous to that in young apes and bats, by which they instinctively embrace or cling to their mothers. However, the question has been raised as to his biological interpretation by several authors. Parmelee Jr. [11] stated that Moro's emphasis in his belief that the reflex is an atavistic lifesaving reflex resulted in some of the present-day confusion. Mitchell [10] described that in the Moro response, the essential component is the extension movement, and thus it should not be called an “embrace” or “clasp” reflex. On the other hand, Amiel-Tison and Grenier [94] commented that the Moro reflex can be seen as a hindrance to voluntary motor activity, and this parasitical movement is a nuisance that the newborn retains until a subsequent stage of maturation when inhibitory brain function begins. The unique movement behavior of the Moro reflex is thus difficult to understand, and its phylogenetic meaning remains unclear.

en as a hindrance to voluntary motor activity, and this parasitical movement is a nuisance that the newborn retains until a subsequent stage of maturation when inhibitory brain function begins. The unique movement behavior of the Moro reflex is thus difficult to understand, and its phylogenetic meaning remains unclear. 4.2. Hierarchy of the Reflex Responses Several authors observed in human newborns that the palmar grasp reflex inhibits the Moro reflex [12, 15, 16, 22, 91]. Later, Katona [17] found in newborn apes and monkeys that the Moro reflex was regularly observed and it was inhibited on elicitation of the palmar grasp reflex. Because this interrelation between the grasp reflex and the Moro reflex would be essential for newborn animals to keep clinging and to prevent a fall, the interrelation observed in human newborns could also be regarded as a phylogenetic rudiment. Katona [17] noted another fact that a sudden auditory stimulus triggered a strong reinforcement of the grasp reflex in such young animals, which also supports the view that the startle reaction is one of the defensive reactions. Pollack [22] found, in term human newborns, that sucking increased the palmar grasp reflex, whereas head rotation had no influence on this grasp reflex. Lippmann noted in human newborns that the Babkin reflex consisting of head flexion and mouth opening in response to pressure on the palms of both hands was not obtained during sucking [95]. Brown and Fredrickson [15] also observed in human newborns that the palmar grasp reflex did not affect the sucking reflex but that the sucking reflex increased the strength and the duration of the palmar grasp reflex. If the same interrelation between the sucking reflex and the palmar grasp reflex exists in monkey infants, it would be very helpful for them to firmly cling to their mother while feeding. The asymmetrical tonic neck reflex (ATNR) elicited by head rotation usually induces extension of the upper and lower limbs on the face side in infants. If the palmar grasp reflex is predominant over the ATNR in monkey infants, as demonstrated in human infants, head rotation would not bring any reduction in the response of the grasp reflex, and the animal infant would keep clinging to its mother. Since the Babkin reflex is regarded as the rudiment of hand-mouth coordination for preying in animals [96, 97], it appears to be rational that the reflex is unable to be elicited during feeding.

n would not bring any reduction in the response of the grasp reflex, and the animal infant would keep clinging to its mother. Since the Babkin reflex is regarded as the rudiment of hand-mouth coordination for preying in animals [96, 97], it appears to be rational that the reflex is unable to be elicited during feeding. Thus, the hierarchical interrelations among the primitive reflexes observed in human newborns seem to be indispensable for monkey newborns for their essential behavior such as feeding, moving, and preventing a fall, although the interrelations have not been fully elucidated in monkey infants. Based on the findings already mentioned in this paper, the Moro reflex in the young monkey would be elicited when the vestibular system is stimulated with abrupt tilting of the body or head while it is being passively held by its mother without active clinging caused by the palmar grasp reflex. In this situation, the mother would notice her baby was off balance from its exaggerated reflex movement, and she would immediately try to seize the baby to prevent a fall. It might be possible to assume that the Moro reflex in monkey neonates plays a role in such interaction between mother and child for protection against a fall. To clarify the meaning of the Moro reflex, it appears necessary to determine in monkeys in what situation the reflex is elicited and how the response works in the mother and child.

to assume that the Moro reflex in monkey neonates plays a role in such interaction between mother and child for protection against a fall. To clarify the meaning of the Moro reflex, it appears necessary to determine in monkeys in what situation the reflex is elicited and how the response works in the mother and child. 5. Conclusions The palmar grasp reflex in infants has diagnostic significance. The absence or a weak response of the reflex during early infancy may reflect peripheral nerve or spinal cord involvement or may predict the development of athetoid type CP, whereas the response may be hyperactive in children with spasticity in their upper limbs. The plantar grasp reflex is also of high clinical significance, especially in terms of the detection of spasticity. No reflex, or a diminished one, during early infancy is often a sensitive predictor of the development of spastic CP. The grasp reflex of the hands and feet is mediated by the spinal reflex mechanism, which, however, appears to be under regulatory control of nonprimary motor areas through the spinal interneurons. The absence of the Moro reflex during the neonatal period and early infancy is highly diagnostic, indicating a variety of compromised conditions. The center of the reflex is probably in the lower region of the pons to the medulla. The grasp reflex of the hands and feet in human infants could be regarded as a rudiment of phylogenetic function, whereas the phylogenetic meaning of the Moro reflex remains unclear. The hierarchical interrelations among the primitive reflexes seem to be essential for monkey newborns for their arboreal life, although it has not been fully elucidated. The possible role of the Moro reflex in these newborns was discussed in relation to the interrelations.

1. Introduction Kawasaki disease (KD) is an acute febrile vasculitis that predominantly affects children ≤5 year of age [1, 2]. Children with KD typically have an acute onset of fever followed by signs of mucosal inflammation and vasodilatation that evolve over the first week of the illness. Laboratory tests reveal a marked systemic inflammatory response [3]. KD is the leading cause of acquired heart disease in children in most developed countries, including Japan and the USA [4, 5]. KD predominantly causes vasculitis in medium-sized arteries, with a striking predilection for the coronary arteries [2, 4]. The earliest pathological change in the vessel wall is edema of endothelial and smooth muscle cells with intense inflammatory infiltration of the vascular wall, initially by polymorphonuclear cells and rapidly thereafter by macrophages, lymphocytes (primarily CD8+ T cells), and plasma cells [4].

ronary arteries [2, 4]. The earliest pathological change in the vessel wall is edema of endothelial and smooth muscle cells with intense inflammatory infiltration of the vascular wall, initially by polymorphonuclear cells and rapidly thereafter by macrophages, lymphocytes (primarily CD8+ T cells), and plasma cells [4]. The cause of KD remains unknown, but it is thought that the immune system is activated by an infectious trigger in genetically susceptible hosts [5]. Although there is considerable controversy regarding the mechanism of immune system activation, recent data suggested that T-cell activation is important in determining the susceptibility and severity of KD [5]. Furthermore, susceptibility genes for KD have been identified by successive studies using a genome-wide approach. These include the inositol 1,4,5-triphosphate kinase-C (ITPKC), caspase-3 (CASP3), CD40 ligand (CD40L), FAM167A-BLK, the Fc fragment of IgG, low affinity IIa, receptor (FCGR2A), and the human leukocyte antigen (HLA) genes [6]. Because KD is a systemic vasculitis, multiple organ involvement can develop, including coronary artery lesions (CALs), carditis, arthritis, hepatitis, central nervous system (CNS) disease [3, 7], KD shock syndrome (KDSS) [8], muscle involvement [7, 9], hyponatremia [10–12], and kidney and urinary tract involvement.

The cause of KD remains unknown, but it is thought that the immune system is activated by an infectious trigger in genetically susceptible hosts [5]. Although there is considerable controversy regarding the mechanism of immune system activation, recent data suggested that T-cell activation is important in determining the susceptibility and severity of KD [5]. Furthermore, susceptibility genes for KD have been identified by successive studies using a genome-wide approach. These include the inositol 1,4,5-triphosphate kinase-C (ITPKC), caspase-3 (CASP3), CD40 ligand (CD40L), FAM167A-BLK, the Fc fragment of IgG, low affinity IIa, receptor (FCGR2A), and the human leukocyte antigen (HLA) genes [6]. Because KD is a systemic vasculitis, multiple organ involvement can develop, including coronary artery lesions (CALs), carditis, arthritis, hepatitis, central nervous system (CNS) disease [3, 7], KD shock syndrome (KDSS) [8], muscle involvement [7, 9], hyponatremia [10–12], and kidney and urinary tract involvement. With the exception of sterile pyuria and trace proteinuria, kidney and urinary tract involvement in KD is uncommon [1, 3, 13]. However, as shown in the paper, such complications are increasingly being reported, including pyuria, prerenal acute kidney injury (AKI), renal AKI caused by tubulointerstitial nephritis (TIN), hemolytic uremic syndrome (HUS), immune-complex mediated nephropathy, renal AKI associated with KDSS or unknown causes, acute nephritic syndrome (ANS), nephrotic syndrome (NS), renal tubular abnormalities, renal abnormalities in imaging studies, and renal artery lesions. Because renal complications can contribute to an acute or chronic deterioration in the condition of patients, it is important to clarify the clinical features and pathogenesis of renal involvement in patients with KD. The following is a review of kidney and urinary tract involvement in KD.

es, and renal artery lesions. Because renal complications can contribute to an acute or chronic deterioration in the condition of patients, it is important to clarify the clinical features and pathogenesis of renal involvement in patients with KD. The following is a review of kidney and urinary tract involvement in KD. 2. Pyuria Patients with KD often present with abnormal urinary findings, including proteinuria, hematuria, and pyuria [1, 12, 13]. Of these, pyuria is the most common abnormal finding [12, 14, 15], with some patients with KD and pyuria being misdiagnosed as having acute pyelonephritis [16, 17]. In a previous study, we showed 43.5% of patients with KD had sterile pyuria, a proportion similar to that reported by Melish et al. [13], Wirojanan et al. [18], Barone et al. [19], and Shike et al. [20] of 62.5%, 33%, 50%, and 74%, respectively. On the other hand, Turner and Coulthard [21] reported that 43% of febrile children without urinary tract infection had moderate pyuria (10–100 cells/high-power field) with only 9% of febrile children having obvious pyuria (>100 cells/high-power field). This finding indicated that pyuria may be a nonspecific feature of fever in acute childhood illness. However, because one-third of patients with KD exhibited obvious pyuria in our study [14], we considered pyuria was a specific feature of the disease.

ebrile children having obvious pyuria (>100 cells/high-power field). This finding indicated that pyuria may be a nonspecific feature of fever in acute childhood illness. However, because one-third of patients with KD exhibited obvious pyuria in our study [14], we considered pyuria was a specific feature of the disease. Sterile pyuria in KD is associated with mononuclear cells in the urine [3]. Ohta et al. [22] demonstrated that the majority of cells in urine specimens of KD patients with pyuria were mononuclear cells, but not neutrophils. Kobayashi et al. [23] also showed that mononuclear cells with intracytoplasmic inclusion were present in the urinary sediments of a patient with KD.

Acknowledgments The author would like to thank Dale Rice, MD, Professor, University of Southern California, Department of Otolaryngology-Head and Neck Surgery, and Dennis Crockett, MD, of Head/Neck Associates of Orange County, for their help in acquiring patient data; H. Wolfgang Beumer, MD, for his invaluable help in translating the German language articles; and Stephen Perlman from the Durham VA Library for his help in acquiring articles from other institutions. None of them received any compensation for their contributions, but their help was greatly appreciated. Conflict of Interests The author declares that there is no conflict of interests regarding the publication of this paper. Figure 1 Length of involved skin relative to patient age. Figure 2 Consistency of presentation between a patient from 1848 (a) and the modern day (b). Table 1 Catalogue of the published world literature by author and year.

cells in the urine [3]. Ohta et al. [22] demonstrated that the majority of cells in urine specimens of KD patients with pyuria were mononuclear cells, but not neutrophils. Kobayashi et al. [23] also showed that mononuclear cells with intracytoplasmic inclusion were present in the urinary sediments of a patient with KD. Sterile pyuria in KD is thought to be due to urethritis caused by a non-specific vasculitis of the urethra [3]. Melish et al. [13] reported four KD patients with sterile pyuria in voided urine samples who did not have leukocytes in bladder urine obtained by bladder taps. This suggested that urethral inflammation was the source of the urinary leukocytes. However, we reported that 5 of 10 KD patients with sterile pyuria in voided urine samples also had leukocytes in bladder urine, with higher levels of urinary protein and β2-microglobulin and serum blood urea nitrogen and creatinine concentrations than patients without leukocytes in their bladder samples or patients without pyuria in voided urine [14]. These results suggest that some patients with KD develop sterile pyuria that originates from the urethra and/or the kidney as a result of mild and subclinical renal injury [14]. In addition, we recently reported a case of acute cystitis in a patient with KD, which suggested sterile pyuria in KD may originate from the bladder due to cystitis [24].

ome patients with KD develop sterile pyuria that originates from the urethra and/or the kidney as a result of mild and subclinical renal injury [14]. In addition, we recently reported a case of acute cystitis in a patient with KD, which suggested sterile pyuria in KD may originate from the bladder due to cystitis [24]. Pyuria is not always sterile in patients with KD. Shiono et al. [25] reported a KD patient with a left vesicoureteral reflux associated with pyuria and pyelonephritis due to Escherichia coli. Wu et al. [26] reported that 8 of 75 patients with KD had bacterial pyuria, while Benseler et al. [27] reported that 42 of 129 (33%) patients with KD had ≥1 confirmed bacterial or viral infection at diagnosis, with 4 (3%) of these patients having a urinary tract infection (UTI). No difference in clinical phenotype or coronary artery outcome was observed in patients with a UTI compared with those without a UTI [26, 27]. It remains unclear whether bacterial UTI induces an immune response in patients with KD or whether this occurs only when the patients have a coexisting infection [26]. However, because the immune system is thought to be activated by an infectious trigger in genetically susceptible hosts with KD [5], a UTI may induce clinical features typical of KD. 3. Acute Kidney Injury AKI is the new consensus term for acute renal failure (ARF) and has replaced ARF in order to emphasize that a continuum of kidney injury exists that begins long before sufficient loss of excretory kidney function can be measured by standard laboratory tests [28].

Pyuria is not always sterile in patients with KD. Shiono et al. [25] reported a KD patient with a left vesicoureteral reflux associated with pyuria and pyelonephritis due to Escherichia coli. Wu et al. [26] reported that 8 of 75 patients with KD had bacterial pyuria, while Benseler et al. [27] reported that 42 of 129 (33%) patients with KD had ≥1 confirmed bacterial or viral infection at diagnosis, with 4 (3%) of these patients having a urinary tract infection (UTI). No difference in clinical phenotype or coronary artery outcome was observed in patients with a UTI compared with those without a UTI [26, 27]. It remains unclear whether bacterial UTI induces an immune response in patients with KD or whether this occurs only when the patients have a coexisting infection [26]. However, because the immune system is thought to be activated by an infectious trigger in genetically susceptible hosts with KD [5], a UTI may induce clinical features typical of KD. 3. Acute Kidney Injury AKI is the new consensus term for acute renal failure (ARF) and has replaced ARF in order to emphasize that a continuum of kidney injury exists that begins long before sufficient loss of excretory kidney function can be measured by standard laboratory tests [28]. AKI is usually divided into three broad pathophysiologic categories according to the cause: prerenal AKI, characterized by decreased kidney perfusion in which there is no parenchymal damage to the kidney, renal AKI, which results from the renal parenchyma injuries, or postrenal AKI, caused by acute obstruction of the urinary tract [28].

ided into three broad pathophysiologic categories according to the cause: prerenal AKI, characterized by decreased kidney perfusion in which there is no parenchymal damage to the kidney, renal AKI, which results from the renal parenchyma injuries, or postrenal AKI, caused by acute obstruction of the urinary tract [28]. In patients with KD, both prerenal and renal AKI have been reported. While TIN, HUS, immune-complex mediated nephropathy, and KDSS have been reported as causes of renal AKI, the cause of renal injury has not been established in other patients. 3.1. Prerenal Acute Kidney Injury Prerenal AKI has been reported in three patients with KD. Nakanishi et al. [29] reported a 12-year-old boy with KD who developed prerenal AKI and acute heart failure (AHF). A renal biopsy demonstrated no abnormal findings in the glomeruli or tubulointerstitium. Senzaki et al. [30] reported an 8-year-old boy with KD who developed prerenal AKI and AHF, while Adachi [31] described a 2-year-old girl with KD who developed prerenal AKI caused by hypovolemia due to gastrointestinal losses and AHF. All three patients had AHF concomitant with AKI with their renal function recovering completely following improvement of AHF. Because AHF causes renal hypoperfusion due to decreased cardiac output [32], AHF may play a central role in the development of prerenal AKI in patients with KD.

gastrointestinal losses and AHF. All three patients had AHF concomitant with AKI with their renal function recovering completely following improvement of AHF. Because AHF causes renal hypoperfusion due to decreased cardiac output [32], AHF may play a central role in the development of prerenal AKI in patients with KD. 3.2. Renal Acute Kidney Injury 3.2.1. Tubulointerstitial Nephritis TIN is a frequent cause of AKI and is characterized by the presence of an inflammatory cell infiltrate in the kidney interstitium [33]. Some cases of TIN are secondary to diseases with immune-mediated mechanisms such as systemic lupus erythematosus, Sjögren's disease, or vasculitis [33]. The presence of T-helper and T-suppressor lymphocytes in the cellular infiltrates suggests that these cells have a pathogenic role in TIN [33].

dney interstitium [33]. Some cases of TIN are secondary to diseases with immune-mediated mechanisms such as systemic lupus erythematosus, Sjögren's disease, or vasculitis [33]. The presence of T-helper and T-suppressor lymphocytes in the cellular infiltrates suggests that these cells have a pathogenic role in TIN [33]. AKI due to TIN has been reported in five patients with KD. Veiga et al. [34] reported a 2-year-old boy with KD who developed AKI due to TIN. A renal biopsy revealed diffuse interstitial infiltration of mononuclear and polymorphonuclear leukocytes. The patient recovered with supportive care alone. Kawamura [35] reported a 9-year-old girl with KD who developed AKI with a renal biopsy showing mild interstitial mononuclear infiltration. The patient underwent hemodialysis and recovered without renal sequelae but with CALs. Ashida et al. [36] reported a 5-year-old boy with KD who developed AKI due to TIN. The patient recovered with supportive care alone. Bonany et al. reported [37] an 8-year-old boy with KD who developed AKI and nephrotic syndrome, in whom renal biopsy showed TIN with mild mesangial expansion and no vessel involvement. The patient recovered with supportive care alone. Papadodima et al. [38] examined the kidney autopsy findings of an 11-year-old boy who died of coronary arterial thrombosis associated with KD and found that the patient had TIN with mild expansion of the mesangial matrix.

d mesangial expansion and no vessel involvement. The patient recovered with supportive care alone. Papadodima et al. [38] examined the kidney autopsy findings of an 11-year-old boy who died of coronary arterial thrombosis associated with KD and found that the patient had TIN with mild expansion of the mesangial matrix. Further evidence was reported by Ogawa [39] who examined kidney specimens from 25 patients who died from KD and showed that these patients frequently had interstitial leukocytes infiltrations (36%), especially in patients with duration of illness <50 days (100%). Asaji et al. also reported that 25% of patients who died from KD had interstitial infiltration of leukocytes [40]. These findings suggest that TIN may occur more commonly in KD patients. Because T cell activation contributes to the development of TIN [33] and KD [5], it may also cause TIN in patients with KD.

s (100%). Asaji et al. also reported that 25% of patients who died from KD had interstitial infiltration of leukocytes [40]. These findings suggest that TIN may occur more commonly in KD patients. Because T cell activation contributes to the development of TIN [33] and KD [5], it may also cause TIN in patients with KD. However, it is necessary to rule out Y. pseudotuberculosis infection in patients with KD and AKI caused by TIN [41]. Y. pseudotuberculosis is a gram-negative coccobacillus that causes diarrhea, abdominal pain, fever, skin rash, conjunctivitis, erythema nodosum, and lymphadenopathy [42, 43]. Typical KD-like findings developed in 8.8%–10.8% of patients with Y. pseudotuberculosis infections [43, 44], while, AKI occurred in 9.6%–13.6% of these patients [43, 45]. In addition, the renal histology of nearly all patients with Y. pseudotuberculosis infection who underwent renal biopsy showed TIN [43, 45–47]. Therefore, AKI is more likely to develop in patients with Y. pseudotuberculosis infection, which mimics KD, than in patients with true KD. 3.2.2. Hemolytic Uremic Syndrome HUS is a disease characterized by nonimmune hemolytic anemia, thrombocytopenia, and renal impairment [48, 49]. Microvascular injury with endothelial cell damage is a pathological characteristic of all forms of HUS [49]. The various etiologies of HUS allow classification into infection-induced, genetic, medication-induced, and HUS associated with systemic diseases characterized by microvascular injury [49].

impairment [48, 49]. Microvascular injury with endothelial cell damage is a pathological characteristic of all forms of HUS [49]. The various etiologies of HUS allow classification into infection-induced, genetic, medication-induced, and HUS associated with systemic diseases characterized by microvascular injury [49]. HUS has been reported in only two patients with KD and AKI [50, 51]. Ferriero and Wolfsdorf [50] reported a 2-year-old girl with KD who had clinical and laboratory features of mild HUS, who recovered with supportive care alone. Heldrich et al. [51] reported a 3-year-old girl with KD who developed HUS and Henoch-Schönlein purpura, which required adjustments in therapy. Although the pathogenesis underlying the development of HUS in both patients was unclear because they did not undergo renal biopsy, it is possible that the vasculitis associated with KD may have involved the kidney producing endothelial injuries in the renal and glomerular endothelium, leading to a clinical picture of HUS [50]. 3.2.3. Immune-Complex Mediated Nephropathy Nagamatsu et al. [52] reported renal histologic findings of a 3-year-old boy with KD who developed renal AKI. Although light microscopy showed that nearly all the glomeruli were normal, electron microscopy revealed electron dense deposits in the subepithelial spaces and in the podocytes. This suggested the possibility of glomerular derangement by immune complex in this patient.

a 3-year-old boy with KD who developed renal AKI. Although light microscopy showed that nearly all the glomeruli were normal, electron microscopy revealed electron dense deposits in the subepithelial spaces and in the podocytes. This suggested the possibility of glomerular derangement by immune complex in this patient. 3.2.4. Renal Acute Kidney Injury Associated with Kawasaki Disease Shock Syndrome KDSS has recently been proposed as severe form of KD characterized by systolic hypotension or clinical signs of poor perfusion [8, 53]. KDSS is associated with more severe changes in laboratory inflammatory markers and greater risk of coronary artery abnormalities, mitral regurgitation, and prolonged myocardial dysfunction [8]. Gatterre et al. [54] studied 11 patients with KDSS and reported that 10 of these patients developed AKI, with 8 of these patients having multiple organ dysfunction syndrome (MODS). All 10 patients recovered from AKI without any renal sequels. Mac Ardle et al. [55] reported a 2-year-old boy with clinical features of KDSS who developed AKI that required peritoneal dialysis. A renal biopsy showed normal glomeruli and a patchy immune-type infiltrate that contained plasma cells and eosinophils, with evidence of recovering acute tubular necrosis (ATN). These findings suggest that AKI in patients with KDSS results from ATN due to MODS.

KDSS who developed AKI that required peritoneal dialysis. A renal biopsy showed normal glomeruli and a patchy immune-type infiltrate that contained plasma cells and eosinophils, with evidence of recovering acute tubular necrosis (ATN). These findings suggest that AKI in patients with KDSS results from ATN due to MODS. 3.2.5. Renal Acute Kidney Injury of Unknown Causes Five patients with AKI of unknown causes have been reported. Rhodes et al. [56] reported an 11-year-old girl with KD who developed renal AKI and AHF due to myocarditis. She underwent mechanical ventilation, intravenous administrations of diuretics, and an intravenous administration of immunoglobulin (IVIG). The patient became well and recovered without any sequels. The authors suggested that AKI in the patient was due to nephritis. Yamawaki et al. [57] reported a 5-year-old boy with KD who developed renal AKI and underwent hemodialysis. The patient recovered completely without any renal sequels. Lande et al. [58] reported a 3-year-old girl with KD who developed renal AKI and underwent hemodialysis. An abdominal ultrasound revealed enlarged and echogenic kidneys. This patient also recovered without any renal sequels. El Karoui et al. [59] reported a 45-year-old man with KD who developed renal AKI, in whom IVIG resulted in a rapid improvement and recovery of normal renal function without any sequels. Nandi and Mondal [60] reported a 4-year-old boy with KD who developed renal AKI and recovered with supportive measures alone without requiring renal replacement therapy.

-year-old man with KD who developed renal AKI, in whom IVIG resulted in a rapid improvement and recovery of normal renal function without any sequels. Nandi and Mondal [60] reported a 4-year-old boy with KD who developed renal AKI and recovered with supportive measures alone without requiring renal replacement therapy. 4. Acute Nephritis Syndrome ANS is the clinical picture of several glomerulonephropathies, characterized typically by hematuria, edema, hypertension, moderate proteinuria, and renal insufficiency [61, 62]. The ANS results from a reduction in the glomerular filtration rate caused by an inflammatory reaction in the glomeruli and is most often seen in patients with acute postinfectious glomerulonephritis [61]. Some forms of vasculitis also present with ANS, including Henoch-Schönlein purpura, Wegener's granulomatosis, and microscopic polyangiitis [62].

ion in the glomerular filtration rate caused by an inflammatory reaction in the glomeruli and is most often seen in patients with acute postinfectious glomerulonephritis [61]. Some forms of vasculitis also present with ANS, including Henoch-Schönlein purpura, Wegener's granulomatosis, and microscopic polyangiitis [62]. Seven cases of ANS have been reported in patients with KD [63–69]. Six patients were infants (median age, 5 months; range, 2–72 months). ANS developed between 2 and 30 days (median 20 days) following the onset of Kawasaki disease. Hematuria (macroscopic hematuria), proteinuria, edema, and hypertension occurred in 7/7 (5/7), 6/7, 5/7, and 4/7 of the patients, respectively. Decreased levels of both serums C3 and C4 were present in 4/7 of the patients. Mild renal insufficiency developed in two patients, which promptly resolved following the recovery of KD. The abnormal renal findings in all patients disappeared completely, while six patients developed CALs. A renal biopsy was only performed in a 3-year-old patient reported by Salcedo et al. [63] which revealed mesangioproliferative glomerulonephritis with interstitial infiltrates of lymphocytes and plasma cells by light microscopy, coarse granular deposition of IgM and C3 in the mesangium by immunofluorescent studies, and electron-dense deposits in the mesangium and in subendothelial space by electron microscopy.

ch revealed mesangioproliferative glomerulonephritis with interstitial infiltrates of lymphocytes and plasma cells by light microscopy, coarse granular deposition of IgM and C3 in the mesangium by immunofluorescent studies, and electron-dense deposits in the mesangium and in subendothelial space by electron microscopy. The pathogenesis underlying the development of ANS in KD remains unclear. Immune-complex mediated mechanisms were suspected in some patients, because the renal biopsy findings of the patient reported by Salcedo et al. showed electron-dense deposits in the glomeruli, with 4 of the 7 patients having decreased levels of both serums C3 and C4, which suggested activation of the classical complement pathway.

lex mediated mechanisms were suspected in some patients, because the renal biopsy findings of the patient reported by Salcedo et al. showed electron-dense deposits in the glomeruli, with 4 of the 7 patients having decreased levels of both serums C3 and C4, which suggested activation of the classical complement pathway. 5. Nephrotic Syndrome Although proteinuria is common in patients with KD, NS has been reported only rarely. Ogino et al. [70] reported a 3-year-old boy with KD who developed NS and azotemia, which resolved following the recovery of KD. Kidney and coronary angiographies showed no abnormal findings. Lee et al. [71] reported a 3-month-old boy with KD who developed NS. Although administration of corticosteroids improved NS in this patient, he died from a massive myocardial infarction on the 34th day of illness. Krug et al. [72] reported three children with KD, aged 4, 4.5, and 8 years, respectively, who developed NS. All three patients were treated with IVIG and aspirin, but none received corticosteroids. Proteinuria in all three patients disappeared within 2 weeks. Gatterre et al. [54] studied 11 patients with KDSS and reported that 3 of the 11 patients developed NS. Despite the severity of symptoms, all patients recovered from NS and survived without any sequels.

th IVIG and aspirin, but none received corticosteroids. Proteinuria in all three patients disappeared within 2 weeks. Gatterre et al. [54] studied 11 patients with KDSS and reported that 3 of the 11 patients developed NS. Despite the severity of symptoms, all patients recovered from NS and survived without any sequels. Because no renal biopsies were performed, the pathogenic mechanism underlying the development of NS in KD remains unclear. However, immune-complex mediated kidney injuries and/or T-cell immune-regulatory abnormalities similar to those seen in minimal change NS have been postulated as possible mechanisms of NS in KD [71]. Joh et al. [73] reported a 4-month-old girl who developed NS 10 weeks following the onset of KD. The early renal histological lesion was similar to the Finnish-type congenital nephrotic syndrome at biopsy, but within 4 months the patient revealed diffuse mesangial sclerosis at autopsy. Because NS of the patient developed 10 weeks after the onset of KD, it seems likely that infantile NS in this patient may have been caused by abnormalities in genes not directly associated with KD such as NPHS1, NPHS2, WT1, or LAMB2 [74]. 6. Renal Tubular Abnormalities Renal tubular abnormalities have sometimes been reported in patients with KD. Kondo et al. [75] studied urinary lysozyme and β2-microglobulin (β2MG) levels, parameters of renal tubular damage, in 16 patients with KD and reported that some patients with KD had elevated urinary lysozyme and β2MG levels during the acute phase.

tubular abnormalities have sometimes been reported in patients with KD. Kondo et al. [75] studied urinary lysozyme and β2-microglobulin (β2MG) levels, parameters of renal tubular damage, in 16 patients with KD and reported that some patients with KD had elevated urinary lysozyme and β2MG levels during the acute phase. Asami et al. [76] examined urinary N-acetyl-β-D-glucosaminidase (NAG) levels, known as a sensitive index of renal tubular disorder, of six patients with KD and reported that all patients had elevated urinary NAG levels during the acute phase of the disease. Ohta et al. [22] measured urinary interleukin-6 (IL-6), β2MG, and NAG levels of 20 patients with KD and showed that urinary IL-6, β2MG, and NAG levels were elevated in the majority of patients during the acute phase, indicating the presence of specific renal parenchymal lesions. Igarashi et al. [77] examined urinary β2MG and NAG levels and serum β2MG concentrations in 12 patients with KD and showed that urinary β2MG and NAG levels were elevated in 10 of these patients during the acute phase, while no patients had elevated serum β2MG concentrations.

Ohta et al. [22] measured urinary interleukin-6 (IL-6), β2MG, and NAG levels of 20 patients with KD and showed that urinary IL-6, β2MG, and NAG levels were elevated in the majority of patients during the acute phase, indicating the presence of specific renal parenchymal lesions. Igarashi et al. [77] examined urinary β2MG and NAG levels and serum β2MG concentrations in 12 patients with KD and showed that urinary β2MG and NAG levels were elevated in 10 of these patients during the acute phase, while no patients had elevated serum β2MG concentrations. We studied urinary β2MG levels in 23 patients with KD [14]. These patients were divided into three groups according to the results of urinalysis: patients without pyuria, patients with pyuria in both voided urine and bladder urine obtained by transurethral catheterization (bladder pyuria), and patients with pyuria only in voided urine (urethral pyuria). We showed that the majority of patients with KD had elevated urinary β2MG levels, while patients with bladder pyuria had higher urinary β2MG levels than patients with urethral pyuria or patients without pyuria. These findings suggest that patients with KD have renal tubular abnormalities, especially patients with renal inflammation. The pathogenesis of renal tubular abnormalities in KD remains unclear. Because KD patients with elevated urinary β2MG and NAG levels also have increased urinary IL-6 concentrations [22], the inflammatory process within the renal parenchyma, such as TIN may cause renal tubular abnormalities in these patients.

We studied urinary β2MG levels in 23 patients with KD [14]. These patients were divided into three groups according to the results of urinalysis: patients without pyuria, patients with pyuria in both voided urine and bladder urine obtained by transurethral catheterization (bladder pyuria), and patients with pyuria only in voided urine (urethral pyuria). We showed that the majority of patients with KD had elevated urinary β2MG levels, while patients with bladder pyuria had higher urinary β2MG levels than patients with urethral pyuria or patients without pyuria. These findings suggest that patients with KD have renal tubular abnormalities, especially patients with renal inflammation. The pathogenesis of renal tubular abnormalities in KD remains unclear. Because KD patients with elevated urinary β2MG and NAG levels also have increased urinary IL-6 concentrations [22], the inflammatory process within the renal parenchyma, such as TIN may cause renal tubular abnormalities in these patients. 7. Renal Abnormalities in Imaging Studies Renal abnormalities in imaging studies using renal ultrasonography (US) or technetium-99m dimercaptosuccinic acid scintigraphy single photon emission computed tomography (DMSA renal SPECT) have been reported in patients with KD.

The pathogenesis of renal tubular abnormalities in KD remains unclear. Because KD patients with elevated urinary β2MG and NAG levels also have increased urinary IL-6 concentrations [22], the inflammatory process within the renal parenchyma, such as TIN may cause renal tubular abnormalities in these patients. 7. Renal Abnormalities in Imaging Studies Renal abnormalities in imaging studies using renal ultrasonography (US) or technetium-99m dimercaptosuccinic acid scintigraphy single photon emission computed tomography (DMSA renal SPECT) have been reported in patients with KD. Nardi et al. [78] examined seven patients with KD using renal US and reported that four patients with AKI had renal sonographic findings of increased cortical echogenicity, enlarged kidneys, and enhanced corticomedullary differentiation. They suggested that the enlargement of the kidneys may have been caused by a vasculitis involving the kidneys with the resulting fibrinoid deposits and cellular infiltrations leading subsequently to ischemia followed by edema.

eased cortical echogenicity, enlarged kidneys, and enhanced corticomedullary differentiation. They suggested that the enlargement of the kidneys may have been caused by a vasculitis involving the kidneys with the resulting fibrinoid deposits and cellular infiltrations leading subsequently to ischemia followed by edema. Huang et al. [79] measured the kidney lengths of 20 patients with KD and with normal renal function, 20 healthy children, and 15 febrile children using renal US. They also measured the levels of plasma hepatocyte growth factor (HGF) and transforming growth factor-β1 (TGF-β1) in all the children. The study showed that 14 (70%) of the patients with KD during the acute phase had absolute nephromegaly, with kidney lengths in these patients being significantly longer than those of healthy children or febrile children. In addition, the ratio of plasma HGF/TGF-β1 during the acute phase and after the recovery phase correlated positively with the degree of nephromegaly in all patients with KD. HGF is a potent mitogen and may act as a renotropic factor [80]. Increased serum levels of HGF have been reported in children with KD [81]. On the other hand, TGF-β1 is an inhibitor of cellular proliferation in many types of epithelial cells [80]. The reciprocal change between HGF and TGF-β1 may suggest a decreased antiproliferative effect of TGF-β1 on renal growth and also potentiate the mitogenic action of HGF, leading to nephromegaly in patients with KD.

n the other hand, TGF-β1 is an inhibitor of cellular proliferation in many types of epithelial cells [80]. The reciprocal change between HGF and TGF-β1 may suggest a decreased antiproliferative effect of TGF-β1 on renal growth and also potentiate the mitogenic action of HGF, leading to nephromegaly in patients with KD. Wang et al. [82] performed DMSA renal SPECT on 50 patients with KD and reported that 26 of these patients (52%) had renal inflammatory foci, in whom 11 of 24 patients (46%) still had renal scarring on a 6-month follow-up DMSA renal SPECT. These findings suggest that the potential long-term clinical impact of KD is not limited to CALs sequelae but also includes renal scar formation. Wu et al. [83] performed DMSA renal SPECT, renal Doppler ultrasonography to measure the pulsatility index (PI) and resistance index (RI), and analysis of urinary IL-6 levels in 50 patients with KD. Their study showed that 10 of 24 patients had renal inflammatory foci, that patients with renal inflammatory foci had significantly higher levels of urinary IL-6 and PI values than patients without renal inflammatory foci, and that there was a significant correlation between urinary IL-6 levels and PI values. These findings suggest that immune-mediated vasculitis is one of the mechanisms causing renal inflammation in KD.

tory foci had significantly higher levels of urinary IL-6 and PI values than patients without renal inflammatory foci, and that there was a significant correlation between urinary IL-6 levels and PI values. These findings suggest that immune-mediated vasculitis is one of the mechanisms causing renal inflammation in KD. 8. Renal Artery Lesions Although CALs are a well-known vascular complication of KD, vasculitis has also been observed in extra-coronary artery including arteries in the kidney. In autopsy studies, panarteritis was observed frequently in the kidney [38, 39] and was localized in the interlobar arteries and occurred rarely in the arcuate and interlobular arteries [83]. Compared with coronary arteritis, arteritis in the kidney developed several days later and inflammation was milder [84]. Renal artery aneurysms have been reported in patients with KD. Sasaguri and Kato [85] presented a patient with bilateral renal artery aneurysms without renovascular hypertension. Kato et al. [86] also reported that 5 of 594 patients with KD (0.8%) developed renal artery aneurysms. All patients with renal artery aneurysms in that report had coronary aneurysms, and no patient progressed to renovascular hypertension.

patient with bilateral renal artery aneurysms without renovascular hypertension. Kato et al. [86] also reported that 5 of 594 patients with KD (0.8%) developed renal artery aneurysms. All patients with renal artery aneurysms in that report had coronary aneurysms, and no patient progressed to renovascular hypertension. Renal artery stenosis has sometimes been reported in patients with KD, with or without renovascular hypertension. Negoro et al. [87] reported that an 11-month-old girl developed CALs, an abdominal aortic aneurysm and left renal artery stenosis 6 months after the onset of KD. Nagao and colleague [88] described a 3-year-old boy who developed right renal artery stenosis with renovascular hypertension 2 years following the acute phase of KD, which was successfully treated with percutaneous transluminal renal artery angioplasty (PTRA). Mawatari et al. [89] presented a 22-year-old man who developed renovascular hypertension due to right renal artery stenosis 14 years after the onset of KD, which was successfully treated with PTRA. Foster et al. [90] described an 18-month-old girl who developed renovascular hypertension due to right renal artery stenosis 6 months after the acute episode of KD. The patient underwent revascularization surgery because PTRA therapy could not conserve the right kidney function. Falcini et al. [91] reported two patients with KD (16-month-old girl and 3-month-old boy) who developed renovascular hypertension due to bilateral renal artery stenosis 3 weeks and 4 weeks after the onset of KD, respectively.

t revascularization surgery because PTRA therapy could not conserve the right kidney function. Falcini et al. [91] reported two patients with KD (16-month-old girl and 3-month-old boy) who developed renovascular hypertension due to bilateral renal artery stenosis 3 weeks and 4 weeks after the onset of KD, respectively. In addition, there are two reports of patients with renovascular hypertension associated with abdominal aortic aneurysms caused by KD [92, 93]. 9. Conclusions Kidney and urinary involvement of KD is uncommon and includes the following conditions: pyuria, prerenal AKI, renal AKI caused by TIN, HUS, and immune-complex mediated nephropathy, renal AKI associated with KDSS or unknown causes, ANS, NS, renal tubular abnormalities, renal abnormalities in imaging studies, and renal artery lesions. Although the precise pathogenesis underlying the development of kidney and urinary tract involvement in patients with KD remains unclear, vasculitis of the arteries in the kidney, immune-complex mediated kidney injuries, and T-cell immune-regulatory abnormalities have been proposed as possible mechanisms. Conflict of Interests The author declares that there is no conflict of interests regarding the publication of this paper.

1. Introduction Midline cervical cleft (MCC) is a rare congenital anomaly whose embryological origin is uncertain. Review of the international literature reveals at least 195 cases reported to date (not including the 10 patients in this series). Taken together, the information from previously published case reports has shown relatively consistent anatomic and pathologic findings, but there have been few series with more than several patients, and no one has undertaken a review of all the published literature. The purpose of this study was to evaluate the physical and pathologic findings associated with this condition and to provide a comprehensive catalogue of published cases in the world's literature. The findings in this large series of ten patients illustrate the clinical and operative findings, demographics, and treatment of this unusual entity.

was to evaluate the physical and pathologic findings associated with this condition and to provide a comprehensive catalogue of published cases in the world's literature. The findings in this large series of ten patients illustrate the clinical and operative findings, demographics, and treatment of this unusual entity. 2. Materials and Methods This study was performed as a retrospective chart review of patients treated by the University of Southern California Keck School of Medicine Department of Otolaryngology-Head and Neck Surgery and the Duke University Division of Otolaryngology-Head and Neck Surgery. Ten patients having the clinical and pathologic diagnosis of midline cervical cleft were identified. IRB approval was obtained from both USC and Duke. The charts were reviewed for history, clinical and pathologic findings as well as timing of surgical intervention and postoperative complications. Complete surgical excision of all components of the cleft was performed in every patient, and closure of the surgical wound was achieved using simple vertical closure or single versus multiple Z- or W-plasties depending on the length of the incision and the laxity of the neck tissue.

on and postoperative complications. Complete surgical excision of all components of the cleft was performed in every patient, and closure of the surgical wound was achieved using simple vertical closure or single versus multiple Z- or W-plasties depending on the length of the incision and the laxity of the neck tissue. The literature search was performed by using the search terms “midline cervical cleft” and “congenital midline cervical cleft” in PubMed and then performing an exhaustive review of all possible publications (journal articles, textbook chapters, doctoral dissertations, and case reports) and references. Copies of original manuscripts were obtained from library and internet resources although it was not possible to acquire some dissertations. In these cases, utilizing email communication, librarians provided the number of new cases described in the dissertation. Whenever possible, authors were contacted directly to clarify whether or not cases had been previously reported and in some cases to provide the gender of the patients. If no information was available on a reference, the publication was not included in the table due to the inability to verify the number of new patients with MCC. Information of the gender of the patients was included whenever possible. 3. Results Demographically, in this series, there were eight male and two female patients ranging in age from 2 months to 12 years. Seven children were Hispanic, two were Caucasian with Northern European lineage, and one was of Filipino descent.

The literature search was performed by using the search terms “midline cervical cleft” and “congenital midline cervical cleft” in PubMed and then performing an exhaustive review of all possible publications (journal articles, textbook chapters, doctoral dissertations, and case reports) and references. Copies of original manuscripts were obtained from library and internet resources although it was not possible to acquire some dissertations. In these cases, utilizing email communication, librarians provided the number of new cases described in the dissertation. Whenever possible, authors were contacted directly to clarify whether or not cases had been previously reported and in some cases to provide the gender of the patients. If no information was available on a reference, the publication was not included in the table due to the inability to verify the number of new patients with MCC. Information of the gender of the patients was included whenever possible. 3. Results Demographically, in this series, there were eight male and two female patients ranging in age from 2 months to 12 years. Seven children were Hispanic, two were Caucasian with Northern European lineage, and one was of Filipino descent. Clinically, there were six consistent findings: (1) a midline, vertical atrophic skin defect, (2) a lack of adnexal elements within this skin defect, (3) a superior skin tag, (4) an inferior blind sinus, (5) a midline subcutaneous fibrous cord, and (6) an increase in the size of the defect commensurate with an increase in the patient's age. Mucous could be expressed from almost all of the patients from the inferior sinus. The length of the skin defect ranged from 3 cm to 12 cm and there was an almost direct correlation between the age of the patient and the length of the defect (Figure 1). Patients were treated with surgical excision at the time of presentation or at one year of age for those patients who presented prior to their first birthday. The fibrous cord also became more prominent as the age of the patient increased with the older patients having some restriction of neck extension. Postoperatively, one patient had a wound infection that was treated with local wound care and healed without sequelae. There were no recurrences.

or to their first birthday. The fibrous cord also became more prominent as the age of the patient increased with the older patients having some restriction of neck extension. Postoperatively, one patient had a wound infection that was treated with local wound care and healed without sequelae. There were no recurrences. Table 1 provides a catalogue of the 195 cases found in the international literature. Author, date of publication, number of cases, and gender are included as well as pertinent notes. Some authors included cases which had been presented before, and in this situation only new patients appear in the table for that publication. Overall, there were 195 cases, 77 females and 58 males and 61 instances of cases presented, but no gender was given. Cases were restricted to the accepted definition of an MCC presentation (superior skin protuberance, midline skin defect, underlying fibrous cord, and inferior blind sinus without involvement of the mandible or sternum). Cases not having at least most of these elements were not included in the final tally.

r was given. Cases were restricted to the accepted definition of an MCC presentation (superior skin protuberance, midline skin defect, underlying fibrous cord, and inferior blind sinus without involvement of the mandible or sternum). Cases not having at least most of these elements were not included in the final tally. 4. Discussion Every patient in this series presented with the classic findings which define midline cervical cleft: a usually erythematous, vertical, and atrophic skin defect in the midline of the neck which lacks adnexal elements, a subcutaneous fibrous cord which is often longer than the overlying skin defect, a superior skin tag, and an inferior blind sinus. This constellation of clinical findings may be found clearly described in Ombredanne's work in 1949 [1]. Initially in the English literature Bailey in 1925 and Gross in 1940 called this entity “thyroglossal fistula,” but their pictures and descriptions are consistent with MCC [2, 3]. By the time he published on this subject again in 1953, Gross' nomenclature had changed to the term “midline cervical cleft” [4].

1949 [1]. Initially in the English literature Bailey in 1925 and Gross in 1940 called this entity “thyroglossal fistula,” but their pictures and descriptions are consistent with MCC [2, 3]. By the time he published on this subject again in 1953, Gross' nomenclature had changed to the term “midline cervical cleft” [4]. Luschka published the first case of MCC in 1848 under the description of “Congenital Fistula of the Neck” (translated) [5]. The drawing in his report is exactly the same as a picture of one of today's patients (Figure 2). In 1864 and 1865, there appeared three reports of neck fistulas by Heusinger but one is more consistent with a bronchogenic cyst (barrel chest, cyanosis, and sinus tract that extends from the left anterior chest toward the lower neck) and the other two have fistula tracts/sinuses involving the lateral aspect of the neck [6]. Those midline cervical clefts are part of a continuum of midline defects which can be seen from two cases that were excluded from the tally: Barsky's case in 1938 in which the patient had a thick midline cord but no epithelial defect [7] and Szenes' case from 1922 in which the cleft in the neck extends inferiorly into the sternum [8].

rvical clefts are part of a continuum of midline defects which can be seen from two cases that were excluded from the tally: Barsky's case in 1938 in which the patient had a thick midline cord but no epithelial defect [7] and Szenes' case from 1922 in which the cleft in the neck extends inferiorly into the sternum [8]. Early reports indicated a significant preponderance of female patients [9–13], but the inclusion of the new cases presented here helps to narrow the gap between the sexes. In 61 of the cases from the world literature, the gender was not reported. This could have an effect on the gender distribution if most of these cases were found in either girls or boys. There is no apparent explanation for why girls would be affected more than boys if in fact this gender predilection is true. Most patients with MCC do not have a family history of congenital anomalies or other birth defects [11, 14], and this was true for the patients in this series as well.

nd in either girls or boys. There is no apparent explanation for why girls would be affected more than boys if in fact this gender predilection is true. Most patients with MCC do not have a family history of congenital anomalies or other birth defects [11, 14], and this was true for the patients in this series as well. MCC is considered part of the midline branchiogenic syndromes [15, 16], but it is not a true cleft in the same way as a cleft palate. Many authors have concluded that incomplete fusion of the second branchial arches is largely responsible for this entity [11, 15, 17]. Several mechanisms have been proposed to explain this incomplete fusion. The presence of amniotic adhesions and vascular anomalies may cause localized tissue ischemia, necrosis, and scarring of the developing branchial arches [18], or pressure on the developing cervical area by the closely juxtaposed pericardial roof during the fifth week of gestation may produce similar results [1]. An underlying mesodermal deficiency [9] or a disturbance in the interaction between the mesoderm and the ectoderm in the developing cervical cleft skin [10] may explain the lack of adnexal elements in the skin defect. One group of authors has proposed that the superior skin tag found in patients with this pathology is formed by a ventral outgrowth of tongue muscle [12]. Some have proposed that MCC is an inferior presentation of Tessier's facial cleft #30 which is a cleft of the mandible [19–21]. Karík included MCC under the category of branchiogenic disorders along with disorders of the mandible, tongue, lower lip, and thorax [16]. Others have proposed that MCC represents a developmental field defect [22, 23]. Bergevin et al. concluded that the surface in MCC is an invagination of endodermal cells [12] whereas Ikuzawa et al. felt that MCC formed because the median sulcus had closed insufficiently allowing a migration of aberrant multipotential cells from which the various pathologic elements of MCC are derived [24].

22, 23]. Bergevin et al. concluded that the surface in MCC is an invagination of endodermal cells [12] whereas Ikuzawa et al. felt that MCC formed because the median sulcus had closed insufficiently allowing a migration of aberrant multipotential cells from which the various pathologic elements of MCC are derived [24]. MCC has been associated with thyroglossal duct and branchial cysts and possibly with accessory bronchi [46, 77, 97]. Indeed, MCC must be correctly diagnosed on pathology to differentiate it from a bronchogenic cyst which can also present as a midline atrophic skin defect with a sinus tract that weeps mucous or serous fluid. Pathologically, a bronchogenic cyst has a lining of pseudostratified ciliated columnar epithelium and often has accessory tissue such as smooth muscle, seromucinous glands, and sometimes cartilage, thus resembling a very immature bronchus [121]. However, the constellation of clinical and pathologic findings can distinguish between the two entities. The erythematous, linear, and atrophic skin defect of MCC can mimic other skin disorders such as linear scleroderma, but linear scleroderma lacks the other salient characteristics of MCC [111].

ature bronchus [121]. However, the constellation of clinical and pathologic findings can distinguish between the two entities. The erythematous, linear, and atrophic skin defect of MCC can mimic other skin disorders such as linear scleroderma, but linear scleroderma lacks the other salient characteristics of MCC [111]. Pathologic examination of the specimens from the 10 new patients presented here revealed the typical findings: the presence of skeletal muscle, minor salivary gland elements, lymphoid tissue and connective tissue, and the absence of adnexal skin elements such as hair or sebaceous glands. Sinopidis and colleagues present a very good tabular summary of the pathologic description of the individual MCC elements: the cephalic skin tag contains stratified squamous epithelium and striated muscle, the atrophic skin defect is comprised of stratified squamous epithelium without adnexal structures but with bundles of striated muscle in the dermis, and the caudal duct has squamous epithelium superficially but pseudostratified ciliated epithelium more deeply along with seromucinous glands [112].

and striated muscle, the atrophic skin defect is comprised of stratified squamous epithelium without adnexal structures but with bundles of striated muscle in the dermis, and the caudal duct has squamous epithelium superficially but pseudostratified ciliated epithelium more deeply along with seromucinous glands [112]. At birth, the external layer of the cleft may consist of a weeping, red membrane which then heals to produce cicatricial skin as the patient grows. The fibrous cord, which usually extends down to the pretracheal fascia as it did in these patients, also becomes more prominent as the child grows. This is because the affected tissues lag behind in vertical growth compared with the surrounding normal neck tissue. Those patients in whom the cord is apparent even without neck extension have difficulty extending their necks. When the fibrous cord extends to the level of the mandible, a bony spur is often seen on the anterior, inferior surface of the bone secondary to the traction placed on the mandible by this tethering cord which may be severe enough to produce an open bite deformity [11, 19, 91, 102, 111].

fficulty extending their necks. When the fibrous cord extends to the level of the mandible, a bony spur is often seen on the anterior, inferior surface of the bone secondary to the traction placed on the mandible by this tethering cord which may be severe enough to produce an open bite deformity [11, 19, 91, 102, 111]. This case series clearly demonstrates an important finding which impacts the timing of intervention in these patients. There was almost a direct correlation between the patient's age and the length of the defect (Figure 1). Whereas some previous publications recommended early excision only in those patients in whom the fibrous cord was prominent and severe producing inability to extend the neck or remodeling of the mandible [10, 11], early excision is recommended to prevent an increased scar length as well as the problems associated with a tethering midline cord [106, 111]. We treated patient surgically at the time of presentation for those over the age of one year and waited until age one for those who presented very early in life.

dible [10, 11], early excision is recommended to prevent an increased scar length as well as the problems associated with a tethering midline cord [106, 111]. We treated patient surgically at the time of presentation for those over the age of one year and waited until age one for those who presented very early in life. It is also important to completely excise the lesion. Simply transecting the fibrous cord or performing incomplete excision of the cutaneous and subcutaneous elements leads to recurrence [9–11, 74]. Closure of the surgical defect is performed with a simple vertical closure if the defect is not long and the surrounding skin is lax. The use of single or multiple W- or Z-plasties is recommended for longer defects to break up the scar and improve the cosmetic and functional results. This has long been proposed as the best way to deal with the vertical defect created by excision of the MCC and has become the usual way in which many patients are closed [40]. Early some patients with a long defect treated with a vertical closure developed neck contractures and an open bite deformity secondary to scarring after the surgery [11].

as the best way to deal with the vertical defect created by excision of the MCC and has become the usual way in which many patients are closed [40]. Early some patients with a long defect treated with a vertical closure developed neck contractures and an open bite deformity secondary to scarring after the surgery [11]. 5. Conclusion This case series demonstrates two interesting points. First, there was a preponderance of male patients (8/10) in contrast to previous case series in which females have predominated. Second, since the length of the defect increased as the patient's age increased, early excision of the lesion to minimize scarring is recommended. The catalogue of cases from the world literature also provides an organized list that may be helpful for future research. Acknowledgments The author would like to thank Dale Rice, MD, Professor, University of Southern California, Department of Otolaryngology-Head and Neck Surgery, and Dennis Crockett, MD, of Head/Neck Associates of Orange County, for their help in acquiring patient data; H. Wolfgang Beumer, MD, for his invaluable help in translating the German language articles; and Stephen Perlman from the Durham VA Library for his help in acquiring articles from other institutions. None of them received any compensation for their contributions, but their help was greatly appreciated. Conflict of Interests The author declares that there is no conflict of interests regarding the publication of this paper. Figure 1 Length of involved skin relative to patient age.

Conflict of Interests The author declares that there is no conflict of interests regarding the publication of this paper. Figure 1 Length of involved skin relative to patient age. Figure 2 Consistency of presentation between a patient from 1848 (a) and the modern day (b). Table 1 Catalogue of the published world literature by author and year. Author Year Number of pts F M Comments von Luschka [5] 1848 1   1 Picture of the patient in the journal is a classic MCC with superior skin tag, superficial skin defect overlying a midline cervical web Roth [25] 1878 1   1 Cusset [26] 1887 1 1 Arndt [27] 1888 1   1 Articles published in 1889 and 1892 by Arndt are further discussions of the same cases but no new cases of midline cleft/sinuses Delkeskamp [28] 1906 1   1 Bailey [2] 1925 2 2   Called “thyroglossal fistula” but picture and description consistent with MCC Nylander [29] 1928 2 1 Hein [30] 1931 1 1 Gross and Connerley [3] 1940 2 2   Called “thyroglossal fistula” but picture and description consistent with MCC Mouchet [31] 1942 1   1 Wynn-Williams [32] 1952 2 2 Gross [4] 1953 4 4   It includes the 2 cases from 1940 in his total of 6; he calls them MCC now rather than thyroglossal fistula Haym [33] 1954 1 1 Karfik [34] 1958 2 Maneksha [35] 1961 1 1 Van Duyn [9] 1963 1   1 Brown [36] 1963 2 Schaller [37] 1963 1 Cronin and Converse [38] 1964 1 1   Same case appears in 1990 in ch 39 Deformities of the Cervical Region in Plastic Surgery, vol. 3, McCarthy JG, ed. 1990, Philadelphia, WB Saunders, pp. 2078–2085 Amr [39] 1964 1   1 Königová [40] 1965 18     Article total is 20 but 2 were already reported by Karfik in 1958; same article appears in Czech language in 1966 Rozhl Chir (the 1966 article's summary indicates 22 cases, but the article text indicates only 20) Gottlieb and Lewin [10] 1966 2 2 Monroe [41] 1966 2     Passing reference to personal experience of two patients with fissura colli medialis Cosman and Crikelair [15] 1969 2     In discussion of the continuum of midline branchiogenic syndromes, 2 cases of MCC are mentioned Kriens and Schuchardt [42] 1969 2 1 1 Extensive bibliography and listing of many cases until 1966 Michalland [43] 1970 6 Sanchez Lopez Tello and Straube [44] 1970 1   1 W. Pirsig and H.

s Cosman and Crikelair [15] 1969 2     In discussion of the continuum of midline branchiogenic syndromes, 2 cases of MCC are mentioned Kriens and Schuchardt [42] 1969 2 1 1 Extensive bibliography and listing of many cases until 1966 Michalland [43] 1970 6 Sanchez Lopez Tello and Straube [44] 1970 1   1 W. Pirsig and H. Pirsig [45] 1972 1 French and Bale [46] 1973 1 1 Sanchez Lopez Tello and Mueller [47] 1973 2   2 Dargallo Reventos [48] 1975 4     In addition, one “incomplete” case is mentioned that was entirely superficial without any deeper elements and was incompletely described so it was excluded from the final tally Ohtsuka et al. [49] 1976 1 Post [50] 1976 2   2 Isono et al. [51] 1977 1 Balcells-Par and Sancho-Cerquella [52] 1977 3 Andersen and Svendsen [53] 1978 1 1 Minami et al. [54] 1980 1   1 Fritzmeier and Kronsbein [55] 1982 1 1 Wood and Deister [56] 1983 1   1 Godbersen and Wiedemann [57] 1984 1 1 Cotin et al. [58] 1984 1 Lardenet et al. [59] 1984 1   1 It states 70 cases had been reported in the world literature Gargan et al. [11] 1985 6 5 1 Article total is 12 because it includes the cases that Gross reported in 1940 and 1953 Soper et al. [60] 1986 1 Breton et al. [61] 1987 4   4 Three of the cases are represented in another article published in 1989 Godbersen et al. [22] 1987 2 1 1 Novák and Jakoubková [62] 1987 1   1 Massardier [63] 1987 4 Desnos et al. [64] 1987 4 4 Mihara and Wada [65] 1987 1 Lindsay et al. [66] 1988 2 Bergevin et al. [12] 1989 1 1 S. G. Fincher and G. G. Fincher [67] 1989 1   1 Breton and Freidel [68] 1989       All 3 cases are included in the 1987 article, no new cases Van der Meulen et al. [69] 1990 1   1 Raffensperger [70] 1990 1 van der Staak et al. [71] 1991 2 1 1 Ikuzawa et al. [24] 1992 1 1 Nicklaus et al. [72] 1992 2 1 1 It includes a case reported by J Friedberg in Dec 1989 in Ped Clinc of North Amer

l [68] 1989       All 3 cases are included in the 1987 article, no new cases Van der Meulen et al. [69] 1990 1   1 Raffensperger [70] 1990 1 van der Staak et al. [71] 1991 2 1 1 Ikuzawa et al. [24] 1992 1 1 Nicklaus et al. [72] 1992 2 1 1 It includes a case reported by J Friedberg in Dec 1989 in Ped Clinc of North Amer Montinet et al. [73] 1992 1   1 Maddalozzo et al. [17] 1993 5 4 1 Liu and Lee [14] 1994 1   1 Kececi et al. [19] 1994 1 1   Description consistent with a MCC Maschka et al. [74] 1995 1 1 Sfeir et al. [75] 1995 4 Ayache et al. [76] 1997 1   1 Andryk et al. [77] 1999 1   1 Soderberg et al. [78] 1999 1 1 Eastlack et al. [79] 2000 1   1 Çetinkurşun [80] 2001 1   1 Ercocen [23] 2002 1 1 Genc et al. [81] 2002 1   1 Sanchez Lopez Tello [82] 2002 1   1 Article total is 4 but includes cases of Sanchez-Lopes-Tello reported in 1970 with Straube and in 1973 with Muller Tsukuno et al. [83] 2002 1 1 Hirokawa et al. [84] 2003 1 1 Joshi et al. [20] 2003 1 1 Mylnarek et al. [85] 2003 1 1 Daw and Patel [86] 2003 1   1 Sannajust et al. [87] 2004 1 1 Derbez et al. [88] 2004 5 3 2 Bajaj et al. [89] 2004 1   1 Tagliarini et al. [90] 2004 2 1 1 Gardner and Moss [91] 2005 1   1 Saha et al. [13] 2005 2 1 1 Cochran et al. [92] 2006 1 1 C. O. Kara and I. G. Kara [93] 2006 1   1 Foley and Fallat [94] 2006 1 Smith et al. [95] 2006 3 2 1 Agag et al. [96] 2007 1   1 Mendis and Moss [97] 2007 2 1 1 Franzese et al. [98] 2008 2 2 Turkyilmaz et al. [99] 2008 1 Cheng and Gottschall [100] 2009 1 1 Renukaswamy et al. [101] 2009 4 1 3 Sharma et al. [102] 2009 1   1 Vure et al. [103] 2009 1 1 Warden and Millar [104] 2010 1 1 Nijkamp and Rijlaarsdam [105] 2011 1 1 Patil et al. [106] 2011 1 1 Grynspan et al. [107] 2012 1   1 Jakobsen et al. [108] 2012 1 1   It presents 3 cases but two involved the mandible McInnes et al. [109] 2012 1   1 Doddamani et al. [110] 2012 1 1 Mendez-Gallart et al. [111] 2012 1 1 Sinopidis et al. [112] 2012 1 1 Tröbs et al. [113] 2012 2   2 Martí Fajardo [114] 2012 1   1 Lillehei and Coran [115] 2012 1 1 Farhadi et al. [116] 2012 2 1 1 Mirza [117] 2013 1   1 Saha et al. [118] 2013 2 2 0 It includes two cases initially described in Saha's 2005 paper; a case of a true cleft from the mandible to the sternum was excluded from the tally Eom et al. [119] 2014 1 1 0 First case reported in the Korean population Crippa et al. [120] 2014 1 1 0

[116] 2012 2 1 1 Mirza [117] 2013 1   1 Saha et al. [118] 2013 2 2 0 It includes two cases initially described in Saha's 2005 paper; a case of a true cleft from the mandible to the sternum was excluded from the tally Eom et al. [119] 2014 1 1 0 First case reported in the Korean population Crippa et al. [120] 2014 1 1 0 Puscas (new cases) 2015 10 2 8 TOTAL   205 79 66

1. Introduction Rubella is a mild viral infection that usually manifests with fever and rash. Its public health significance primarily lies in its ability to cause congenital rubella syndrome (CRS) that results in devastating malformations and significant long-term disability [1–3]. Rubella infection has a teratogenic effect on the developing embryo, which may cause abortion if a woman is infected early in the gestational period or if pregnancy continues, may result in major cardiac anomalies, sensorineural hearing loss, cataracts, or death. With the use of effective rubella vaccines, rubella was targeted for elimination in two regions of the World Health Organization (WHO) by 2015. But by the end of 2015, only the Region of the Americas was declared to be free of endemic rubella transmission. Worldwide, use of rubella-containing vaccine (RCV) is increasing but despite RCV introduction in 74% of 194 WHO member states, in 2014, global infant immunization coverage remained low at 46% [4]. Surveillance for rubella and CRS is crucial in monitoring the impact of immunization programs to assess disease burden before and after RCV introduction. Although rubella surveillance is being conducted in most countries in conjunction with measles surveillance, surveillance activities for CRS, particularly in developing countries, have proven to be more challenging. Out of 194 member states, only 75 countries began reporting in 2000, which increased to 114 in 2014 but only 14 countries reported positive case identification [4, 5]. Thus, the true burden of CRS remains underestimated [6].

e activities for CRS, particularly in developing countries, have proven to be more challenging. Out of 194 member states, only 75 countries began reporting in 2000, which increased to 114 in 2014 but only 14 countries reported positive case identification [4, 5]. Thus, the true burden of CRS remains underestimated [6]. Since 2003, the Western Pacific Region of the World Health Organization (WHO) has resolved to accelerate the control of rubella and prevention of CRS through integration with measles elimination activities [7]. In October 2014, the Western Pacific Region included rubella including CRS elimination as one of eight regional immunization goals specified by the Regional Framework for Implementation of the Global Vaccine Action Plan in the Western Pacific [8]. In the Philippines, RCVs, either singly or in combination with measles and mumps vaccines, had been available in the private sector for more than two decades [9, 10] but this sector is estimated to cover only 5%–10% of the population. RCV was only included in the country's Expanded Programme of Immunization (EPI) in 2010 and given routinely to all infants in 2011 at 12 months of age. Since 2015, through a school-based immunization strategy, children are given a second dose of RCV at school entry (7 years of age) to ensure that children receive at least two doses of measles-containing vaccines. No adult immunization is provided in the public sector and rubella serologic screening of pregnant women is not mandated by the government.

d immunization strategy, children are given a second dose of RCV at school entry (7 years of age) to ensure that children receive at least two doses of measles-containing vaccines. No adult immunization is provided in the public sector and rubella serologic screening of pregnant women is not mandated by the government. Surveillance for rubella was integrated into the measles surveillance in 2010; however passive laboratory-confirmed surveillance for rubella has been conducted since 2005. In 2009 and 2010, there were 310 and 1,092 serologically confirmed rubella cases out of 1,279 and 4,085 specimens tested, respectively, nationwide [11]. Because there is minimal baseline information on the burden of rubella in the Philippines, we conducted a systematic review of studies on rubella susceptibility, as well as a review on CRS to provide information on rubella and CRS prior to routine vaccine introduction.

5 specimens tested, respectively, nationwide [11]. Because there is minimal baseline information on the burden of rubella in the Philippines, we conducted a systematic review of studies on rubella susceptibility, as well as a review on CRS to provide information on rubella and CRS prior to routine vaccine introduction. 2. Methods 2.1. Systematic Review For the systematic review, published and unpublished studies on rubella were sought. PubMed searches were performed using the search terms “Philippines” and “rubella” without any limitations on dates. References of the identified publications were reviewed. Literature search in the Philippine databases, namely, HERDIN (Health Research Information and Development Network) and PIMEDICUS (Philippine Index Medicus), was performed using the search terms “rubella,” “CRS,” or “congenital infection.” HERDIN and PIMEDICUS include both locally published and unpublished works. Content experts and specialists were also asked for ongoing and unpublished works on the subject that we may have missed in the local and international database search. Two authors (ALL and PFR) reviewed the retrieved articles and tabulated information in Microsoft Excel™.

both locally published and unpublished works. Content experts and specialists were also asked for ongoing and unpublished works on the subject that we may have missed in the local and international database search. Two authors (ALL and PFR) reviewed the retrieved articles and tabulated information in Microsoft Excel™. 2.2. Seroprevalence Survey To determine the proportion of women who were susceptible, that is, seronegative for rubella, a cross-sectional study of pregnant women who consulted in the obstetric outpatient clinic of the Philippine General Hospital (PGH) in the city of Manila for antenatal care was conducted in 2001-2002. Sample size was computed based on the seroprevalence study of Alday [14] that 27% were seronegative in the urban population, and using an interval of 5 with a 95% confidence level, and a 25% dropout rate, then a minimum of 373 pregnant women were required. After obtaining informed consent, data collection forms were completed specifically noting the following: age, occupation, educational attainment, age of gestation, number of previous gestations, and history of rubella immunization. Blood was obtained from the subjects and stored at −20°C until ELISA for rubella was performed. Rubella-specific IgG was tested using commercially available enzyme-linked immunosorbent assay kit (CAPTIA™; Trinity Biotech, Clark Laboratories, TX, USA). ELISA was performed following the manufacturer's instructions.

tion. Blood was obtained from the subjects and stored at −20°C until ELISA for rubella was performed. Rubella-specific IgG was tested using commercially available enzyme-linked immunosorbent assay kit (CAPTIA™; Trinity Biotech, Clark Laboratories, TX, USA). ELISA was performed following the manufacturer's instructions. Data was entered in Microsoft Excel and analysed using Stata 7™ (Stata Corporation, TX, USA). Descriptive statistics for maternal data (age, educational attainment, number of previous gestations, age of gestation, occupation, and history of rubella immunization) and 95% confidence intervals (CI) were calculated. Tests for association of above factors with rubella susceptibility were determined using chi-square test or Fisher's exact test when data were sparse, and statistical significance was set at p < 0.05. The study was reviewed and approved by the University of the Philippines Manila-Review Ethics Board prior to patient enrollment. Written informed consent was obtained. Women who had no detectable rubella IgG were informed of the result and advised on the need for immunization.

tistical significance was set at p < 0.05. The study was reviewed and approved by the University of the Philippines Manila-Review Ethics Board prior to patient enrollment. Written informed consent was obtained. Women who had no detectable rubella IgG were informed of the result and advised on the need for immunization. 3. Results 3.1. Systematic Literature Review Our search yielded 11 locally published [10, 12–21] and 4 unpublished studies [22–25] (Figure 1), none of which were indexed in PubMed. From 1973–1997, 5 rubella seroprevalence studies were published, 4 of which were conducted in Metro Manila with one including a rural area [10, 12–14], and 1 was in urban Cebu [15] (Table 1). Rubella susceptibility was higher in studies conducted from the 1970s–1980s [10, 12–14] as well as from participants in the rural areas [10, 14].

rubella seroprevalence studies were published, 4 of which were conducted in Metro Manila with one including a rural area [10, 12–14], and 1 was in urban Cebu [15] (Table 1). Rubella susceptibility was higher in studies conducted from the 1970s–1980s [10, 12–14] as well as from participants in the rural areas [10, 14]. Ten studies were on CRS [15–24] with one study on pregnancy outcomes [25]. Of the 10 studies on CRS, seven were published (Table 1) and three were unpublished (Table 2). Nine of the 10 studies were retrospective reviews and only one was cross-sectional [18]. Seven of the studies were conducted in a tertiary public hospital (PGH) located in Manila [16, 17, 19, 20, 22–24]. Among published studies, one study proportionately described CRS in congenital cataract cases [19] and two studies described CRS on patients with hearing impairment [18, 21]. Among all patients with congenital cataract, CRS was found to be the most common cause of secondary congenital cataract (28.7% of all congenital cataracts) [19]. CRS was also the most common cause of secondary congenital sensorineural hearing loss [18, 21]. In a prospective study students in a School for the Deaf, 28 (4.8%) children reported other associated impairments (e.g., cataract, blindness, and cardiac defect) suggesting CRS, while 136 (23.5%) reported possible maternal rubella exposure [18].

common cause of secondary congenital sensorineural hearing loss [18, 21]. In a prospective study students in a School for the Deaf, 28 (4.8%) children reported other associated impairments (e.g., cataract, blindness, and cardiac defect) suggesting CRS, while 136 (23.5%) reported possible maternal rubella exposure [18]. Only one unpublished study reported on pregnancy outcomes among women who were tested for rubella titers. Out of 124 pregnant women, there were 4 (3.2% of all tested) cases of maternal rubella infection: 1 had fetal death in utero, and 3 babies were diagnosed with CRS [25]. Two studies characterized the clinical profile of CRS among physician-diagnosed cases [17, 20]. The most common clinical manifestations of CRS in the reports were cataract [20] or congenital heart disease [17]. Patent ductus arteriosus and pulmonary stenosis were the most common cardiac manifestations seen in about 45% of the reviewed cases of children with CRS [17].

CRS among physician-diagnosed cases [17, 20]. The most common clinical manifestations of CRS in the reports were cataract [20] or congenital heart disease [17]. Patent ductus arteriosus and pulmonary stenosis were the most common cardiac manifestations seen in about 45% of the reviewed cases of children with CRS [17]. 3.2. Cross-Sectional Study From March 5, 2002, to June 5, 2002, 383 pregnant females participated in the study, of which, 324 (84.6%) had positive rubella IgG by ELISA and 59 (15.4%) were negative or had equivocal results (Table 3). There were no significant differences seen in the rubella immune and susceptible groups as regards age, level of education, place of work, or history of rubella immunization. Only 3 women reported being immunized against rubella in the past, all of whom were seropositive. Among the 12 subjects who were unsure of their rubella immunization, 2 were found to be seronegative. No association was noted between rubella seropositivity and the factors studied (age, educational attainment, more than one pregnancy, age of gestation, and occupation). 4. Discussion Our findings suggest that rubella, including congenital rubella syndrome, is an important public health problem in the Philippines, causing long-term disabilities such as deafness and other congenital disabilities. No prospective studies were conducted in the country on CRS detection, all the studies were retrospective, and most were hospital based; hence most of the cases with CRS that were reported were the ones that required acute care such as congenital heart diseases.

ities such as deafness and other congenital disabilities. No prospective studies were conducted in the country on CRS detection, all the studies were retrospective, and most were hospital based; hence most of the cases with CRS that were reported were the ones that required acute care such as congenital heart diseases. Based on the results of the seroprevalence studies, the number of susceptible women appeared to have declined in the Philippines since the 1980s, wherein 37% of all women tested were susceptible to rubella [14]. In 2002, 15% of pregnant women in an urban setting remained susceptible to rubella. It is likely that the number of susceptible women has not changed substantially as RCV was introduced in the public health setting only in 2010 in young children. It is also likely that rubella susceptibility is higher in rural areas, as was observed in previous rubella susceptibility studies [10, 14]. In 2002, there were 1,666,773 live births in the Philippines [26]; if at least 15% of pregnant women were susceptible to rubella, then a substantial number of pregnancies would have been at risk for CRS. In the Philippines it is estimated that there are 100 to 149 CRS cases per 1,000 live births annually [27].

Based on the results of the seroprevalence studies, the number of susceptible women appeared to have declined in the Philippines since the 1980s, wherein 37% of all women tested were susceptible to rubella [14]. In 2002, 15% of pregnant women in an urban setting remained susceptible to rubella. It is likely that the number of susceptible women has not changed substantially as RCV was introduced in the public health setting only in 2010 in young children. It is also likely that rubella susceptibility is higher in rural areas, as was observed in previous rubella susceptibility studies [10, 14]. In 2002, there were 1,666,773 live births in the Philippines [26]; if at least 15% of pregnant women were susceptible to rubella, then a substantial number of pregnancies would have been at risk for CRS. In the Philippines it is estimated that there are 100 to 149 CRS cases per 1,000 live births annually [27]. Similar to developing countries in the Western Pacific Region (WPR), the Philippines has recently introduced RCV. Routine surveillance data for CRS and rubella in the region are just being reported. In Myanmar, the burden of CRS during interepidemic periods was similar to that seen prior to the introduction of RCV in developed countries [28]. Following an outbreak of rubella in Vietnam, surveillance for CRS in Vietnam yielded 292 cases of CRS [29]. In 2014 and 2015, 6 and 7 countries in the WPR have reported >1 confirmed rubella case per 1,000,000 population while no country reported a case of CRS [30]. In the Philippines, reporting CRS cases is not mandatory; hence the disease is likely underestimated. Surveillance on rubella has been integrated in measles-rubella surveillance, wherein patients presenting with rash and fever who tested negative for measles are eventually tested for rubella. This may also be an underestimate as most cases of rubella have an indolent course. Nevertheless, the country reported 3.4 and 1.5 and 1.4 (annualised) laboratory-confirmed rubella cases per 1,000,000 population in 2014, 2015, and 2016, respectively [30].

o tested negative for measles are eventually tested for rubella. This may also be an underestimate as most cases of rubella have an indolent course. Nevertheless, the country reported 3.4 and 1.5 and 1.4 (annualised) laboratory-confirmed rubella cases per 1,000,000 population in 2014, 2015, and 2016, respectively [30]. The WHO recommends two approaches for rubella immunization. The first recommends immunization of adolescent and adult women of childbearing age, aimed at reducing the burden of CRS. The second, which is suitable for elimination of both rubella and CRS, interrupts the transmission of rubella virus by the introduction of RCV into the routine childhood immunization schedule in combination with the vaccination of older age groups [3]. In view of the regional goal of rubella elimination, RCVs have been introduced in the Philippines at 9 and 12–15 months of age as well as at 7 years of age in schools. However, rubella susceptibility screening is not routinely conducted among pregnant women and immunization of women of child bearing age has not been included in the public health setting. Continued vigilance and enhanced surveillance for CRS and rubella are important in identifying cases to attain the regional goal of rubella elimination. 5. Conclusions Our review presents the baseline information on rubella and CRS and supports the establishment of CRS surveillance in the country to identify the burden, as well as for programmatic implementation to monitor the impact of immunization. Competing Interests The authors declare that they have no competing interests.

5. Conclusions Our review presents the baseline information on rubella and CRS and supports the establishment of CRS surveillance in the country to identify the burden, as well as for programmatic implementation to monitor the impact of immunization. Competing Interests The authors declare that they have no competing interests. Figure 1 Flow of articles reviewed. Table 1 Published studies of serologic surveys and clinical manifestations of congenital rubella syndrome (CRS) in the Philippines. (a) Serologic surveys (n = 5). Author Date Setting Methods Inclusion criteria Confirmatory test Results Espiritu-Campos et al. 1973 [12] 1973 Community setting (urban) Prospective Male and female volunteers (0–40 years old) Hemagglutination-Inhibition test 32.2% rubella susceptible Del Mundo 1973 [10] 1973 Community setting (urban and rural) Prospective Male and female volunteers (0–40 years old) Hemagglutination-Inhibition test 26% rubella susceptible Chan et al. 1979 [13] 1978 Community setting (urban) Prospective Nonpregnant women (6–45 years old) Hemagglutination-Inhibition test 29.6% were rubella susceptible (38.1% for women ≤ 30 years old, 12.5% for >30 years old) Alday et al. 1982 [14] 1982 Tertiary hospitals (urban Manila City and rural Las Pinas) Prospective Nonpregnant women aged (16–45 years old) Hemagglutination-Inhibition test 37.3% were rubella susceptible (urban women ≤ 30 years old, 80%; urban women > 30 years old, 20%; and rural women ≤ 30 years old, 64%, rural women > 30 years old, 36%)

4] 1982 Tertiary hospitals (urban Manila City and rural Las Pinas) Prospective Nonpregnant women aged (16–45 years old) Hemagglutination-Inhibition test 37.3% were rubella susceptible (urban women ≤ 30 years old, 80%; urban women > 30 years old, 20%; and rural women ≤ 30 years old, 64%, rural women > 30 years old, 36%) Yu et al. 1997 [15] 1997 Tertiary hospital (Cebu City) Prospective Pregnant women (17–45 years old) Microparticle Enzyme Immunoassay (IgG) 10.9% were rubella susceptible (b) Congenital rubella syndrome (n = 6). Author Month/year study was conducted CRS manifestation(s) primarily used to identify cases Methods Included subjects Diagnostic criteria for CRS Results Nueva-Espana et al. 1988 [16] January 1981 to June 1986 Hearing loss Retrospective review of audiology clinic records and audiologic test Children 0–16 years old presenting in a tertiary hospital with hearing loss (mean 4.7 years old) Hearing loss plus maternal history of febrile rash and other signs associated with CRS 17 cases (out of 496) had maternal history of rubella, and only 3 cases of the 17 were diagnosed with CRS Santos-Cabaero et al. 1998 [17] January 1990 to December 1994 Cardiac, hearing loss, and cataract Retrospective review of patient records Children diagnosed with CRS by physicians Rubella IgM confirmed CRS cases Deafness and cataract were the most predominant symptoms with 26 (62%) and 28 (67%) cases, respectively. Other clinical findings included were psychomotor delay (50%), congenital heart disease (45%), neonatal cholestasis (31%), and glaucoma (4 cases or 10%)

agnosed with CRS by physicians Rubella IgM confirmed CRS cases Deafness and cataract were the most predominant symptoms with 26 (62%) and 28 (67%) cases, respectively. Other clinical findings included were psychomotor delay (50%), congenital heart disease (45%), neonatal cholestasis (31%), and glaucoma (4 cases or 10%) Yu and Rameriz 2003 [18] September to October 2002 Hearing loss Prospective, cross-sectional study using questionnaire Students enrolled in a School for the Deaf (mean 14.2 years old) Hearing loss plus maternal history of rubella 136 (23.5%) deaf students had maternal history of rubella, among which 15 (2.6%) had visual problems and 8 (1.4%) had congenital heart disease Tecson and Santiago 2004 [19] January 2000 to August 2003 Cataract Retrospective review of patient records Infants 0–12 mos with atraumatic cataract in a tertiary hospital Congenital cataract with history of maternal measles and heart disease 45 cases (out of 218 cases, or 20.5%) had CRS. 18 cases (out of 218 cases or 8.2%) had suspected CRS CRS was the most common cause of secondary cataract Agnas 2005 [20] January 1995 to December 2002 Cardiac, cataract, and hearing loss Retrospective review of patient records Children diagnosed with CRS by physicians Rubella IgM confirmed CRS cases Cataract was the most common clinical manifestation, followed by patent ductus arteriosus at 24 (49%) and 15 cases (31%), respectively. The other clinical manifestations were hepatomegaly (10%), jaundice (10%), pulmonary artery stenosis (6%), extrauterine growth retardation (4%), glaucoma (2%), and hemolytic anemia (2%)

ataract was the most common clinical manifestation, followed by patent ductus arteriosus at 24 (49%) and 15 cases (31%), respectively. The other clinical manifestations were hepatomegaly (10%), jaundice (10%), pulmonary artery stenosis (6%), extrauterine growth retardation (4%), glaucoma (2%), and hemolytic anemia (2%) Tipayno 2008 [21] January 1996 to December 2005 Hearing loss Retrospective review of audiologic records Patients tested in audiologic clinic of a pediatric specialty tertiary hospital (mean 3.9 years old) Clinically diagnosed CRS (criteria used for diagnosis not mentioned) 48 patients out of 2,783 (1.7%) physician-diagnosed CRS 44 patients with CRS (91%) with hearing loss, 80% of which were severe to profound hearing loss Table 2 Unpublished studies (n = 4) on congenital rubella syndrome. Author Year Sign Methods Included subjects Results Tanglao-Salazar 1993 [22] 1993 Hearing loss Retrospective chart review Children with hearing impairment 18.4% with maternal history of rubella infection during pregnancy Rodriguez 1995 [23] 1995 Hearing loss Retrospective chart review Children with hearing impairment 13% diagnosed with congenital rubella Santos-Gonzales and Santiago 2013 [24] 2013 Cataract Retrospective review of records in pediatric ophthalmology clinic Children with suspected, probable, or laboratory-confirmed CRSa Out of 23 cases, there were 6 (26%) rubella IgM confirmed, 11 (48%) probable, and 6 (26%) CRS suspected cases. Cataract was seen in 21 (91%) of the cases of CRS. Two patients presented with pigmentary retinopathy

f records in pediatric ophthalmology clinic Children with suspected, probable, or laboratory-confirmed CRSa Out of 23 cases, there were 6 (26%) rubella IgM confirmed, 11 (48%) probable, and 6 (26%) CRS suspected cases. Cataract was seen in 21 (91%) of the cases of CRS. Two patients presented with pigmentary retinopathy Limgenco 2000 [25] 2000 Pregnancy outcome Retrospective review Pregnant women with TORCHb titers 4 cases of maternal rubella infection, 1 had fetal death in utero, and 3 babies were diagnosed with congenital rubella syndrome

f records in pediatric ophthalmology clinic Children with suspected, probable, or laboratory-confirmed CRSa Out of 23 cases, there were 6 (26%) rubella IgM confirmed, 11 (48%) probable, and 6 (26%) CRS suspected cases. Cataract was seen in 21 (91%) of the cases of CRS. Two patients presented with pigmentary retinopathy Limgenco 2000 [25] 2000 Pregnancy outcome Retrospective review Pregnant women with TORCHb titers 4 cases of maternal rubella infection, 1 had fetal death in utero, and 3 babies were diagnosed with congenital rubella syndrome aThe 1997 US CDC criteria for CRS include the following. Clinical description: an illness is usually manifesting in infancy resulting from rubella infection in utero and characterized by signs or symptoms from the following categories: (A) cataracts/congenital glaucoma, congenital heart disease (most commonly patent ductus arteriosus, or peripheral pulmonary artery stenosis), loss of hearing, and pigmentary retinopathy; (B) purpura, splenomegaly, jaundice, microcephaly, mental retardation, meningoencephalitis, and radiolucent bone disease. Case classification: suspected: a case with some compatible clinical findings but not meeting the criteria for a probable case; probable: a case that is not laboratory confirmed and that has any two complications listed in paragraph (A) of the clinical description or one complication from paragraph (A) and one from paragraph (B) and lacks evidence of any other etiology; confirmed: a clinically compatible case that is laboratory confirmed; infection only: a case that demonstrates laboratory evidence of infection, but without any clinical symptoms or signs. Laboratory confirmation is through viral isolation, rubella IgM, and rubella antibody level that persists at a higher level and for a longer period than expected from passive transfer of maternal antibody.

ction only: a case that demonstrates laboratory evidence of infection, but without any clinical symptoms or signs. Laboratory confirmation is through viral isolation, rubella IgM, and rubella antibody level that persists at a higher level and for a longer period than expected from passive transfer of maternal antibody. bTORCH: toxoplasma, rubella, cytomegalovirus, herpes titers. Table 3 Characteristics of pregnant women and their rubella serologic status, 2002. Variable N Immune n (%, 95%  CIa) Susceptible n (%, 95%  CIa) P value Age 0.33 16–25 176 152 (86.36, 80.39–91.06) 24 (13.64, 8.94–19.61) 26–35 168 142 (84.52, 78.15–89.63) 26 (15.48, 10.37–21.85) 36–45 39 30 (76.92, 60.67–88.87) 9 (23.08, 11.13–39.33) Education 0.88 Elementary 21 17 (80.95, 58.9–94.55) 4 (19.05, 5.45–41.91) High school 180 152 (84.44, 78.31–89.41) 28 (15.56, 10.59–21.69) College 182 155 (85.16, 79.15–89.99) 27 (14.84, 10.01–20.85) Work 0.21 Primarily at home 294 245 (83.33, 78.57–87.41) 49 (16.67, 12.59–21.43) Primarily outside 89 79 (88.76, 80.30–94.47) 10 (11.24, 5.52–19.69) History of rubella immunization 0.75 None 368 311 (84.51, 80.40–88.05) 57 (15.49, 11.95–19.6) Yes 3 3 (100) 0 Unknown 12 10 (83.33, 51.58–97.91) 2 (16.67, 2.09–48.41) aExact binomial 95% confidence interval.

1. Introduction Missed clinic appointments present a significant burden to the health care system and prevent optimal care for patients. Pediatric visits are particularly important for several reasons. Growth at this age is rapid, and frequent well-child visits in the beginning years are crucial to evaluate development. Pediatricians monitor height and weight gain and motor and speech development, and administer vaccinations at proper times per recommendations set forth by the American Academy of Pediatrics. For the practice, a scheduled appointment slot represents time and effort invested by the clinic and scheduling personnel to prepare for the visit. In addition, failure to attend an appointment represents a missed opportunity for other patients that may require a timely follow-up, disrupting optimal care. With many advances in technology in the last few decades, the use of electronic reminders has become increasingly prevalent. Text messages are a convenient mode of communication that is widely used and can effectively reach a large population. The most recent data show that 83% of American adults own a cell phone [1]. Although many practices exist in urban settings where patients may have limited resources, cell phone ownership remains prevalent. In addition to the convenience of receiving text messages, there is the added benefit of a saved reminder that can be accessed again, a benefit not always available with phone call reminders.

though many practices exist in urban settings where patients may have limited resources, cell phone ownership remains prevalent. In addition to the convenience of receiving text messages, there is the added benefit of a saved reminder that can be accessed again, a benefit not always available with phone call reminders. A 2013 Cochrane Review showed low to moderate quality evidence that text message reminders improve appointment attendance rates [2]. A recent BMJ article published in 2016 also showed that text-based electronic notifications improved adherence [3]. However, the studies included in these reviews were heterogeneous, encompassing different interests. The scope of research on text message studies is broad and continues to expand. Studies have been done in various subspecialty areas spanning from dental clinics to endoscopy clinics, and the outcomes studied include appointment adherence [4], medication adherence [5], vaccination rates [6], and chronic disease care [7]. Many of the results have been positive. For example, text messages improved attendance in adolescents attending outpatient mental health treatment [4]. A systematic review also showed that text messages improved patients' medication adherence, especially in those with chronic diseases such as HIV, asthma, diabetes, and heart disease [5]. However, there are also studies where text messages did not demonstrate a positive impact [8, 9]. Pediatric data regarding text messages is still limited. However, there is an emerging body of literature focusing on issues pertaining to the adolescent population, such as follow-up for sexually transmitted illnesses [10, 11], vaccination adherence [12, 13], and contraception adherence [14]. Still, there are not many studies focusing on text messages for pediatric appointment adherence. To date, there are no published studies regarding whether text messages are advantageous in an inner-city resident clinic population. With high no-show rates and lack of continuity, resident clinics could substantially benefit from an intervention, such as a text messages, that would improve appointment adherence.

nce. To date, there are no published studies regarding whether text messages are advantageous in an inner-city resident clinic population. With high no-show rates and lack of continuity, resident clinics could substantially benefit from an intervention, such as a text messages, that would improve appointment adherence. Our resident clinic services mostly low-income families that are predominantly African American. There are no previous studies reporting the relationship between text message reminders and no-show rate in a predominantly African American population. Preliminary data from our institution indicated that more than 95% of families at the clinic have cell phones, 75% of families use text messaging daily, and 85% have unlimited text messages, suggesting that a text messaging reminder system would be feasible for our patient population. The survey also showed that while 32% preferred only a phone call reminder and 20% preferred just a text message reminder, 45% preferred both a text message and a phone call reminder. Based on the lack of text message studies performed with our patient demographics and our preliminary findings, this randomized controlled trial was initiated to study the hypothesis that a text message reminder would improve appointment adherence at our primary care pediatric practice.

ssage and a phone call reminder. Based on the lack of text message studies performed with our patient demographics and our preliminary findings, this randomized controlled trial was initiated to study the hypothesis that a text message reminder would improve appointment adherence at our primary care pediatric practice. 2. Methods 2.1. Experimental Design 2.1.1. Study Type A randomized controlled trial was designed to test the impact of text reminders on attendance adherence at a pediatric resident clinic servicing an inner-city population. Patients who scheduled an appointment were randomized to receive text or not receive text, in addition to the standard clinic phone message, at a 1 : 1 allocation ratio using a simple randomization method.

mpact of text reminders on attendance adherence at a pediatric resident clinic servicing an inner-city population. Patients who scheduled an appointment were randomized to receive text or not receive text, in addition to the standard clinic phone message, at a 1 : 1 allocation ratio using a simple randomization method. 2.1.2. Participants This study was approved by the hospital's Institutional Review Board. Patients and guardians were enrolled from a single, urban, pediatric primary care clinic staffed by residents at an academic center which serves an inner-city population. Data was collected on a rolling basis over a period of seven months from August 2014 to February 2015. All patients scheduled in the resident clinic were considered eligible except for those without cell phones and those whose projected subsequent appointment fell outside the planned period of data collection. Hence, only those whose next appointment fell within the seven-month period were included in the study. Guardians were either approached in the waiting room or referred to the researchers by physicians. Patients over age 18 consented for themselves, while guardians consented for all patients under age 16. For patients between ages 16–18, guardians consented but patients gave assent and could choose to receive a personal text message appointment reminder in addition to the one sent to their guardian.

ers by physicians. Patients over age 18 consented for themselves, while guardians consented for all patients under age 16. For patients between ages 16–18, guardians consented but patients gave assent and could choose to receive a personal text message appointment reminder in addition to the one sent to their guardian. 2.1.3. Survey Once enrolled, participants filled out a survey. Survey questions included patient's name, birthday, age, gender, and race, participant's relationship to the patient, if their cell phone receives text messages, cell phone number, purpose of the current visit, projected time of next visit, purpose of the next visit, and highest level of education of the participant completing the form. Survey was conducted as a multiple choice questionnaire to standardize answers. There were no exclusions for the visit type at the time of enrollment, and there were no exclusions for the next scheduled appointment visit type. Visit types included well visits or immunization visits, sick visits, follow-up visits, and others. Sick visit reasons were colds, fevers, vomiting, diarrhea, rashes, and others. Follow-up reasons included newborn weight checks, neonatal bilirubin checks, and birth contraception, such as Depo-Provera injections.

type. Visit types included well visits or immunization visits, sick visits, follow-up visits, and others. Sick visit reasons were colds, fevers, vomiting, diarrhea, rashes, and others. Follow-up reasons included newborn weight checks, neonatal bilirubin checks, and birth contraception, such as Depo-Provera injections. 2.1.4. Intervention On a weekly basis, the clinic schedule was surveyed to determine the scheduled appointment date for the participants. Preliminary survey data demonstrated that parents prefer texts within 3 days of the appointment. Hence, all the participants randomized to the intervention group received a text within 3 days of the appointment. The text message was sent from a private number. The Institutional Review Board and Office of Research Compliance did not approve the use of computer automated texting programs due to HIPAA concerns. Therefore, each text message was manually sent by the researchers performing the study from a prepaid cell phone with a unique number used only for the purposes of this study. The text messages were standardized to include the name of the pediatric practice along with the date and time of the appointment. The patient's name and other identifying data were not included in the text message. Recipients were asked to not reply to the sender, but the number of the pediatric practice was provided for patients to reschedule if needed.

o include the name of the pediatric practice along with the date and time of the appointment. The patient's name and other identifying data were not included in the text message. Recipients were asked to not reply to the sender, but the number of the pediatric practice was provided for patients to reschedule if needed. 2.1.5. Outcome Charts were periodically reviewed to determine which patients had arrived for their scheduled appointments. Data was also collected on rescheduled or canceled appointments in both groups. Our primary end-point was whether patients were no-show for their appointments. 2.2. Statistical Methods 2.2.1. Sample Size Determination Sample size was estimated based on internal institutional data from the previous year's no-show rate, which was 30%. We aimed to determine whether the no-show rate could be reduced to 15% by adding text reminders. Assuming a no-show rate of 30% in the group without text, a sample size of 160 (80 in each group) would achieve 90% power to detect a difference of 15% based on the two-sided Z-test with pooled variance and 5% significance level. To account for no cell phone ownership, no text message capacity in cell phone, or declined participation, we aimed to recruit approximately 200 patients.

le size of 160 (80 in each group) would achieve 90% power to detect a difference of 15% based on the two-sided Z-test with pooled variance and 5% significance level. To account for no cell phone ownership, no text message capacity in cell phone, or declined participation, we aimed to recruit approximately 200 patients. 2.2.2. Randomization Of the 196 parents or patients approached, 95% (186) agreed to participate (Figure 1). Those that declined to participate in the research study or stated their preference for only voice message reminders were immediately excluded. Of those assessed for eligibility, 17 (8.7%) were excluded. Ten participants did not meet inclusion criteria because they either did not own a cell phone or had a phone that could not receive text messages. One excluded patient planned to switch practices. Six excluded participants refused to provide their cell phone numbers for research purposes due to concerns for privacy. After exclusion, 169 were eligible, enrolled, and randomized to either the control or intervention group using simple randomization method. Patients were randomized on a rolling basis. The control group (n = 84) did not receive a text message reminder and only received the standard voice message reminder. The experimental group (n = 85) received a text message reminder in addition to the standard voice message reminder no more than three days prior to the appointment.

e randomized on a rolling basis. The control group (n = 84) did not receive a text message reminder and only received the standard voice message reminder. The experimental group (n = 85) received a text message reminder in addition to the standard voice message reminder no more than three days prior to the appointment. 2.2.3. Statistical Analysis Caregiver's demographics and child's age between the two randomized groups were first compared using Chi-squared test, t-test, and Wilcoxon rank test to ensure the randomization was properly performed. Patients who did not show up for their appointment and did not cancel or reschedule the appointment were identified as “no-show.” Overall no-show rate was determined, and the rate between text sent and no text sent was compared using Chi-squared test. Child's age, reason for the visit, and caregiver's demographics were also compared between the two groups using Wilcoxon two-sample test or Chi-squared test. Child's age was dichotomized to less than 7 months of age and greater than or equal to 7 months of age, based on the age of frequent visits for well-child checks and immunizations. A multivariate logistic regression analysis was performed to estimate adjusted OR (95% CI) of kept/canceled/rescheduled visit by controlling for child's age, caregiver's education, and whether caregiver who received the text was mother or another guardian. All analyses were performed using SAS version 9.4. Two-sided p values are presented, and a p value less than 0.05 is considered statistically significant.

of kept/canceled/rescheduled visit by controlling for child's age, caregiver's education, and whether caregiver who received the text was mother or another guardian. All analyses were performed using SAS version 9.4. Two-sided p values are presented, and a p value less than 0.05 is considered statistically significant. 3. Results To validate the randomization, caregiver's and child's characteristics of the control and intervention groups were examined. At baseline, there were no significant differences between the demographics of the control and intervention groups, including sex and race of the patient, median age of the patient, caregiver providing consent, and education level of the caregiver (data not shown). The overall no-show rate during the period of data collection was 30.8%. The no-show rate was significantly lower in those who received text reminders with the standard phone message compared to those who only received phone message reminders (23.5% versus 38.1%, p = 0.04) (Figure 2).

3. Results To validate the randomization, caregiver's and child's characteristics of the control and intervention groups were examined. At baseline, there were no significant differences between the demographics of the control and intervention groups, including sex and race of the patient, median age of the patient, caregiver providing consent, and education level of the caregiver (data not shown). The overall no-show rate during the period of data collection was 30.8%. The no-show rate was significantly lower in those who received text reminders with the standard phone message compared to those who only received phone message reminders (23.5% versus 38.1%, p = 0.04) (Figure 2). In addition to text message intervention, associations between other factors and no-show were also determined (Table 1). One set of parameters examined was regarding the caregiver, including who consented to the study and their education level. In each of these cases, there was no significant association with the no-show population. Another set of parameters examined was related to the patient, including race, gender, age, and whether they were under 7 months old. Patient's age was of interest because children younger than 7 months old have more frequent visits. These factors were also not predictive of no-show events. The reason for the patient's visit at the time of enrollment and reason for the next appointment were not predictive of no-show events. On the other hand, text message intervention did in fact have a significant association in this univariate analysis.

t visits. These factors were also not predictive of no-show events. The reason for the patient's visit at the time of enrollment and reason for the next appointment were not predictive of no-show events. On the other hand, text message intervention did in fact have a significant association in this univariate analysis. Although not statistically significant, trends were noted when comparing the median patient age of the no-show group and other group (Table 1; 7.6 mo versus 5.3 mo, p = 0.29), the percentage of caregivers with some college education or higher (32.7% versus 45.3%, p = 0.12), and whether consent was obtained from the mother (90.4% versus 83.8%, p = 0.25). Hence, a multivariate logistic regression analysis was performed to determine whether the impact of text message reminders would remain significant after controlling for these factors. Analysis showed that when adjusted for age, caregiver education, and caregiver who gave consent, the intervention group was two times more likely (OR 2.12, CI 1.06–4.21) to keep, cancel, or reschedule the appointment (Table 2). Therefore, text message reminders can serve as an effective means to decrease pediatric patient no-show rates among low-income populations.

egiver education, and caregiver who gave consent, the intervention group was two times more likely (OR 2.12, CI 1.06–4.21) to keep, cancel, or reschedule the appointment (Table 2). Therefore, text message reminders can serve as an effective means to decrease pediatric patient no-show rates among low-income populations. 4. Discussion Our results indicate that sending text message reminders is an effective means to improve appointment adherence at a pediatric resident clinic in an urban setting. Based on our pilot survey, where 33.8% of parents reported to have forgotten their scheduled visit, the text messages could serve as a valuable prompt for keeping the scheduled appointment. This is consistent with previous literature demonstrating the utility of text message appointment reminders on different patient populations [3, 4].

survey, where 33.8% of parents reported to have forgotten their scheduled visit, the text messages could serve as a valuable prompt for keeping the scheduled appointment. This is consistent with previous literature demonstrating the utility of text message appointment reminders on different patient populations [3, 4]. Our study is unique in that it focuses on a pediatric, urban, low-income population comprised mostly of African Americans. A literature search for similar studies with low-income populations largely points to studies done in developing countries [15], while many urban-based studies focus on issues such as HIV [16] and diabetes [17]. Few studies have looked at the impact of text messages on appointment adherence in an urban setting. There are even fewer studies that focus on pediatric populations in this setting. One study examined whether text message reminders improved rates of second dose of influenza vaccination in an urban clinic servicing mostly Latinos aged 6 months through 8 years, and it showed that the low-income, urban, minority population responded positively to text messages [18]. Many of the pediatric studies focus on adolescents, including one study utilizing text messaging for Depo-Provera injection reminders for urban adolescents and young adults, which also showed a positive effect with a text message reminder [14]. Our population is unique due to the relatively young age. Our findings add to the body of literature supporting the use of text message for pediatric patients in an urban setting.

o-Provera injection reminders for urban adolescents and young adults, which also showed a positive effect with a text message reminder [14]. Our population is unique due to the relatively young age. Our findings add to the body of literature supporting the use of text message for pediatric patients in an urban setting. To our knowledge, there are no studies done examining text message reminders in an inner-city resident clinic setting, where residents provide direct patient care under the supervision of faculty doctors. Continuity of care is challenging to achieve at a resident clinic [19]. We show here that sending a text message reminder can help increase the show rate and likely will improve continuity at the clinic. Our study results contrast with a similar study done in a dental school pediatric dentistry clinic, where text messaging did not improve appointment adherence [20]. This shows that more research needs to be done in this unique setting.

er can help increase the show rate and likely will improve continuity at the clinic. Our study results contrast with a similar study done in a dental school pediatric dentistry clinic, where text messaging did not improve appointment adherence [20]. This shows that more research needs to be done in this unique setting. No-show can be a major burden to health care [21]. Missed appointments can lead to missed opportunities for disease detection and, in pediatrics, delayed vaccinations or recognition of developmental delays. Delayed care can potentially worsen outcomes for patients and increase health care costs for the government. A missed appointment also represents a financial burden to the clinic. A study performed at the VA Medical Center estimated that the average cost of no-show per patient was $196 in 2008 [21]. Financially, providing a text message is a promising and worthwhile way to reduce no-show rate. It is low cost and can be easily performed now with many automated systems on the market. However, the clinic must work with their legal system to ensure that the messages are delivered in a HIPAA compliant manner.

008 [21]. Financially, providing a text message is a promising and worthwhile way to reduce no-show rate. It is low cost and can be easily performed now with many automated systems on the market. However, the clinic must work with their legal system to ensure that the messages are delivered in a HIPAA compliant manner. This study was performed in a clinic that serves mostly Medicaid patients. The results of this study are applicable to similar populations but may not be generalized to other populations or clinics with different demographics. This study was also limited by the nature of a resident clinic. Given that schedules cannot be made far in advance, this study naturally selected for infants and toddlers, who require more frequent visits. Another limitation is the lack of confirmation of text message receipt. The study did not allow for subjects to reply to the text message. Phone numbers change frequently; hence it could not be ascertained how many participants in the experimental group received the text message sent to them. A study looking at communication with adolescents via text messages in the context of Depo-Provera injections showed that phone number changes, unpaid cell phone bills, and phone loss significantly contribute to adolescent text nonresponsiveness rate [22].

the experimental group received the text message sent to them. A study looking at communication with adolescents via text messages in the context of Depo-Provera injections showed that phone number changes, unpaid cell phone bills, and phone loss significantly contribute to adolescent text nonresponsiveness rate [22]. Even with the addition of a text message, 24% of our patients still did not attend their appointment. Future research will be focused on this population. Another questionnaire via phone call can be administered to this population. Questions would include whether the text message was received, whether transportation was an issue, and so forth. Given the lower socioeconomic status of the study population, further research can delve into how limited resources play a role in show rates, so that the clinic can target ways to better serve the patient population. Potential future research can also look at parental response to the text message reminder and if frequent text messages would lead to eventual desensitization towards this type of reminder. 5. Conclusion Text message reminders are an effective way to improve rate of attendance at an urban pediatric resident clinic with high no-show rates. This promising and modern technology represents an underutilized resource in primary care that will improve appointment adherence and potentially enhance patient care.

Even with the addition of a text message, 24% of our patients still did not attend their appointment. Future research will be focused on this population. Another questionnaire via phone call can be administered to this population. Questions would include whether the text message was received, whether transportation was an issue, and so forth. Given the lower socioeconomic status of the study population, further research can delve into how limited resources play a role in show rates, so that the clinic can target ways to better serve the patient population. Potential future research can also look at parental response to the text message reminder and if frequent text messages would lead to eventual desensitization towards this type of reminder. 5. Conclusion Text message reminders are an effective way to improve rate of attendance at an urban pediatric resident clinic with high no-show rates. This promising and modern technology represents an underutilized resource in primary care that will improve appointment adherence and potentially enhance patient care. Acknowledgments The authors would like to thank the Rainbow Ambulatory Practice, particularly the pediatric residents, for helping to identify and recruit patients for this study. The authors would also like to thank Rena Chandra for data from the pilot study. Competing Interests The authors declare that there are no competing interests regarding the publication of this paper. Figure 1 Flow chart of randomization. Figure 2 No-show rate for text sent and not sent. Table 1 Comparison of characteristics in outcome groups.

Acknowledgments The authors would like to thank the Rainbow Ambulatory Practice, particularly the pediatric residents, for helping to identify and recruit patients for this study. The authors would also like to thank Rena Chandra for data from the pilot study. Competing Interests The authors declare that there are no competing interests regarding the publication of this paper. Figure 1 Flow chart of randomization. Figure 2 No-show rate for text sent and not sent. Table 1 Comparison of characteristics in outcome groups. Factor No-show (n = 52) Other ∗ (n = 117) p value Caregiver parameters Consent from mother (%) 47 (90.4) 98 (83.8) 0.25 College education or higher (%) 17 (32.7) 53 (45.3) 0.12 Child parameters Race, African American (%) 49 (94.2) 112 (95.7) 0.67 Gender, male (%) 26 (50.0) 52 (44.4) 0.50 Age (median, IQR) 7.6 (1.6, 23.8) 5.3 (1.6, 12.0) 0.29 Patients < 7 months old (%) 26 (50.0) 65 (55.6) 0.50 Reason for visit at study enrollment Well visit (%) 26 (50) 61 (52.1) 0.80 Sick visit (%) 14 (26.9) 25 (21.4) 0.43 Follow-up visit (%) 12 (23.1) 31 (26.5) 0.64 Reason for visit of interest Well visit (%) 34 (65.4) 81 (69.2) 0.62 Follow-up visit (%) 18 (34.6) 34 (29.1) 0.47 Intervention Text message (%) 20 (38.5) 65 (55.6) 0.04 ∗Show, cancelled, or rescheduled. Table 2 Factors associated with reduced no-show probability. Unadjusted Adjusted ∗ OR (95% CI) p value OR (95% CI) p value Randomization Text sent 2.00 (1.03, 3.9) 0.04 2.12 (1.06, 4.21) 0.03 No text sent 1 1 Child's age (months) Less than 7 1.25 (0.65, 2.41) 0.5 1.31 (0.66, 2.60) 0.44 7 or more 1 1 Education

Text message (%) 20 (38.5) 65 (55.6) 0.04 ∗Show, cancelled, or rescheduled. Table 2 Factors associated with reduced no-show probability. Unadjusted Adjusted ∗ OR (95% CI) p value OR (95% CI) p value Randomization Text sent 2.00 (1.03, 3.9) 0.04 2.12 (1.06, 4.21) 0.03 No text sent 1 1 Child's age (months) Less than 7 1.25 (0.65, 2.41) 0.5 1.31 (0.66, 2.60) 0.44 7 or more 1 1 Education College or higher 1.71 (0.86, 3.38) 0.13 1.70 (0.84, 3.43) 0.14 Lower than college 1 1 Caregiver Mother 1.82 (0.64, 5.18) 0.26 2.19 (0.73, 6.55) 0.16 Other 1 1 ∗Adjusted factors listed in the same table.

1. Background Neurocysticercosis (NCC), the commonest parasitic infection of the nervous system, is also the most common cause of acquired epilepsy in childhood [1]. In a systematic review, 29% of epileptics were found to have NCC [2]. NCC is endemic in most developing countries of Asia, Latin America, and Central and South Africa [3]. A seroprevalence study from Chandigarh, India, reported anticysticercus antibodies in 17.3% [4] while seroprevalence among the healthy blood donors from Pondicherry, India, was 6.5% using both antigen and antibody detection methods [5]. In a WHO study conducted in rural pig-farming community of UP, India, the prevalence of Taeniasis was 18.6% and epilepsy was found in 5.8% and 48.3% of them had NCC [6]. There is no single therapeutic approach to NCC. The approach depends upon the number and location of lesions, cysts' viability, and host's immune response to the parasite [7]. A combination of symptomatic and cysticidal therapy is usually used. Of the later, albendazole is superior to praziquantel. The use of cysticidal therapy has been questioned, as it destroys the cysts but do not modify the disease course [8]. Based on the available literature, the American Academy of Neurology advised the use of cysticidal therapy with albendazole in cases of NCC, as it was found to be associated with both reduction of long-term seizure and resolution of the lesions [9].

ed, as it destroys the cysts but do not modify the disease course [8]. Based on the available literature, the American Academy of Neurology advised the use of cysticidal therapy with albendazole in cases of NCC, as it was found to be associated with both reduction of long-term seizure and resolution of the lesions [9]. With this background, we conducted a prospective, observational study to find prevalence of NCC, its response to cysticidal therapy, and adverse effects of therapy and also to find the risk factors for persistent seizure in Indian children with NCC.

ed, as it destroys the cysts but do not modify the disease course [8]. Based on the available literature, the American Academy of Neurology advised the use of cysticidal therapy with albendazole in cases of NCC, as it was found to be associated with both reduction of long-term seizure and resolution of the lesions [9]. With this background, we conducted a prospective, observational study to find prevalence of NCC, its response to cysticidal therapy, and adverse effects of therapy and also to find the risk factors for persistent seizure in Indian children with NCC. 2. Methods Children between 1 year to 14 years of age visiting the pediatric outpatient department (OPD) and pediatric emergency of a tertiary care hospital in Northern India over 2 years period (May 2012 to August 2014) with first episode of seizure or acute focal neurological deficit were screened. Ethical clearance was obtained from the Institute Ethics Committee. A written informed consent was taken from either of the parents/legal guardians. Patients were excluded if they had fever, known neurological illness, or epilepsy with known etiology (e.g., structural brain anomaly), were critically sick (respiratory failure, cardiovascular instability), and already received antihelminthic (albendazole or praziquantel) or steroid therapy. Patients who were found to have an alternate diagnosis or only calcified NCC or ocular cysticercosis after enrolment were also excluded. Therefore, cases of NCC in vesicular stage, colloid stage, and granular nodular stage, either parenchymal or extraparenchymal, were finally included. Patients' demographic data, presenting illness, past history, family history of seizure/similar illness, dietary history, treatment history, history of exposure to animals, household contact with NCC, and socioeconomic status were recorded. General physical and detailed nervous system examination including direct ophthalmoscopy was done for all the patients.

llness, past history, family history of seizure/similar illness, dietary history, treatment history, history of exposure to animals, household contact with NCC, and socioeconomic status were recorded. General physical and detailed nervous system examination including direct ophthalmoscopy was done for all the patients. A complete blood count, liver and renal function tests, mantoux test, chest X-ray, contrast enhanced computed tomography (CT) scan of brain were obtained in all the patients. Magnetic resonance imaging (MRI) of brain (in cases with normal or doubtful CT scan findings), cerebrospinal fluid (CSF) examination, stool microscopic examination for evidence of Taenia solium, Taenia solium antibody in serum and CSF (ELISA kit; UB-Magiwell Cysticercosis Kit™), and EEG were done. Diagnosis was based on clinical picture and radiological findings highly suggestive of neurocysticercosis.

indings), cerebrospinal fluid (CSF) examination, stool microscopic examination for evidence of Taenia solium, Taenia solium antibody in serum and CSF (ELISA kit; UB-Magiwell Cysticercosis Kit™), and EEG were done. Diagnosis was based on clinical picture and radiological findings highly suggestive of neurocysticercosis. All the children were treated with albendazole at a dose of 15 mg/kg/day (maximum 800 mg/day) in 2 divided doses for 28 days along with dexamethasone at a dose of 0.15 mg/kg/dose (maximum 4 mg/dose) every six hours (started 2 days before albendazole and continued for 3 days along with albendazole) for a total duration of 5 days, irrespective of the number of NCC. Children also received antiepileptic for seizure control. Children were followed up at 1-week, 2-week, 1-month, 3-month, 6-month, and 1-year period after starting the cysticidal therapy. At each followup, frequency of seizure occurrence, symptomatic improvement/worsening of symptoms (other associated features like raised intracranial tension, focal neurological deficits, encephalopathy, and visual problems), and adverse drug reactions (vomiting, abdominal pain, rash, and jaundice) were noted. Laboratory tests were repeated at 1 week, 2 weeks, and the end of therapy (i.e., at 1 month). Various presenting features and laboratory parameters between children who had a persistence/recurrence of seizure at 1 year of followup and those who did not have seizure at 1 year were compared in an attempt to find out the possible risk factors. All the data were entered in a predesigned performa and managed in Microsoft excel spreadsheet. The data was expressed as number and percentage (for categorical variables) or mean with standard deviation (for continuous variable). Statistical analysis was done using Stata 11 software (Statacorp, TX, USA). Student's t test and Chi square test were used for determining the P value (<0.05 was taken as significant).

The data was expressed as number and percentage (for categorical variables) or mean with standard deviation (for continuous variable). Statistical analysis was done using Stata 11 software (Statacorp, TX, USA). Student's t test and Chi square test were used for determining the P value (<0.05 was taken as significant). 3. Results A total of 3006 patients were screened. Of 169 initially included, 35 turned out to be tuberculoma and 14 had calcified cysts, so they were further excluded. Finally 120 cases of NCC in vesicular stage, colloid stage, and granular nodular stage, either parenchymal or extra-parenchymal according to neuroimaging, were included for treatment and further followup (Figure 1).

lly included, 35 turned out to be tuberculoma and 14 had calcified cysts, so they were further excluded. Finally 120 cases of NCC in vesicular stage, colloid stage, and granular nodular stage, either parenchymal or extra-parenchymal according to neuroimaging, were included for treatment and further followup (Figure 1). The baseline characteristics of these children have been described (Table 1). Most of the patients were in the 10–14 years of age group, and around 56% were male. Commonest presenting complaint was seizure with generalized seizure in 92 cases (86.7%). Commonest type of lesion observed on neuroimaging was a solitary lesion in brain (67.5%), while none of the children had >20 lesions. In majority (85%) of the cases, lesions were located in the brain parenchyma only, and rests were having lesions both in brain parenchyma and ventricles. None of the cases were noted to have isolated ventricular or cisternal lesions (Table 2). Amongst the parenchymal lesions, parietal lobe was the most common site of involvement (79%), either isolated or along with other site of involvement followed by frontal lobe in around 15% cases. No children were noted to have associated extracranial cysticercosis. Thirty-nine children had “multiple lesion” NCC; 30 (76.9%) of them had raised ICT, and 4 (10.3%) had focal neurological deficits. All the 18 cases with intraventricular NCC had more than 5 NCC lesions with 10 cases having 5–10 lesions and 8 cases with >10 lesions.

re noted to have associated extracranial cysticercosis. Thirty-nine children had “multiple lesion” NCC; 30 (76.9%) of them had raised ICT, and 4 (10.3%) had focal neurological deficits. All the 18 cases with intraventricular NCC had more than 5 NCC lesions with 10 cases having 5–10 lesions and 8 cases with >10 lesions. Serum Taenia solium antibodies were tested in 43 children and a positivity rate of 65.1% was observed (Table 3). CSF antibodies against T solium was done in 20 children only but it revealed a very high positivity (90%) (Table 4). Therapy, treatment outcome, appearance of new symptoms, and side effects of therapy in these children on followup has been described (Table 5). Recurrence of seizure was observed in 28 children over 1-year period, and only 5 children had persistent seizure at 6 months and also at 1-year followup. Repeat CT scan revealed calcified lesions (data not shown). The antiepileptic drugs were continued in these cases. Worsening of preexisting symptoms was mostly in the form of worsening preexisting raised intracranial pressure and headache. One child developed pseudoseizure after 6 months of therapy. The number of children who had developed new symptoms attributable to NCC (e.g., headache, features of raised intracranial tension, focal neurological deficits, encephalopathy, and visual problems) during 1 year of followup was very small. Regarding the adverse events, 2 cases developed new onset raised intracranial tension after 1 week and 3 cases after 2 weeks, respectively; cysticidal therapy was stopped in these 5 cases. Eleven cases developed gastrointestinal symptoms (abdominal pain, nausea, and vomiting) during the course of albendazole therapy. They were managed conservatively and did not require stopping of therapy. Three children developed elevation of transaminases (>2 times of upper limit of normal) after 1 week of therapy requiring temporary stoppage of therapy; they normalized over next 2 weeks. Neutropenia occurred in 2 cases at 2 weeks and in 6 cases at 4 weeks requiring temporary stoppage of albendazole therapy. They were also followed up with serial blood counts and improved over next 2 weeks.

mit of normal) after 1 week of therapy requiring temporary stoppage of therapy; they normalized over next 2 weeks. Neutropenia occurred in 2 cases at 2 weeks and in 6 cases at 4 weeks requiring temporary stoppage of albendazole therapy. They were also followed up with serial blood counts and improved over next 2 weeks. While trying to find out any risk factors that could account for persistence of seizure at 1 year, various parameters were compared between children having persistence of seizure at 1 year and those in whom seizure subsided (Table 6). But none of the compared variables were found to be statistically significant between the two groups. 4. Discussion In this hospital based, prospective study, the prevalence of NCC was 4.5% in children presenting with acute onset seizure/focal neurological deficit. The 10–14-year age group was most commonly affected. Similar findings are reported in previous studies as well [6, 10]. This could be explained by the fact that children share the family diet usually late, and though symptoms may occur as early as months after primary infestation, it usually occurs 2–6 years after exposure [11]. No predilection of sex could be identified. NCC was found to be more prevalent in the children belonging to lower socioeconomic strata. This could be explained by poorer standards of living and lack of hygiene that goes hand in hand with poverty. Another postulate is that, with pork being a cheaper meat, it is more consumed by poor, and eating pork has been found to be a risk factor for the disease [6].

children belonging to lower socioeconomic strata. This could be explained by poorer standards of living and lack of hygiene that goes hand in hand with poverty. Another postulate is that, with pork being a cheaper meat, it is more consumed by poor, and eating pork has been found to be a risk factor for the disease [6]. Seizure was the commonest (86.7%) presenting feature with generalized seizure (76.7%) being more common than focal. This may seem to be in contrast to the common belief and also most of the previously reported studies from India and abroad [12–14]. But two large case series published recently from Nepal [15] and India [16] cited incidences of generalized seizure at presentations 52% and 65%, respectively. But this should be interpreted cautiously, as the chances of focal seizure leading to secondary generalization being interpreted as GTCS cannot be ruled out as the seizure seminology was discerned from parental history and the onset may have been missed by them. Features suggestive of raised intracranial pressure and focal neurological deficit are less common in children compared to adults. Our figures of 28.3% and 4% are similar to the 30% and 8% reported by another study [12].

seizure seminology was discerned from parental history and the onset may have been missed by them. Features suggestive of raised intracranial pressure and focal neurological deficit are less common in children compared to adults. Our figures of 28.3% and 4% are similar to the 30% and 8% reported by another study [12]. Around 11% children had peripheral blood eosinophilia (absolute eosinophil count >500/mm3); therefore, it was of little benefit in the diagnosis. Only 5.8% children with NCC were found to be excreting ova of Taenia solium in their stool. The stool positivity rates for T solium were always found to be very scarce in previous pediatric studies [12, 15], but interestingly, a recent study from India [17] has reported a very high rate (44.7%) of isolation of T solium ova in stool. CSF examination was normal in most (45%) cases. The commonest abnormal finding was noted to be pleocytosis followed by raised protein. These findings were also partially congruent with previously conducted pediatric and adult studies [11, 18]. Few (13.3%) cases showed CSF eosinophilia which is considered a suggestive finding towards the diagnosis of CNS helminthic infection though not specific of neurocysticercosis [19]. An abnormal EEG was observed in 56% of cases, the commonest abnormality being sharp wave and spikes followed by nonspecific background abnormalities. However, EEG has minor relation to the symptoms and CT lesions in NCC [20].

ards the diagnosis of CNS helminthic infection though not specific of neurocysticercosis [19]. An abnormal EEG was observed in 56% of cases, the commonest abnormality being sharp wave and spikes followed by nonspecific background abnormalities. However, EEG has minor relation to the symptoms and CT lesions in NCC [20]. Commonest radiological pattern of NCC was a solitary lesion (67.5%). Parietal lobe was the most common site of involvement (79%) followed by frontal lobe (15%). Commonest site of involvement was brain parenchyma, either alone or in combination with ventricle. These are consistent with findings from other studies [12, 15–18]. Only 65% of 40 cases were found positive for serum antibody in our study, whereas CSF antibody positivity rates of 90% were observed in 20 children tested for the same. Serum ELISA has always been associated with poor sensitivity [1]. Other possible explanations are that false negative serum antibody results by ELISA may be higher in patients with only parenchymatous disease, especially with single lesions [1]. False negative results in serum with positive CSF serology may also occur because of local production of antibodies without a parallel increase in titres in the peripheral blood [21]. On the other hand, false positive results in immune tests are sometimes seen in individuals from endemic areas because of contact with the parasite without development of the disease [1].

erology may also occur because of local production of antibodies without a parallel increase in titres in the peripheral blood [21]. On the other hand, false positive results in immune tests are sometimes seen in individuals from endemic areas because of contact with the parasite without development of the disease [1]. Seizure recurrence was observed in 6 (5%), 5 (4.2%), and 5 (4.2%) cases at 3-month, 6-month, and 1-year followup, respectively. These rates have been variable being high (10% and 13% at 3 months and 6 months, resp.) [22] and low (4.8% at 1 year) [23]. On the other hand, a more recent study from India reported a very high rate of seizure recurrence (23.7%) at 2-year followup [17]. None of our children had persistence of any other symptoms attributable to NCC at the end of 1-year followup. Only one child developed pseudoseizure, which would be unrelated to NCC.

[23]. On the other hand, a more recent study from India reported a very high rate of seizure recurrence (23.7%) at 2-year followup [17]. None of our children had persistence of any other symptoms attributable to NCC at the end of 1-year followup. Only one child developed pseudoseizure, which would be unrelated to NCC. A follow-up CT scan was performed only in those with recurrence of seizure (28), worsening of preexisting symptoms (13), or new neurological symptoms (2). Among the children with recurrence of seizure (12) or worsening of preexisting symptoms (11) or development of new symptoms (1) during therapy 5 children were found to have raised ICP. Notably all of them had 2–5 parenchymal NCC (no intraventricular NCC) and only had evidence of perilesional edema with no evidence of hydrocephalus. All of them responded well to short course (7 days) of dexamethasone at a dose of 0.15 mg/kg/dose q 6 hourly along with osmotherapy (mannitol and/or hypertonic saline) and other supportive measures with temporary cessation of cysticidal therapy.

only had evidence of perilesional edema with no evidence of hydrocephalus. All of them responded well to short course (7 days) of dexamethasone at a dose of 0.15 mg/kg/dose q 6 hourly along with osmotherapy (mannitol and/or hypertonic saline) and other supportive measures with temporary cessation of cysticidal therapy. All the patients in whom a CT scan was performed at 3 months and beyond showed complete calcification of NCC. Calcification in NCC is a common long-term outcome with observed rates of calcification in pediatric NCC varying widely from 6.2% to as high as 60% [24]. Interestingly, cysticidal therapy was found to have no significant effect in the calcification of solitary parenchymal NCC [25, 26]. A retrospective analysis of solitary parenchymal NCC from India [24] reported that the risk of breakthrough seizures was significantly higher in calcified lesions on follow-up scan when compared to normal scans or persisting lesions. The same has been echoed in another prospective study involving both children and adults [27].

pective analysis of solitary parenchymal NCC from India [24] reported that the risk of breakthrough seizures was significantly higher in calcified lesions on follow-up scan when compared to normal scans or persisting lesions. The same has been echoed in another prospective study involving both children and adults [27]. There are very few pediatric studies which document adverse effects of albendazole therapy other than neurological symptoms [24]. Gastrointestinal, hepatic, and hematological side effects of albendazole are well known, and guidelines for monitoring of patients on long-term therapy advise laboratory monitoring twice weekly [28]. In our study, gastrointestinal side effects were observed in 11 (9.2%) children, elevated transaminases in 3 (2.5%), and neutropenia in 8 (6.7%) requiring temporary cessation of therapy. Both hepatic and hematological abnormalities normalized on followup within few weeks. Variable gastrointestinal side effects have been observed in various studies (from 6% to 26%) [29, 30].

d in 11 (9.2%) children, elevated transaminases in 3 (2.5%), and neutropenia in 8 (6.7%) requiring temporary cessation of therapy. Both hepatic and hematological abnormalities normalized on followup within few weeks. Variable gastrointestinal side effects have been observed in various studies (from 6% to 26%) [29, 30]. Sharma et al. [27] found that apart from calcification of the lesions family history of seizures, serial seizures, and presence of EEG abnormalities is to be associated with increased risk of recurrence of seizures. While trying to find out risk factors for persistence of seizure at 1 year in our population, none of the perceived risk factors (e.g., mean age at onset of seizure, partial seizure, >5 lesions in brain, and EEG abnormality) were found to be important. Similar findings were observed by others [17, 23], who also failed to find any risk factors for persistence of seizures in children with NCC.

in our population, none of the perceived risk factors (e.g., mean age at onset of seizure, partial seizure, >5 lesions in brain, and EEG abnormality) were found to be important. Similar findings were observed by others [17, 23], who also failed to find any risk factors for persistence of seizures in children with NCC. The present study has many strengths: longitudinal design, robust methodology, and prolonged clinical and laboratory followup of the cases after therapy to look for response as well as adverse effects which is rarely described in previous pediatric studies. Following are the limitations. First, neuroimaging and EEG were not repeated in the children after treatment in the followup. It would have revealed response to antihelminthic therapy and also helped in deciding on the duration of antiepileptic. Then again, repeating a CT scan definitely increases radiation exposure to the developing brain of these children. MRI though does not involve radiation; it is associated with need for prolonged sedation or general anesthesia and the associated risk of the same. The cost of the above mentioned procedures is also to be kept in mind in the context of a resource poor setting like India. Second, being a tertiary care hospital based study, the participants may not be representative of the population intended. 5. Conclusion Present study shows that the response to cysticidal therapy is very good in NCC with infrequent adverse effects. Competing Interests The authors declare that they have no competing interests. Figure 1 Flow diagram showing recruitment of cases into the study.

The present study has many strengths: longitudinal design, robust methodology, and prolonged clinical and laboratory followup of the cases after therapy to look for response as well as adverse effects which is rarely described in previous pediatric studies. Following are the limitations. First, neuroimaging and EEG were not repeated in the children after treatment in the followup. It would have revealed response to antihelminthic therapy and also helped in deciding on the duration of antiepileptic. Then again, repeating a CT scan definitely increases radiation exposure to the developing brain of these children. MRI though does not involve radiation; it is associated with need for prolonged sedation or general anesthesia and the associated risk of the same. The cost of the above mentioned procedures is also to be kept in mind in the context of a resource poor setting like India. Second, being a tertiary care hospital based study, the participants may not be representative of the population intended. 5. Conclusion Present study shows that the response to cysticidal therapy is very good in NCC with infrequent adverse effects. Competing Interests The authors declare that they have no competing interests. Figure 1 Flow diagram showing recruitment of cases into the study. Table 1 Baseline characteristics and investigations in children with NCC (N = 120). Characteristics Children with NCC (N = 120) Age distribution 1–4 years 8 (6.7%) 5–10 years 48 (40%) 11–14 years 64 (53.3%) Boys 67 (55.8%) Socioeconomic status (modified Kuppuswamy scale)

Figure 1 Flow diagram showing recruitment of cases into the study. Table 1 Baseline characteristics and investigations in children with NCC (N = 120). Characteristics Children with NCC (N = 120) Age distribution 1–4 years 8 (6.7%) 5–10 years 48 (40%) 11–14 years 64 (53.3%) Boys 67 (55.8%) Socioeconomic status (modified Kuppuswamy scale) Upper 8 (6.7%) Upper-middle 25 (20.8%) Lower-middle 25 (20.8%) Upper-lower 40 (33.3%) Lower 22 (18.3%) Nonvegetarian 77 (64.2%) Symptoms/signs Seizure 104 (86.7%) Raised intracranial pressure 34 (28.3%) Altered sensorium 32 (26.7%) Focal neurological deficit 8 (6.7%) Peripheral blood eosinophilia 13 (10.8%) Stool examination positive for Taenia solium ova 7 (5.8%) CSF findings Pleocytosis 56 (46.7%) Elevated protein 41 (34.2%) Low glucose (<1/2 of concurrently measured blood glucose) 32 (26.7%) Eosinophils in centrifuged sediment 16 (13.3%) Normal CSF 54 (45%) Abnormal EEG 56 (46.7%) Table 2 Neuroimaging findings in children with NCC (N = 120). Neuroimaging findings Number (%) Number of lesions Single 81 (67.5%) 2–5 17 (14.2%) 6–10 14 (11.7%) 11–20 8 (6.7%) >20 0 Location in brain Parenchyma only 102 (85%) Combined (parenchyma and ventricular) 18 (15%) Cistern only 0 Ventricles only 0 Location of the parenchymal lesions Parietal lobe 79% Frontal lobe 15% Temporal lobe 0 Occipital lobe 6% Table 3 Serum T solium antibodies by ELISA in children with neurocysticercosis (N = 43). Results Number Percentage Positive 28 65.1% Negative 15 34.9% Table 4 CSF T. solium antibodies in children with neurocysticercosis (N = 20).

Location of the parenchymal lesions Parietal lobe 79% Frontal lobe 15% Temporal lobe 0 Occipital lobe 6% Table 3 Serum T solium antibodies by ELISA in children with neurocysticercosis (N = 43). Results Number Percentage Positive 28 65.1% Negative 15 34.9% Table 4 CSF T. solium antibodies in children with neurocysticercosis (N = 20). Results Number Percentage Positive 18 90.0% Negative 2 10.0% Table 5 Therapy, treatment outcome, appearance of new symptoms, and side effects of therapy in children with neurocysticercosis. Follow-up duration N Seizure Worsening of other preexisting symptoms New symptoms Side effects of medication 1 week 114 0 2 0 5 2 weeks 109 4 5 0 9 1 month 105 8 4 1 10 3 months 98 6 2 1 NA 6 months 90 5 0 0 NA 1 year 81 5 0 0 NA Table 6 Comparison between patients with and without persistence of seizure at 1 year. Parameters Patients with persistence of seizure at 1 year (n = 5)∗ Patients without persistence of seizure at 1 year (n = 76)∗ P value Mean age of onset (years) 9.6 ± 2.9 10.8 ± 3.2 0.41 Male gender 4 (75%) 42 (55.3%) 0.53 Partial seizure at presentation 3 (60%) 22 (29%) 0.33 Lesions >5 in neuroimaging 2 (40%) 17 (22.4%) 0.72 EEG abnormality at presentation 3 (60%) 40 (52.6%) 0.75 ∗A total of 81 patients could be followed up for 1 year.

1. Introduction Nowadays cardiovascular diseases are among the leading causes of death all around the world [1]. Among the different predisposing causes of these problems hypertension is considered as an important risk factor [2]. About 80 percent of global hypertension cases occur in developing countries. In Iran 50 percent of deaths due to cardiovascular diseases is related to hypertension [3]. Hypertension which usually occurs in adults is determined by different factors such as gender, weight, height, genetic, physiologic, lifestyle, diet, geographical conditions, and ethnicity [4, 5]. Also hypertension occurred in all age groups, but it is obvious that prevalence of hypertension in a population rises along with the increase of the age [6]. Systemic hypertension is an indication of an underlying pathophysiology in childhood and adolescence which if not controlled effectively could be resulted to the increased risk of myocardial and cerebral infarctions in adulthood [7]. Studies show that hypertension in early adulthood age has its root in adolescence [8].

Systemic hypertension is an indication of an underlying pathophysiology in childhood and adolescence which if not controlled effectively could be resulted to the increased risk of myocardial and cerebral infarctions in adulthood [7]. Studies show that hypertension in early adulthood age has its root in adolescence [8]. Hypertension in children is often asymptomatic and does not receive any particular treatment. Hypertension pharmaceutical treatment usually starts in adulthood, whereas high blood pressure (BP) has left its considerable damaging effect on vascular system and other organs of the body from childhood which are often irreversible [9]. Many prospective studies indicated that children with sustainable high BP had higher risk of hypertension in adulthood [10]. Therefore hypertension screening in childhood and adolescence could result in recognition of people with early hypertension and those with higher risk of developing the problem [11]. Forasmuch BP in childhood and adolescence is affected by different factors such as age, gender, height, weight, ethnicity, nutrition, geographic zone, stage of sexual development, and even fatal growth pattern. BP level and its changes vary in different populations. Many studies have reported prevalence of hypertension among children and adolescence from 1 to 16.6 percent [12–14].

ent factors such as age, gender, height, weight, ethnicity, nutrition, geographic zone, stage of sexual development, and even fatal growth pattern. BP level and its changes vary in different populations. Many studies have reported prevalence of hypertension among children and adolescence from 1 to 16.6 percent [12–14]. Studies done on children and adolescents have shown that different factors (such as gender, age, height, and body size) are associated with BP in these groups [15]. It should be considered that as these factors vary in different communities and since that children and adolescents are in a dynamic growth process, there are no specific standard criteria for analysis of BP pattern and also diagnosis and treatment of hypertension in these groups [16]. This study aimed to investigate factor associated with BP and also to determine its standard percentiles in adolescents of Jahrom city in southern Iran. 2. Methods This cross-sectional community-based study was done in Jahrom city of Iran in 2014. Jahrom with 220000 inhabitants is one of the most populated cities of Fars province (the fourth province of the country in terms of area and population) which is located in southern Iran. We used a multistage random cluster sampling method for sampling and accordingly 8 high schools (4 male and 4 female schools) were selected from different areas of the city as clusters. Different educational levels (4 classes) were considered as strata and then we selected samples from these strata randomly. Finally 983 students including 498 male and 454 female students were selected.

accordingly 8 high schools (4 male and 4 female schools) were selected from different areas of the city as clusters. Different educational levels (4 classes) were considered as strata and then we selected samples from these strata randomly. Finally 983 students including 498 male and 454 female students were selected. Necessary data were gathered from the entire participants who include age, gender, height, weight, systolic and diastolic BP, physical activity, and some demographic information. To measure the variables we used 6 public health technicians who were working under the supervision of a general physician. Age of the participants retrieved from their registration documents in the schools. We used a digital stadiometer (Seca 274) for measurement of height and weight of the students. In order to reduce error of measurement, the stadiometer was calibrated with a standard weight every day after using it. Weight of participants measured with an accuracy of 0.1 Kg. In order to increase the accuracy of measurement participants were assessed with minimum possible cloths and also they were asked not to wear shoes and hat.

e error of measurement, the stadiometer was calibrated with a standard weight every day after using it. Weight of participants measured with an accuracy of 0.1 Kg. In order to increase the accuracy of measurement participants were assessed with minimum possible cloths and also they were asked not to wear shoes and hat. We used a digital sphygmomanometer (Omron M6 Comfort) to measure BP of which length of its cuff was at least 2.3 times more than arm circumference of participants. BP of students measured twice with an interval of 5 minutes in a sitting position. The average value of two measurements was considered as BP of participants. In order to increase reliability of the data, all of the BP measurements were done by one of the technicians using the same method. In all cases diaphragm of the stethoscope located on brachial artery, the cuff filled 30 to 40 mmHg more than expected systolic blood pressure (SBP), and then the air released by 3 mmHg in second. The appearance point of the first Korotkoff sounds was considered as SBP. In the cases that Korotkoff sounds were audible to 0 mmHg, the appearance point of forth Korotkoff sounds was considered as diastolic blood pressure (DBP). The sphygmomanometer and stethoscope were checked for likely error every day. Four team groups were trained to collect data by measurement height, weight, BP, and other variables.

at Korotkoff sounds were audible to 0 mmHg, the appearance point of forth Korotkoff sounds was considered as diastolic blood pressure (DBP). The sphygmomanometer and stethoscope were checked for likely error every day. Four team groups were trained to collect data by measurement height, weight, BP, and other variables. In order to increase measurement accuracy some advices were recommended including the following: all participants got rest in a calm and quiet place at least 5 minutes before BP measurement; to reduce the likely stresses all participants were informed about the process; to avoid the effects of dilated urinary bladder all participant were asked to dispose their urine before measurement if needed; the participants were asked not to use caffeine and any adrenergic stimulant such as eye and nose drops and should not be involved in a hard physical activity or severe exercise 1 hour before measurement. All of the participants were in healthy children situation, had not any known disease, and were not under any drug therapy plan. Suspected participants checked by the physician and cases with illnesses and comorbidities (including genetic and congenital disorders, cardiovascular diseases, physical disabilities, Asthma, and chronic kidney disease) were not included in sampling. This study was approved by the ethical committee of Jahrom University of Medical Sciences. All of the cases were informed about the study and participated with the consent.

All of the participants were in healthy children situation, had not any known disease, and were not under any drug therapy plan. Suspected participants checked by the physician and cases with illnesses and comorbidities (including genetic and congenital disorders, cardiovascular diseases, physical disabilities, Asthma, and chronic kidney disease) were not included in sampling. This study was approved by the ethical committee of Jahrom University of Medical Sciences. All of the cases were informed about the study and participated with the consent. 2.1. Statistical Analyses Data were entered into the statistical software for windows (SPSS 16). The descriptive variables such as mean and standard deviations were used. After checking normality assumption by Kolmogorov-Smirnov test, independent t-test was used to compare SBP and DBP with gender. One-way analysis of variance (ANOVA) was performed for finding out significance difference among mean of SBP and DBP with body mass index, father and mother education levels, regular physical exercise, and participants' education levels. P value less than 0.05 was considered statistically significant.

DBP with gender. One-way analysis of variance (ANOVA) was performed for finding out significance difference among mean of SBP and DBP with body mass index, father and mother education levels, regular physical exercise, and participants' education levels. P value less than 0.05 was considered statistically significant. Smoothed gender specific reference plots showing 3th, 10th, 15th, 25th, 50th, 75th, 85th, 90th, and 97th percentiles were derived using LMS method (LMS Chartmaker Pro version 2.4, 2008; by Dr. Huiqi Pan and Dr. Tim Cole) [17]. SBP and DBP were summarized by three smooth curves plotted against age, representing the median (M), coefficient of variation (S), and skewness (L) of the measurement distribution [18]. Models were checked for goodness of fit using the detrended Q-Q plot, Q tests, and worm plots [19]. The LMS method was found to be appropriate to use for this data as the measure of skewness of the data was 1.2 with a standard error of 0.07. 2.2. Results In this study 983 students from 8 high schools of Jahrom city were included. 50.56% (497) of the participants were male with the mean ± SD age of 16.3 ± 1.1 (range: 14.2–20.7) and 49.44% (486) of them were female with the mean ± SD age of 15.8 ± 1.2 (range: 13.8–19.8). Mean ± SD SBP of the students was 110.27 ± 11.44 mmHg (range: 80.60–151.3) and the mean ± SD DBP was 71.76 ± 8.61 mmHg (range: 49.3–105). Also, mean ± SD weight and height were 59.33 ± 13.3 (range 33.7–122.9) and 165.2 ± 7.8 (rang 138–189), respectively.

e female with the mean ± SD age of 15.8 ± 1.2 (range: 13.8–19.8). Mean ± SD SBP of the students was 110.27 ± 11.44 mmHg (range: 80.60–151.3) and the mean ± SD DBP was 71.76 ± 8.61 mmHg (range: 49.3–105). Also, mean ± SD weight and height were 59.33 ± 13.3 (range 33.7–122.9) and 165.2 ± 7.8 (rang 138–189), respectively. Averages of SBP and DBP of the students in terms of gender, body mass index, parents' education level, physical activity, and student's education level are presented in Table 1. After completion of the study the standard percentiles of BP were estimated for students of Jahrom city in both genders. The model was considered a good fit as per the shape of the worm plot; the Q statistic curves for L, M, and  S were within −2 and +2, and the detrended Q-Q plot indicated that the population was approximately normal. Figures 1 and 2 show the standard percentiles of BP in terms of gender and age. 3. Discussion Hypertension in adults could result in various complications imposing financial and nonfinancial burdens to individuals and societies [20]. Hypertension in adolescents is considered as one of the important causes of this problem. Therefor monitoring of BP in childhood and adolescence could be helpful for early detection and prevention of hypertension in adults [21].

omplications imposing financial and nonfinancial burdens to individuals and societies [20]. Hypertension in adolescents is considered as one of the important causes of this problem. Therefor monitoring of BP in childhood and adolescence could be helpful for early detection and prevention of hypertension in adults [21]. According to the results of this study average of SBP among the male students was 5.2 mmHg more than females, whereas the average of DBP among females was 2.5 mmHg more than males. Results of the studies that have evaluated the relationship between gender and BP are very different. Whereas many studies, such as Nielsen and Andersen in Denmark [22], Al-Sendi et al. in Bahrain [14], and Nader et al. [23], in Iran have suggested that average of BP among male adolescents was more than females, some other studies such as Omisore et al. in Nigeria [24] and Falah et al. [13] in Iran have indicated different findings. On the other hand results of some studies such as de Rezende et al. in Brazil [10] and Ataei et al. in Iran [25] did not report a significant difference between average of BP among male and female adolescents. It seems that gender difference in average blood pressure could be result of different life styles, dietary habits, and sociodemographic factors in societies and also different study designs.

] and Ataei et al. in Iran [25] did not report a significant difference between average of BP among male and female adolescents. It seems that gender difference in average blood pressure could be result of different life styles, dietary habits, and sociodemographic factors in societies and also different study designs. Many similar studies, in line with our findings, have confirmed a direct relationship between BMI and BP. Studies of Ribeiro et al. in Portugal [26], He et al. [27] in China, and Soleymani et al. [28] in Iran indicated that average of blood pressure significantly increased along with the body mass index. It could be noted that dietary habits and pattern of physical activity among the obese and overweight people are important factors associated with hypertension.

6], He et al. [27] in China, and Soleymani et al. [28] in Iran indicated that average of blood pressure significantly increased along with the body mass index. It could be noted that dietary habits and pattern of physical activity among the obese and overweight people are important factors associated with hypertension. Findings of this study suggested that there was no significant correlation between regular physical exercise and average of SBP and DBP among adolescents. Similar studies which have been done in this area have found conflicting results. Study of Strazzullo et al. [29] in Australia revealed that there was an indirect relationship between regular physical exercise and hypertension among children and adolescents, while Corrêa Neto et al. [30] did not find a strong correlation between physical exercise and blood pressure among Brazilian children and adolescents. Given that physical activity in children and adolescents is not limited to regular sport activities, definite conclusion about the relationship between BP and regular physical activity is impossible and the correlation should be evaluated through the longitudinal studies.

ssure among Brazilian children and adolescents. Given that physical activity in children and adolescents is not limited to regular sport activities, definite conclusion about the relationship between BP and regular physical activity is impossible and the correlation should be evaluated through the longitudinal studies. This study showed that average of SBP and DBP of students was indirectly associated with the education level of parents, so that average of blood pressure was higher among the students whose parents had lower education levels. It is believed that parental education level is correlated with healthy dietary pattern and life style of students. Consumption of salt, sugar, and cooking oils which contain polyunsaturated fatty acids is higher among the families with lower health knowledge. This kind of nutrition pattern as well as lack of enough attention to regular appropriate physical activity could lead to hypertension in children and adolescents. Also it is assumed that some parents with lower educational level consider their children's overweight and obesity as a sign of health which will result in overnutrition of children. Hypertension among children and adolescents that is always asymptomatic is a leading risk factor of cardiovascular diseases in adults. Therefor detection and control of high blood pressure in this period are considered as an appropriate preventive approach to reducing complications in adults.

This study showed that average of SBP and DBP of students was indirectly associated with the education level of parents, so that average of blood pressure was higher among the students whose parents had lower education levels. It is believed that parental education level is correlated with healthy dietary pattern and life style of students. Consumption of salt, sugar, and cooking oils which contain polyunsaturated fatty acids is higher among the families with lower health knowledge. This kind of nutrition pattern as well as lack of enough attention to regular appropriate physical activity could lead to hypertension in children and adolescents. Also it is assumed that some parents with lower educational level consider their children's overweight and obesity as a sign of health which will result in overnutrition of children. Hypertension among children and adolescents that is always asymptomatic is a leading risk factor of cardiovascular diseases in adults. Therefor detection and control of high blood pressure in this period are considered as an appropriate preventive approach to reducing complications in adults. Findings of this study indicated that different factors are associated with average of blood pressure in students. Body mass index is one of the main factors that is influenced by various determinants such as nutrition pattern, physical activity, and also heredity. Intervention strategies should focus on the most important factors that have been determined through these kinds of studies.

with average of blood pressure in students. Body mass index is one of the main factors that is influenced by various determinants such as nutrition pattern, physical activity, and also heredity. Intervention strategies should focus on the most important factors that have been determined through these kinds of studies. Acknowledgments This article is the result of a thesis project by Dr. Jafari for M.D. degree in Jahrom University of Medical Sciences with the approval code 1394.130 [31]. The project was financially supported by Jahrom University of Medical Sciences. Competing Interests There is no conflict of interests to be declared by the authors regarding the manuscript. Figure 1 The smoothed standard percentiles of systolic blood pressure among 13–21-year-old adolescents of Jahrom city. Figure 2 The smoothed standard percentiles of diastolic blood pressure among 13–21-year-old adolescents of Jahrom city. Table 1 Average of systolic and diastolic blood pressure among Jahrom high school students in terms of demographic and anthropometric factors. Characteristics Blood pressure∗∗∗ P value Gender Male SBP∗ 112.8 ± 11.1 <0.001 Female 107.6 ± 11.1 Total 110.27 ± 11.4 Male DBP∗∗ 70.5 ± 8.4 <0.001 Female 73 ± 8.6 Total 71.76 ± 8.6 Body mass index <18.5 SBP 106.3 ± 10.9 <0.001 18.5–25 110.3 ± 11.1 25–30 114.2 ± 10.4 >30 117.5 ± 13.3 <18.5 DBP 71.2 ± 8.3 <0.001 18.5–25 70.9 ± 8.4 25–30 74 ± 8.4 >30 78 ± 9.3

Characteristics Blood pressure∗∗∗ P value Gender Male SBP∗ 112.8 ± 11.1 <0.001 Female 107.6 ± 11.1 Total 110.27 ± 11.4 Male DBP∗∗ 70.5 ± 8.4 <0.001 Female 73 ± 8.6 Total 71.76 ± 8.6 Body mass index <18.5 SBP 106.3 ± 10.9 <0.001 18.5–25 110.3 ± 11.1 25–30 114.2 ± 10.4 >30 117.5 ± 13.3 <18.5 DBP 71.2 ± 8.3 <0.001 18.5–25 70.9 ± 8.4 25–30 74 ± 8.4 >30 78 ± 9.3 Father's education level Illiterate SBP 114.4 ± 9.4 0.01 Under diploma 110.1 ± 11.6 Diploma 109.2 ± 11 University 111.3 ± 11.7 Illiterate DBP 73 ± 8.1 0.77 Under diploma 71.7 ± 8.7 Diploma 71.7 ± 8.7 University   71.5 ± 8.2 Mother's education level Illiterate SBP 117.4 ± 10.3 0.001 Under diploma 110.3 ± 11.2 Diploma 108.5 ± 10.6 University 111.7 ± 12.7 Illiterate DBP 74.4 ± 7.8 0.05 Under diploma 71.3 ± 8.4 Diploma 71.4 ± 8.6 University 72.7 ± 8.9 Regular physical exercise Without physical activity SBP 109.7 ± 11.7 0.74 With physical activity 110.3 ± 11.4 Without physical activity DBP 72.3 ± 8.4 0.40 With physical activity 71.6 ± 8.6 Participants' education level The first year of high school SBP 109.4 ± 10.7 0.50 The second year of high school 110.1 ± 11.9 The third year of high school 110.8 ± 12.1 Per university 110.7 ± 10.4 The first year of high school DBP 72.4 ± 8.6 0.02 The second year of high school 71.5 ± 8.7 The third year of high school 70.6 ± 8.5 Per university 72.9 ± 8.2 ∗Systolic blood pressure; ∗∗diastolic blood pressure; ∗∗∗data are mean ± SD.

1. Introduction Congenital anomalies are important causes of childhood death, chronic illness, and disability in many countries. Congenital anomalies are also known as birth defects, congenital disorders, or congenital malformations [1]. According to WHO factsheet on 2000–2013 child causes of death, every year, around 276,000 babies die within 4 weeks of birth, worldwide, from congenital anomalies [2]. Congenital anomalies can be defined as structural or functional anomalies (e.g., metabolic disorders) that occur during intrauterine life and can be identified prenatally, at birth, or later in life. Birth defects may be the result of genetic or environmental factors which include errors of morphogenesis, infection, epigenetic modifications on a parental germline, or a chromosomal abnormality. The outcome of the disorder will depend on complex interactions between the prenatal deficit and the postnatal environment [3]. Congenital anomalies can result in long-term disability, which may have significant impacts on individuals, families, health-care systems, and societies. The outcome of children with congenital anomalies in developing countries is worse than in developed countries due to lack of appropriate resources for their management. Congenital anomalies account for 8–15% of perinatal deaths and 13–16% of neonatal deaths in India [4]. As other causes of infant mortality like infections and nutritional deficiencies are being brought under control, congenital malformations are rapidly emerging as one of the major worldwide problems [5, 6]. The prevalence rate of congenital anomalies is increasing due to exposure of teratogens of various kinds [7].

4]. As other causes of infant mortality like infections and nutritional deficiencies are being brought under control, congenital malformations are rapidly emerging as one of the major worldwide problems [5, 6]. The prevalence rate of congenital anomalies is increasing due to exposure of teratogens of various kinds [7]. The present study was carried out with an aim to study the percentage of various congenital anomalies among the patients admitted over a period of four years from 2011 to 2015 in our centre. According to English literature this is the first such study of North India.

4]. As other causes of infant mortality like infections and nutritional deficiencies are being brought under control, congenital malformations are rapidly emerging as one of the major worldwide problems [5, 6]. The prevalence rate of congenital anomalies is increasing due to exposure of teratogens of various kinds [7]. The present study was carried out with an aim to study the percentage of various congenital anomalies among the patients admitted over a period of four years from 2011 to 2015 in our centre. According to English literature this is the first such study of North India. 2. Material and Methods A retrospective analysis was conducted in Department of Pediatric Surgery at Pt. B.D. Sharma, Post Graduate Institute of Medical Sciences, Rohtak, Haryana, from July 2011 to June 2015. The study population comprised 1374 patients admitted with us with congenital anomalies. Relevant information regarding age, sex, birth weight, birth order, and consanguinity was documented. Significant antenatal history like maternal illness, ingestion of drugs, exposure to radiation, mode of delivery, and complications of labor was recorded. Antenatal ultrasonography (USG) findings were noted. Relevant radiological and histohematological tests were carried out. Computed tomography (CT) scan was advised only for certain special cases. The major malformations were divided into central nervous system (CNS), gastrointestinal (GIT), genitourinary (GU), and miscellaneous disorders. Comparison between percentage of affected males and females was made. Many patients with isolated anomalies of lip, palate (like cleft lip/palate), limb deformity (CTEV, syndactyly, polydactyly, etc.), chest deformity (like pectus excavatum), and ear anomalies (microtia, anotia, etc.) were excluded from our study as they did not require immediate surgery and hence were not admitted and managed on OPD basis. Patients who presented in casualty in terminal stage were also not admitted as they could not be operated on and hence were not included in our study. Patients with anomalies of cardiovascular system were referred to pediatric cardiologist and were not admitted with us as there is no pediatric cardiologist at our centre. Hence they were also excluded from this study.

age were also not admitted as they could not be operated on and hence were not included in our study. Patients with anomalies of cardiovascular system were referred to pediatric cardiologist and were not admitted with us as there is no pediatric cardiologist at our centre. Hence they were also excluded from this study. 3. Results During the study period 1374 patients were admitted with us with congenital anomalies. Various congenital anomalies were classified according to the system affected (Table 1). Percentage of various anomalies was calculated and compared as shown in Figure 1. NTD was found to be the most common anomaly in 24.3% of the children admitted. It included meningomyelocele, encephalocele, and hydrocephalus. Lumbosacral meningomyelocele was most common anomaly among them. Next most common anomaly was ARM in 20.74% patients followed by TOF in 20.08% of the patients. Intestinal obstruction was found to be a major anomaly in this study, occurring among 14.84% patients. It included intestinal atresia in most cases. Others were malrotation of gut, duplication cyst, hypertrophic pyloric stenosis, Hirschsprung disease, meconium plug, and, in few cases, intussusception. Abdominal wall defects like gastroschisis, omphalocele, patent vitellointestinal duct, exstrophy bladder, and cloacal exstrophy accounted for 8.15% of the anomalies. Inguinal, lumbar, and umbilical hernia were found in 4.44% of cases. Genitourinary anomalies were found in 3.42%. They comprised posterior urethral valve, pelvic ureteral junction obstruction, and multicystic kidneys.

Abdominal wall defects like gastroschisis, omphalocele, patent vitellointestinal duct, exstrophy bladder, and cloacal exstrophy accounted for 8.15% of the anomalies. Inguinal, lumbar, and umbilical hernia were found in 4.44% of cases. Genitourinary anomalies were found in 3.42%. They comprised posterior urethral valve, pelvic ureteral junction obstruction, and multicystic kidneys. Congenital diaphragmatic hernia (CDH) was seen in 2.04% of cases. Other anomalies seen in few cases were teratoma, hemangioma, cystic hygroma, ranula tongue, diencephaly, ectopia cordis, and parasitic twinning. Out of the total 1374 patients in our study group, 831 (60.5%) were males while only 543 (39.5%) of them were females. Figure 2 shows comparison between number of males and females with congenital anomalies.

Congenital diaphragmatic hernia (CDH) was seen in 2.04% of cases. Other anomalies seen in few cases were teratoma, hemangioma, cystic hygroma, ranula tongue, diencephaly, ectopia cordis, and parasitic twinning. Out of the total 1374 patients in our study group, 831 (60.5%) were males while only 543 (39.5%) of them were females. Figure 2 shows comparison between number of males and females with congenital anomalies. 4. Discussion Congenital malformations are rapidly emerging as one of the major worldwide problems as they can result in long-term disability, which may have significant impacts on individuals, families, health-care systems, and societies. Since the ancient times, congenital anomalies have been topic of frequent discussion and research. The exact causes of malformations remain unknown in a large number of the cases. According to ancient beliefs of negative or supernatural forces, birth defects were a result of divine punishment for wickedness. Over the years, numerous studies have been carried out to determine the prevalence, patterns, possible causations, and other factors of congenital malformations. However despite massive advancements, the magnitude of the problem still to this day causes significant health impacts.

t of divine punishment for wickedness. Over the years, numerous studies have been carried out to determine the prevalence, patterns, possible causations, and other factors of congenital malformations. However despite massive advancements, the magnitude of the problem still to this day causes significant health impacts. In this study we found that the most common anomalies were NTD (24.3%) followed by ARM (20.7%), followed by TOF (20%), and that males were affected more than females. According to the English literature, there is no study of this kind in North India. Hence this is the first one. Some of the previous studies, performed at other regions, have results similar to our present study. In a study in Dhaka, by Fazle Mubarak Bari [8], GIT accounted for majority of the cases 27% followed by nervous system, 15.7%. 52.8% of the patients were males. According to study by Taksande et al. [4], cardiovascular malformations were most common. 62% males and 38% females were affected in this study which is almost the same as in ours. Sarkar et al. [9] found that the predominant system involved was musculoskeletal system (33.2%) followed by gastrointestinal (GI) system (15%) and central nervous system (CNS) (11.2%) congenital anomalies affected significantly higher proportion of male babies than their female counterparts. Chaturvedi and Banerjee [10] studied rural population of Maharashtra and found that the most common system involved was musculoskeletal (23.65%), followed by CNS (16.12%) and GIT (13.97%). Basavanthappa et al. [11] found that musculoskeletal malformations were the commonest malformation and accounted for 27.5% of all the malformations in a hospital of South India. This was followed by cutaneous 19.16%, genitourinary 15.83%, gastrointestinal 12.5%, neurological 10%, and cardiac malformations 5.83%. in a study in West Bengal, by Pal et al. [12], cardiovascular, musculoskeletal, and genitourinary system were found to be most commonly involved. 62% males and 38% females were affected in this study too.

cutaneous 19.16%, genitourinary 15.83%, gastrointestinal 12.5%, neurological 10%, and cardiac malformations 5.83%. in a study in West Bengal, by Pal et al. [12], cardiovascular, musculoskeletal, and genitourinary system were found to be most commonly involved. 62% males and 38% females were affected in this study too. Despite the high risk of recurrence of congenital malformations, there are no well accepted preventive measures in developing countries like India. It indicates that strong preventive measures for congenital anomalies are needed. Increasing awareness about maternal care during pregnancy, educational programs on congenital malformations and the consequences of consanguineous marriages need to be highlighted to decrease the incidence of congenital anomalies and their comorbidities. Nutritional status of women needs to be improved which includes improving their general nutrition; ensuring adequate intake of specific micronutrients including folic acid, iodine, and iron; and removing harmful substances from the diet, especially alcohol, which may damage the developing embryo or fetus. The periconception period (three months before and after conception) can be targeted by folic acid supplementation [13]. Studies suggest that 70% of neural tube defects can be prevented by the intake of daily dose of 400 μg synthetic folic acid for women of childbearing age [14]. Hence, regular antenatal visits and prenatal diagnosis are recommended for prevention, early intervention, and even planned termination, when needed.

tion [13]. Studies suggest that 70% of neural tube defects can be prevented by the intake of daily dose of 400 μg synthetic folic acid for women of childbearing age [14]. Hence, regular antenatal visits and prenatal diagnosis are recommended for prevention, early intervention, and even planned termination, when needed. 5. Conclusion In this study conducted at our centre, we found that NTD and GI anomalies are very common among those admitted for surgery and that males are affected much more than females. Competing Interests The authors declare that they have no competing interests. Figure 1 Percentage of children with various congenital anomalies. Figure 2 Comparison between number of males and females with congenital anomalies. Table 1 Total number of cases in each group of congenital anomalies. S. number Anomaly Total cases Percentage 1 CNS/NTD 334 24.31% 2 Anorectal malformation (ARM) 285 20.74% 3 Tracheoesophageal fistula (TOF) 276 20.08% 4 Neonatal intestinal obstruction 204 14.84% 5 Abdominal wall defects 112 8.15% 6 Hernia 61 4.44% 7 Genitourinary anomalies 47 3.42% 8 Congenital diaphragmatic hernia (CDH) 28 2.04% 9 Miscellaneous 27 1.96%

1. Introduction Trauma is the most common cause of death in children aged one year and older and the main cause of permanent disabilities [1]. The treatment of injuries and life-threatening emergencies in children is a major cognitive challenge and an emotional burden for the treatment teams providing care in the trauma room, even in a Level 1 trauma center for children. Various studies have been able to identify massive deficiencies preclinically as well as in the pediatric surgery trauma room and in pediatric trauma care [2]. The deficits described in emergency care are found not only in the medical-specialty area, but also in the area of the so-called nontechnical skills. These nonmedical errors account for approximately 70 percent of errors according to a report published in 2000 titled “To Err is Human” [3]. Therefore, team-training concepts are increasingly being implemented in many high-risk medical fields as a tool to ensure that interdisciplinary medical care teams are best prepared for emergency situations. These courses allow medical professionals to receive training and some medical-specialty skills and learn team-oriented and behavior-oriented techniques [4]. The European division of the WHO showed that high-quality trauma care can reduce the mortality rate following trauma by up to 30 percent [1]. Numerous studies have been able to demonstrate that time delays in trauma care occur when activities are not carried out in coordinated succession [2, 5, 6].

Therefore, team-training concepts are increasingly being implemented in many high-risk medical fields as a tool to ensure that interdisciplinary medical care teams are best prepared for emergency situations. These courses allow medical professionals to receive training and some medical-specialty skills and learn team-oriented and behavior-oriented techniques [4]. The European division of the WHO showed that high-quality trauma care can reduce the mortality rate following trauma by up to 30 percent [1]. Numerous studies have been able to demonstrate that time delays in trauma care occur when activities are not carried out in coordinated succession [2, 5, 6]. Until now, team trainings have not been available focused on pediatric trauma room care for children with major trauma in German speaking countries. The objective of the pilot project described in this study was, therefore, to establish interdisciplinary simulation-based team training in Germany as a tool to improve the care of trauma patients in the pediatric surgery trauma room.

Until now, team trainings have not been available focused on pediatric trauma room care for children with major trauma in German speaking countries. The objective of the pilot project described in this study was, therefore, to establish interdisciplinary simulation-based team training in Germany as a tool to improve the care of trauma patients in the pediatric surgery trauma room. 2. Methods The first interdisciplinary, simulation-based team training in the pediatric surgery trauma room was held in Tuebingen. The training included 14 medical doctors and 4 nurses from the medical fields of pediatric surgery, pediatric intensive care and emergency medicine, and anesthesia. All of the medical doctors were attending physicians of pediatric surgery, pediatrics, and anesthesia having more than 10 years of practice each. Two weeks before the course started, for theoretical preparation the participants received the guidelines for emergency pediatric care based on the ERC guidelines from the European Resuscitation Council 2010 (ERC). The faculty consisted of eight CRM-trained instructors from the fields of pediatric intensive care, pediatric surgery, anesthesia, and pediatric emergency medicine. Training was held in a mock trauma room in the Tuebingen Patient Safety and Simulation Center (TüPASS) of the University Hospital Tuebingen.

il 2010 (ERC). The faculty consisted of eight CRM-trained instructors from the fields of pediatric intensive care, pediatric surgery, anesthesia, and pediatric emergency medicine. Training was held in a mock trauma room in the Tuebingen Patient Safety and Simulation Center (TüPASS) of the University Hospital Tuebingen. The course lasted 1.5 days. On the first day of the course, the participants received three hours of theoretical introduction to the topics “Trauma Room Management in Pediatric Patients” and “Errors and Patient Safety in Pediatric Emergency Care.” On the second day, participants then had a one-hour introduction to get familiarized with the simulator and the training environment. Also, skill stations focused on airway management and IO access were set up. Next, the training went through six scenarios exclusively from the field of pediatric surgery trauma room care (Table 1). The training goals were adapted to each scenario. Attention was given to ensure that all steps of the diagnostic and therapeutic algorithm of the European Pediatric Life Support (EPLS) [7] or Advanced Trauma Life Support (ATLS) courses [8] had a thematic focus in the scenarios. Here the patient was evaluated using the graduated approach following the ABCD scheme (where A = airway, B = breathing, C = circulation, and D = disability). The scenarios with each of the key medical areas and the CRM learning objectives are illustrated in Table 1. The time schedule of each training sequence was 15 minutes for the scenario with subsequently a 45-minute video-based debriefing for the participants. Therefore a higher weighting was focused on the debriefing allowing sufficient time for complete discussion of the key aspects. It was led by a two-person, interdisciplinary and multiprofessional instructor team. The ratio of medical content to CRM-related aspects was estimated to be approximately 1 : 1. As suggested by other authors, only short video sequences oriented to the learning objective were selected for the debriefing [9]. Each scenario included the active participation of 4–6 doctors and nurses as team members. The participants took on roles that corresponded to their position and their level of clinical training. The course participants who were not actively involved in the scenario observed the scenario from an adjoining room via video transmission. All participants were actively involved in at least two scenarios.

members. The participants took on roles that corresponded to their position and their level of clinical training. The course participants who were not actively involved in the scenario observed the scenario from an adjoining room via video transmission. All participants were actively involved in at least two scenarios. The simulation training was implemented using full-scale patient simulators: a SIMBaby (Laerdal, Stavanger, Norway) as a baby simulator and a Pediatric HAL Five Year (Gaumard, Florida, USA) as a small child simulator. The simulators were controlled from a control room outside of the simulated trauma room. The course evaluation was conducted using anonymized pre- and postsurveys for evaluating the course and for a self-evaluation in regard to medical competency and CRM aspects. Participants were able to select from six response options on a scale between “I fully agree” = “1” and “I do not agree at all” = “6” for each item. The participants were given an anonymized code to allow for comparison between the pre- and postintervention surveys. The participants filled out the surveys and participated in the simulation training voluntarily. Statistical analysis was conducted using Microsoft Excel 2010. Significance was analyzed with Student's t-test; statistical significance was set at an alpha level of p = 0.05. All data were irreversibly made anonymous. The Ethics Committee of the Ludwig-Maximilians-University of Munich granted ethical clearance for this study, as only anonymized data were collected.

ft Excel 2010. Significance was analyzed with Student's t-test; statistical significance was set at an alpha level of p = 0.05. All data were irreversibly made anonymous. The Ethics Committee of the Ludwig-Maximilians-University of Munich granted ethical clearance for this study, as only anonymized data were collected. 3. Results A total of 18 pre- and 17 postsurveys were evaluated. The training included the participation of 14 doctors and 4 nurses from various fields. Of the doctors, 43 percent were residents. The allocation of the occupational groups was 43 percent pediatric surgery, 14 percent pediatrics, and 29 percent anesthesia. All participants treat severely injured children in their daily routine in a pediatric surgical emergency care outpatient center or an interdisciplinary emergency care outpatient center including trauma room. Of the participants, 71 percent reported having more than six years of work experience. 61 percent of the participants completed regular emergency training and 22 percent reported having already participated in simulation-based team training at least once in the past. Only 39 percent of the participants had completed an official pediatric emergency course of the established organizations of the European Resuscitation Council or the American Heart Association (ATLS [8], PALS [10], and EPLS [7]) over the course of the last two years prior to the trauma training.

east once in the past. Only 39 percent of the participants had completed an official pediatric emergency course of the established organizations of the European Resuscitation Council or the American Heart Association (ATLS [8], PALS [10], and EPLS [7]) over the course of the last two years prior to the trauma training. Overall the individual course elements received a very positive evaluation (Table 2). The course was evaluated throughout as very realistic and relevant to the daily routine. Likewise, the detailed debriefings were evaluated as positive in the evaluation. Individual aspects of this trauma training showed that even though this course was short, the individual participants felt there was a benefit for real care of children with critical trauma and found the feedback within the debriefings to be important and applicable to the clinical routine (Table 3). Contrary to the participants' expectations before the course, the video recordings taken during the scenarios for the debriefing were seen as slightly uncomfortable. Likewise, the participants did not feel like they were being put on display in the debriefings (Table 2).

nd applicable to the clinical routine (Table 3). Contrary to the participants' expectations before the course, the video recordings taken during the scenarios for the debriefing were seen as slightly uncomfortable. Likewise, the participants did not feel like they were being put on display in the debriefings (Table 2). The participants reported of a feeling of individual improvement in almost all categories of the medical problems they worked through (Figure 1). Special medical aspects in this regard were pediatric airway management (pretrauma course: median 3, range 1–6; posttrauma course: median 2, range 1–4; not significant (ns)), circulatory problems (pretrauma course: median 3, range 1–6; posttrauma course: median 2, range 1–4; ns), polytrauma management (pretrauma course: median 3, range 1–6; posttrauma course: median 2, range 1–4; ns), management of severe head-brain trauma (pretrauma course: median 3, range 1–6; posttrauma course: median 3, range 2–4; ns), and cardiopulmonary resuscitation (pretrauma course: median 4, range 1–6; posttrauma course: median 2, range 1–4; p < 0.001). An improvement was likewise achieved in nontechnical skills using the example of setting priorities (pre: median 3, range 2–5; post: median 2, range 1–3; p < 0.001) and reliable and effective communication (pre: median 3, range 2–5; post: median 2, range 2-3; ns) (Figure 1).

uma course: median 2, range 1–4; p < 0.001). An improvement was likewise achieved in nontechnical skills using the example of setting priorities (pre: median 3, range 2–5; post: median 2, range 1–3; p < 0.001) and reliable and effective communication (pre: median 3, range 2–5; post: median 2, range 2-3; ns) (Figure 1). 4. Discussion The pilot project described here was the first to introduce a simulator-based course concept focused on pediatric surgical trauma care in the German-speaking countries. For this project, the course concept of the PAEDSIM Working Group that was already tested in many trainings in pediatric emergencies involving over 1.000 participants was adapted to the special needs of a pediatric trauma emergency room [11]. In regard to the interdisciplinary character of trauma care, it seems to be necessary to integrate nonmedical aspects also, such as teamwork and communication, into established training concepts [12].

ergencies involving over 1.000 participants was adapted to the special needs of a pediatric trauma emergency room [11]. In regard to the interdisciplinary character of trauma care, it seems to be necessary to integrate nonmedical aspects also, such as teamwork and communication, into established training concepts [12]. This becomes even more relevant in the time-critical emergency care of a pediatric trauma. In addition to a clear organizational structure and assignment of tasks, closed loop communication and clear team leadership are required in other course concepts [13–15]. Such was also demonstrated impressively in the scenarios practiced in this pilot project. The participants recognized for themselves the need for a clear team structure, especially in complex situations, for example, the maintenance of two patients after MVA in the trauma room (double scenario). This self-recognition is the basis for deep, experienced-based learning (deliberate practice) [16, 17]. In this respect, we believe the opportunity of having a video debriefing is an essential basis. This view coincides with the experience of other authors [18]. Participants reported that they did not feel uncomfortable with this video recording and did not feel “paraded” during scenarios. This evaluation is important to maximize the impact of training sessions and avoiding retention, in particular for simulation-based training in Germany.

the experience of other authors [18]. Participants reported that they did not feel uncomfortable with this video recording and did not feel “paraded” during scenarios. This evaluation is important to maximize the impact of training sessions and avoiding retention, in particular for simulation-based training in Germany. In our opinion, a simulator-based course concept with a focus on CRM cannot replace the known guideline courses of medical societies (e.g., ATLS, EPLS). It is, however, a very helpful additional course focused on the important nontechnical skills during time-critical care in a large team. The “common language” of the current algorithm-oriented course formats must continue to be a basis of interdisciplinary trauma care in pediatric patients. Only those who know what is meant by the ABCDE care algorithm can function as good team members and think with foresight. To act in this way, all trainees as stakeholders in their field work as multiplicators and are responsible for the training's acceptance within their teams.

linary trauma care in pediatric patients. Only those who know what is meant by the ABCDE care algorithm can function as good team members and think with foresight. To act in this way, all trainees as stakeholders in their field work as multiplicators and are responsible for the training's acceptance within their teams. The participants' positive evaluation of the course format in regard to the relevance to daily practice and the reality of the practiced scenarios show that the instructors have selected the right scenarios. The relevance to the daily work of each participant is an important criterion of a simulation scenario developed by the PAEDSIM Working Group. According to the opinion of the authors, standard treatment of a life-threatening injury in a child is already so demanding for a multiheaded team that a conscious decision was made to exclude other devised snares, such as a power outage and other technical problems, or the presentation of rare diagnoses [12, 19].

oup. According to the opinion of the authors, standard treatment of a life-threatening injury in a child is already so demanding for a multiheaded team that a conscious decision was made to exclude other devised snares, such as a power outage and other technical problems, or the presentation of rare diagnoses [12, 19]. The positive evaluation of the debriefings supports the course format with a temporary focus on the debriefing. An appropriate amount of time is needed to work through a complex incident of this type, involving the provision of care for severely injured children. Therefore, the course planners calculated 45 minutes for each debriefing. The importance of structured debriefing for the aforementioned “deep learning” of the participants is also emphasized by other authors [20, 21]. The debriefing structure proposed by Cheng et al. was effective in the course presented here as well [22]. This debriefing method, which avoids any assignment of blame and premature judgment, is a focus of the instructor training of the PAEDSIM Working Group and contributes to the good evaluation of the debriefings by the participants (Table 3).

oposed by Cheng et al. was effective in the course presented here as well [22]. This debriefing method, which avoids any assignment of blame and premature judgment, is a focus of the instructor training of the PAEDSIM Working Group and contributes to the good evaluation of the debriefings by the participants (Table 3). The participants' rather average evaluation of the theoretical part calls for a revision of this section of the course. We speculate that the presentations on trauma care in pediatric patients were not adequately adjusted to the level of the participants. Due to the pilot character of the course, the majority of participants already had many years of experience in pediatric trauma care. Here a more accurate evaluation of the participants' knowledge would have been necessary at the time of course planning. Such could eventually be evaluated using an online survey distributed prior to the course. An alternative would be to reduce the amount of time for this part of the course as a way to offer more skill stations for smaller groups. Leaning on the model of the ATLS [8] courses, thematic preparation for the course would be done in private study with the help of a special manual. If participants were less experienced, an alternative would be to extend the duration of the course to two days, allowing sufficient time for interactive theoretical processing of the course content. In addition to resuscitation, pediatric trauma has been identified as an uncommon event that requires practice in managing. Furthermore to the ATLS trauma courses only a small component is devoted to pediatrics and in the EPLS courses specific performance about pediatric trauma has a low representation [23].

e content. In addition to resuscitation, pediatric trauma has been identified as an uncommon event that requires practice in managing. Furthermore to the ATLS trauma courses only a small component is devoted to pediatrics and in the EPLS courses specific performance about pediatric trauma has a low representation [23]. 5. Summary The objective of the present pilot project was to apply the concept of the PAEDSIM Working Group to the interdisciplinary management of pediatric trauma. We were able to show that a high-quality, simulation-based course concept can be implemented even within a narrow time frame of 1.5 days. The number of course participants is not sufficient, however, to demonstrate a subjective improvement in medical techniques and nontechnical skills based on the participants' self-evaluation. The selection of these subjective parameters as a measure of the effect of training is being viewed with increasing criticism. To evaluate the effect of simulation-based training, the personnel and organizational structure would have to remain as constant as possible and clinical quality parameters, such as the change in inner clinical care time, safety in diagnostic activities, and the quality of pediatric surgical therapy of a severely injured child, would have to be analyzed in the trauma room.

e personnel and organizational structure would have to remain as constant as possible and clinical quality parameters, such as the change in inner clinical care time, safety in diagnostic activities, and the quality of pediatric surgical therapy of a severely injured child, would have to be analyzed in the trauma room. The present project is intended to serve as an impetus for further expansion of the modern and innovative educational concept of simulation-based training in pediatric surgery. This course concept is scheduled to be continued as in situ training within hospitals in the real clinical setting of interdisciplinary pediatric surgery trauma room care. This study demonstrates that simulation-based training even in pediatric trauma scenarios is feasible in an interdisciplinary setting in Germany. Disclosure An earlier version of this work was presented as an abstract at the 4th International Pediatric Simulation Symposia and Workshop (IPSSW 2011). Competing Interests The authors declare that there is no conflict of interests regarding the publication of this paper. Figure 1 Assessment of various elements of the course pre- and posttrauma training (1 = very good, 6 = unsatisfactory, median, 25th and 75th percentiles, and span). ∗p < 0.001. Table 1 The different scenarios with their respective medical and CRM priorities. Trauma scenario Category Training goal CRM goal Hypovolemic shock in a child with blunt abdominal trauma C Recognition and treatment of hypovolemic shock Reevaluation effective communication

Figure 1 Assessment of various elements of the course pre- and posttrauma training (1 = very good, 6 = unsatisfactory, median, 25th and 75th percentiles, and span). ∗p < 0.001. Table 1 The different scenarios with their respective medical and CRM priorities. Trauma scenario Category Training goal CRM goal Hypovolemic shock in a child with blunt abdominal trauma C Recognition and treatment of hypovolemic shock Reevaluation effective communication Maintenance two patients after MVA in the trauma room (double scenario) A B C D Recognition and treatment of respiratory failure, rapid sequence intubation, CPR Detecting and treating a hematothorax Team and time management prioritization Mobilization of all available resources Get help early Tracheal tube dislocation after repositioning the patient A B DOPES Avoidance of fixing errors of the tracheal tube Mobilization of all available resources Battered child B D Differential diagnosis of unconsciousness CPR algorithm Dealing with parents Double check Use of any information Tension pneumothorax in a major injured child with thoracic contusion A B C Differential diagnosis of acute circulatory insufficiency Treatment of a tension pneumothorax CPR Prioritization team leadership anticipation

Battered child B D Differential diagnosis of unconsciousness CPR algorithm Dealing with parents Double check Use of any information Tension pneumothorax in a major injured child with thoracic contusion A B C Differential diagnosis of acute circulatory insufficiency Treatment of a tension pneumothorax CPR Prioritization team leadership anticipation Traumatic brain injury (TBI) with secondary deterioration and seizure following sledge accident D Treatment of TBI and seizure neuroprotection Prioritization (diagnostic procedures versus surgical care) A = airway, B = breathing, C = circulation, D = disability; CRM = crisis resource management; CPR: cardiopulmonary resuscitation. DOPES: D = displacement (tube), o = Obstruction (tube), P = pneumothorax, E = equipment failure, S = stomach pressure; MVA = motor vehicle accident. Table 2 Evaluation of the individual course elements (1 = very good, 6 = unsatisfactory, and n = number of participants). Parameter 1 2 3 4 5 6 Overall impression 14 3 — — — — Lessons (CRM + acute trauma care, emergencies) 2 2 9 4 — — Realism of scenarios 9 6 1 1 — — Relevance of the scenarios for the practice 12 3 2 — — — Debriefings 11 4 — 1 — Table 3 Individual marks of the course elements (n = number of participants).

Table 2 Evaluation of the individual course elements (1 = very good, 6 = unsatisfactory, and n = number of participants). Parameter 1 2 3 4 5 6 Overall impression 14 3 — — — — Lessons (CRM + acute trauma care, emergencies) 2 2 9 4 — — Realism of scenarios 9 6 1 1 — — Relevance of the scenarios for the practice 12 3 2 — — — Debriefings 11 4 — 1 — Table 3 Individual marks of the course elements (n = number of participants). Parameter I totally agree I agree I tend to agree I tend to disagree I do not agree I do not agree at all In this course I got benefit for my clinical practice? 13 4 — — — — The feedback from the instructors is useful for my clinical practice? 10 7 — — — — I felt uncomfortable with video recordings during the scenarios. — 1 — — 7 9 I feel “paraded” during scenarios. — — — 2 2 14

1. Introduction Kawasaki disease (KD) is an acute febrile illness that predominantly affects infants and children [1]. This disease is characterized by a systemic vasculitis that may have coronary artery lesions (CALs) [1, 2]. Although intravenous immunoglobulin (IVIG) is the established treatment for acute KD [2, 3], more than 10% of such patients are resistant to this therapy. KD has a diverse array of clinical manifestations and laboratory findings, and the histopathological examinations demonstrate arteritis and inflammation in multiple organs including the kidney [4, 5]. Abnormal urinary findings, such as sterile pyuria, proteinuria, and microscopic hematuria, are often seen in patients with KD during the acute phase [5, 6]. Although sterile pyuria in KD had been thought to be caused by urethritis [5], Watanabe et al. have reported that these urinary cells originated from both the urethra and the kidney as a result of mild and subclinical renal injury [7, 8].

opic hematuria, are often seen in patients with KD during the acute phase [5, 6]. Although sterile pyuria in KD had been thought to be caused by urethritis [5], Watanabe et al. have reported that these urinary cells originated from both the urethra and the kidney as a result of mild and subclinical renal injury [7, 8]. Lactate dehydrogenase (LDH) is an enzyme found in nearly all living cells and consists of at least five isozymes (LDH1–LDH5) [9]. Urinary LDH (U-LDH) activity, especially U-LDH4 and U-LDH5, was reported to be elevated in child patients with upper urinary tract infection (UTI) [10–12]. However, there has been no report of U-LDH activity or its isozyme patterns in KD. The aim of the present study was to investigate whether or not U-LDH activity is increased in the acute phase of KD and, if so, which of the isozymes are increased. Furthermore, to investigate whether or not U-LDH levels reflect the severity of KD vasculitis, the U-LDH activity and its isozyme patterns were compared between IVIG responders and IVIG nonresponders. 2. Materials and Methods 2.1. Ethical Statement The present study was approved by the institutional review board at the National Defense Medical College.

Lactate dehydrogenase (LDH) is an enzyme found in nearly all living cells and consists of at least five isozymes (LDH1–LDH5) [9]. Urinary LDH (U-LDH) activity, especially U-LDH4 and U-LDH5, was reported to be elevated in child patients with upper urinary tract infection (UTI) [10–12]. However, there has been no report of U-LDH activity or its isozyme patterns in KD. The aim of the present study was to investigate whether or not U-LDH activity is increased in the acute phase of KD and, if so, which of the isozymes are increased. Furthermore, to investigate whether or not U-LDH levels reflect the severity of KD vasculitis, the U-LDH activity and its isozyme patterns were compared between IVIG responders and IVIG nonresponders. 2. Materials and Methods 2.1. Ethical Statement The present study was approved by the institutional review board at the National Defense Medical College. 2.2. Study Subjects We retrospectively reviewed the clinical records of 120 patients with KD, 18 patients with viral infection (VI), and 43 patients with UTI. All patients were hospitalized at the National Defense Medical College hospital between January 2002 and December 2015. All KD patients were enrolled within 6 days of the onset of illness, with day 1 defined as the first day of the fever, and all patients met the diagnostic criteria for KD established by the Diagnostic Guidelines for Kawasaki Disease (5th revision) [13]. No bacterial species were identified in the blood or urine cultures from KD patients. All KD patients were treated with oral aspirin (30 mg/kg/day), IVIG (2 g/kg/day), and intravenous ulinastatin (15000 U/kg in 3 divided doses) [14]. “IVIG nonresponder” was defined as a persistent fever lasting >24 h after the completion of IVIG or a recrudescent fever associated with KD symptoms after an afebrile period. The Kobayashi score [15], which is a common predictor for IVIG resistance in Japan, was calculated in all KD patients. Although six patients had transient dilatation of coronary arteries from the acute through the subacute phases, they had no segmental aneurysm at the convalescent phase. Urine samples were obtained from KD patients in the acute febrile phase (pre-IVIG), subacute phase (post-IVIG), and convalescent (afebrile) phase. U-LDH activity was analyzed using the Japanese Society of Clinical Chemistry (JSCC) transferable method [16]. U-LDH isozyme was analyzed using a cellulose acetate membrane electrophoresis.

ere obtained from KD patients in the acute febrile phase (pre-IVIG), subacute phase (post-IVIG), and convalescent (afebrile) phase. U-LDH activity was analyzed using the Japanese Society of Clinical Chemistry (JSCC) transferable method [16]. U-LDH isozyme was analyzed using a cellulose acetate membrane electrophoresis. The VI group included 18 children with fever (≥38°C). The type of virus in three patients was definitively diagnosed (two with adenovirus infection by a rapid diagnostic test using throat swab and one with Epstein-Barr virus infection by a serum antibody test). As 15 patients (9 with acute upper respiratory infection and 6 with bronchopneumonia) had low CRP levels (<2.0 mg/dl) and their fever went down without the administration of antibiotics, all of them were clinically diagnosed with some type of viral infection. None of the VI patients met the diagnostic criteria for KD, and no bacterial species were detected in the urine culture from them. The UTI group included 43 children with a fever (≥38°C) before antibiotic therapy. Urinary specimens for culture were collected by catheter. The diagnosis of UTI was based on a quantitative urine culture yielding greater than 105 colony-forming units (CFU) of bacteria per milliliter of urine. The UTI group included 37 children with Escherichia coli, 4 with Klebsiella oxytoca, and 2 with Enterococcus faecalis, detected in urine culture. Thus, all UTI patients were diagnosed as upper UTI.

on a quantitative urine culture yielding greater than 105 colony-forming units (CFU) of bacteria per milliliter of urine. The UTI group included 37 children with Escherichia coli, 4 with Klebsiella oxytoca, and 2 with Enterococcus faecalis, detected in urine culture. Thus, all UTI patients were diagnosed as upper UTI. 2.3. Statistical Analysis All of the data are presented as the mean ± standard error (SE). The statistical analyses were performed using JMP11-0 statistical software (SAS Institute, Inc., Cary, NC). Pearson's chi-square test was used for the comparison of categorical variables. We compared data among the three groups using the analysis of variance (ANOVA) tests for continuous variables and Tukey's honestly significant difference (HSD) test. We compared variables between two groups using Student's t-test. A P value < 0.05 was considered to be statistically significant. 3. Results The clinical and laboratory data were compared among the KD, VI, and UTI groups (Table 1). The KD group tended to have significantly higher levels of neutrophil proportion (P < 0.001) and significantly lower levels of sodium (P < 0.01) than the VI or UTI groups. The KD group also tended to be older, have significantly higher levels of alanine aminotransferase (ALT), and have significantly lower levels of albumin (Alb) and white blood cells (WBCs) than the UTI group (P < 0.01).

proportion (P < 0.001) and significantly lower levels of sodium (P < 0.01) than the VI or UTI groups. The KD group also tended to be older, have significantly higher levels of alanine aminotransferase (ALT), and have significantly lower levels of albumin (Alb) and white blood cells (WBCs) than the UTI group (P < 0.01). Total U-LDH activity and its isozyme patterns were compared among the KD, VI, and UTI groups (Table 2). The UTI group had significantly higher total LDH activity (66.0 ± 8.0 IU/L, P < 0.01) than the KD (35.4 ± 4.8 IU/L) and VI groups (17.0 ± 6.2 IU/L), and the KD group had significantly higher total LDH activity (P < 0.05) than the VI group. On comparison of the U-LDH isozyme patterns among the three groups, the KD group had significantly higher levels of U-LDH1 (12.5 ± 0.9 IU/L) than the VI (5.5 ± 2.4 IU/L, P < 0.01) and UTI (7.4 ± 1.5 IU/L, P < 0.05) groups and significantly higher levels of U-LDH2 (8.1 ± 0.6 IU/L, P < 0.01) than the VI group (3.9 ± 1.5 IU/L). In contrast, the UTI groups had significantly higher levels of U-LDH3 (9.5 ± 1.2 IU/L), U-LDH4 (14.4 ± 2.0 IU/L), and U-LDH5 (27.4 ± 3.9 IU/L) than the KD (4.2 ± 0.7, 4.0 ± 1.2, and 6.6 ± 2.3 IU/L) and VI groups (2.0 ± 2.0, 1.6 ± 3.3, and 3.9 ± 6.4 IU/L) (P < 0.01). There was no correlation between the serum LDH levels and total U-LDH activity in the KD, VI, and UTI groups (data not shown).

5 ± 1.2 IU/L), U-LDH4 (14.4 ± 2.0 IU/L), and U-LDH5 (27.4 ± 3.9 IU/L) than the KD (4.2 ± 0.7, 4.0 ± 1.2, and 6.6 ± 2.3 IU/L) and VI groups (2.0 ± 2.0, 1.6 ± 3.3, and 3.9 ± 6.4 IU/L) (P < 0.01). There was no correlation between the serum LDH levels and total U-LDH activity in the KD, VI, and UTI groups (data not shown). The total U-LDH activity was compared among the pre-IVIG, post-IVIG, and convalescent phases of KD. The mean levels of U-LDH were significantly higher in the pre-IVIG phase (35.4 ± 2.1 IU/L) than in the post-IVIG (19.2 ± 3.3 IU/L) and convalescent (13.0 ± 4.4 IU/L) phases (P < 0.001) (Figure 1(a)). The time course changes in the total U-LDH activity in the nine KD patients whose data could be consecutively acquired in these three phases are shown in Figure 1(b). The increased levels of total U-LDH activity in the pre-IVIG phase tended to decrease from the subacute phase throughout the convalescent phase. In the KD group (n = 120), 89 patients were responders to IVIG therapy, and 31 patients were nonresponders. Total U-LDH activity and its isozyme patterns during pre-IVIG phase were compared between the IVIG responders and nonresponders in the KD group (Table 3). The mean Kobayashi score in the IVIG nonresponder group (5.6 ± 0.4) was significantly higher than that in the IVIG responder group (3.5 ± 0.2) (P < 0.05). The IVIG nonresponder group had significantly higher total U-LDH activity (45.1 ± 4.7 IU/L, P < 0.05) than the IVIG responder group (32.0 ± 2.8 IU/L). Furthermore, the IVIG nonresponder group had significantly higher levels of U-LDH1 (17.4 ± 1.9 IU/L) and U-LDH2 (10.8 ± 1.1 IU/L) isozymes than the IVIG responder group (10.9 ± 1.1 IU/L, 7.1 ± 0.6 IU/L) (P < 0.05). Furthermore, urinary levels of N-acetyl-β-D-glucosaminidase (NAG, n = 83) and β2-microglobulin (n = 87) were measured in KD patients. There was a significant positive correlation between the total U-LDH activity and urinary NAG levels (R = 0.52; P < 0.001) but not between the total U-LDH activity and urinary β2-microglobulin levels (R = 0.22; P = 0.038).

tyl-β-D-glucosaminidase (NAG, n = 83) and β2-microglobulin (n = 87) were measured in KD patients. There was a significant positive correlation between the total U-LDH activity and urinary NAG levels (R = 0.52; P < 0.001) but not between the total U-LDH activity and urinary β2-microglobulin levels (R = 0.22; P = 0.038). 4. Discussion The present study revealed that total U-LDH activity increased in the acute phase of KD and decreased from the subacute phase throughout the convalescent phase. In the isozyme pattern analysis, KD patients had high levels of U-LDH1 and U-LDH2, while UTI patients had high levels of U-LDH3, U-LDH4, and U-LDH5. Furthermore, the IVIG nonresponder group of KD patients had significantly higher total U-LDH activity, especially U-LDH1 and U-LDH2 levels, than the IVIG responder group.

In the isozyme pattern analysis, KD patients had high levels of U-LDH1 and U-LDH2, while UTI patients had high levels of U-LDH3, U-LDH4, and U-LDH5. Furthermore, the IVIG nonresponder group of KD patients had significantly higher total U-LDH activity, especially U-LDH1 and U-LDH2 levels, than the IVIG responder group. Total U-LDH activity has been reported to be increased in a wide spectrum of renal diseases, including an upper UTI and malignant diseases [17, 18]. The U-LDH level is reported to have no correlation with the serum LDH level [19], and the same finding was also seen in the present study. These findings suggest that U-LDH does not translocate from the bloodstream but originates from the urinary tract, including the kidney, ureter, bladder, and urethra. U-LDH is composed of five isozymes (U-LDH1–5) [9], and U-LDH1 and U-LDH2 predominate in normal, healthy children [10]. In patients with a lower UTI, such as cystitis, the predominance of U-LDH1 and U-LDH2 is maintained as it is, without significant increase in total U-LDH activity [10]. Under conditions of an upper UTI, such as pyelonephritis, the isozyme composition changes from predominance of U-LDH1 and U-LDH2 (fast-zone pattern) to that of U-LDH4 and U-LDH5 (slow-zone pattern) [10, 11]. Since leukocytes contain abundant LDH4 and LDH5, the presence of elevated numbers of leukocytes in the urine may result in a shift to the predominance of U-LDH4 and U-LDH5 [9]. Furthermore, it has been reported that infection, ischemia, or necrosis can alter the isozyme composition of renal tissue [11].

tern) [10, 11]. Since leukocytes contain abundant LDH4 and LDH5, the presence of elevated numbers of leukocytes in the urine may result in a shift to the predominance of U-LDH4 and U-LDH5 [9]. Furthermore, it has been reported that infection, ischemia, or necrosis can alter the isozyme composition of renal tissue [11]. The present study is the first to find that KD patients had high levels of U-LDH activity. The predominance of U-LDH1 and U-LDH2 isozymes in the acute phase of KD is thought to be derived from the kidney and urinary tract. Amano et al. reported that urinary abnormalities such as proteinuria and pyuria in KD may be caused by inflammation in the urinary system including focal interstitial nephritis, cystitis, and prostatitis [20]. Ohta et al. demonstrated that the urinary levels of urinary NAG and β2-microglobulin were elevated during the acute phase of KD, suggesting the presence of an inflammatory process within the renal parenchyma, such as interstitial nephritis [21]. The present study revealed that the U-LDH activity was positively correlated with the urinary levels of NAG. Renal echographic evaluation demonstrated enlarged kidneys with increased corticomedullary differentiation [22], and follow-up technetium-99m dimercaptosuccinic acid scintigraphy single-photon-emission computed tomography revealed renal scarring in 46% patients of KD [23]. Wu et al. reported that immune-mediated vasculitis may be responsible for renal involvement in KD [24]. Therefore, the elevation of U-LDH1 and U-LDH2 isozymes may be mainly caused by renal injury due to KD vasculitis. However, because not all of KD patients have urinary abnormalities, such as proteinuria, pyuria, or increased levels of NAG, β2-microglobulin, and U-LDH, these renal findings do not directly lead to the diagnosis of KD.

e elevation of U-LDH1 and U-LDH2 isozymes may be mainly caused by renal injury due to KD vasculitis. However, because not all of KD patients have urinary abnormalities, such as proteinuria, pyuria, or increased levels of NAG, β2-microglobulin, and U-LDH, these renal findings do not directly lead to the diagnosis of KD. Our results also revealed that the IVIG nonresponder group had significantly higher U-LDH activity, especially U-LDH1 and U-LDH2 isozymes, than the IVIG responder group. IVIG nonresponders are suggested to be at higher risk of developing CALs than IVIG responders [2, 3]. Fatal KD cases are reported to have findings of renal vasculitis, hypercellularity in glomeruli, and interstitial cell infiltration [5]. Recently, the new term “Kawasaki disease shock syndrome” has been proposed as a severe form of KD, and these patients are reported to be resistant to IVIG therapy and associated with greater risk of CALs [25]. Gatterre et al. reported that 10 out of 11 patients with Kawasaki disease shock syndrome developed acute kidney injury [26]. Thus, it cannot be denied that these high levels of U-LDH1 and U-LDH2 may reflect the degree of severity of this disease.

resistant to IVIG therapy and associated with greater risk of CALs [25]. Gatterre et al. reported that 10 out of 11 patients with Kawasaki disease shock syndrome developed acute kidney injury [26]. Thus, it cannot be denied that these high levels of U-LDH1 and U-LDH2 may reflect the degree of severity of this disease. The patterns of U-LDH isozymes differed between the KD and UTI groups (Table 2). Namely, U-LDH1 and U-LDH2 dominated in KD, while U-LDH3, U-LDH4, and U-LDH5 dominated in UTI. Benseler et al. reported that 42 (33%) out of 129 patients with KD had concurrent infections at the time of diagnosis, and 4 patients had UTIs due to E. coli and Klebsiella [27]. Wu et al. revealed that 8 (10.7%) out of 75 patients with KD had bacterial pyuria in urine culture, indicating that pyuria is not always sterile in KD patients [28]. There is a case report of KD misdiagnosed as acute pyelonephritis, because fever and pyuria preceded the appearance of classical mucocutaneous signs [29]. Therefore, the measurement of U-LDH isozymes may aid in the differential diagnosis of KD and UTI.

, indicating that pyuria is not always sterile in KD patients [28]. There is a case report of KD misdiagnosed as acute pyelonephritis, because fever and pyuria preceded the appearance of classical mucocutaneous signs [29]. Therefore, the measurement of U-LDH isozymes may aid in the differential diagnosis of KD and UTI. Several limitations associated with the present study must be mentioned. First, we investigated a small number of patients in only one institution. The VI group is thought to include a wide variety of viral infections, and the number of patients was small. To determine the sensitivity and specificity of U-LDH for distinguishing KD from UTI and other infectious diseases, further studies will be needed in the future. Second, since the present study was a retrospective observation without randomization, it may have had some bias. It is possible that the patients in the present study were not representative of patients with KD in general. A prospectively designed study should be performed to minimize the survey bias. Third, since none of KD patients had CALs as sequelae in the present study, we could not investigate the association between U-LDH level and CALs. Further studies will be needed to determine whether or not U-LDH level is a predictor for CALs. Despite these limitations, however, the present results offer new information that KD patients have increased total U-LDH activity, especially U-LDH1 and U-LDH2 isozymes, unlike UTI.

association between U-LDH level and CALs. Further studies will be needed to determine whether or not U-LDH level is a predictor for CALs. Despite these limitations, however, the present results offer new information that KD patients have increased total U-LDH activity, especially U-LDH1 and U-LDH2 isozymes, unlike UTI. In conclusion, the total U-LDH activity increased in the acute phase of KD and decreased from the subacute throughout the convalescent phases. While UTI patients had high levels of U-LDH3, U-LDH4, and U-LDH5, KD patients had high levels of U-LDH1 and U-LDH2. IVIG nonresponders with KD had significantly higher total U-LDH activity, especially U-LDH1 and U-LDH2, than IVIG responders. These results indicate that KD patients have a unique pattern of U-LDH isozymes that differ from those in UTI patients. Acknowledgments This study was supported by grants from the Japan Kawasaki Disease Research Center and the Kawasaki Disease Research from Japan Blood Products Organization. Disclosure The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the paper. Competing Interests The authors declare that they have no competing interests. Figure 1 Comparison of total U-LDH activity among pre-IVIG, post-IVIG, and convalescent phases of KD (a) and the time course changes in the total U-LDH activity (b). Table 1 A comparison of the clinical data among the KD, VI, and UTI groups. KD (n = 120) VI (n = 18) UTI (n = 43) P

Competing Interests The authors declare that they have no competing interests. Figure 1 Comparison of total U-LDH activity among pre-IVIG, post-IVIG, and convalescent phases of KD (a) and the time course changes in the total U-LDH activity (b). Table 1 A comparison of the clinical data among the KD, VI, and UTI groups. KD (n = 120) VI (n = 18) UTI (n = 43) P Male, n (%) 70 (58.3) 10 (55.6) 28 (65.1) 0.689# Age, month 33.4 ± 2.78† 41.1 ± 7.22 11.9 ± 4.63 <0.001 T-Bil, mg/dl 1.24 ± 0.12 0.76 ± 0.47 0.65 ± 0.21 0.060 AST, IU/L 150.1 ± 23.6 45.4 ± 97.1 42.3 ± 43.8 0.089 ALT, IU/L 138.0 ± 16.4† 28.0 ± 67.2 28.2 ± 30.3 0.006 LDH, IU/L 349.8 ± 13.6 305.9 ± 54.6 283.0 ± 24.8 0.074 TP, g/dl 6.50 ± 0.07 6.46 ± 0.20 6.49 ± 0.10 0.980 Alb, g/dl 3.77 ± 0.05† 4.03 ± 0.17 4.16 ± 0.08 <0.001 Sodium, mEq/L 134.5 ± 0.25∗∗ 138.3 ± 2.05 137.2 ± 0.47 <0.001 CRP, mg/dl 9.05 ± 0.47 2.36 ± 1.65 6.51 ± 0.78 <0.001 WBC, ×103/μL 15.0 ± 0.55† 13.2 ± 2.18 19.6 ± 1.01 <0.001 Neutrophil, % 71.9 ± 1.40∗ 42.8 ± 5.74 59.5 ± 2.57 <0.001 Hb, g/dl 11.4 ± 0.12 11.8 ± 0.42 11.4 ± 0.18 0.680 Hct, % 34.2 ± 0.27 34.5 ± 1.15 33.2 ± 0.51 0.194 Plt, ×104/μL 38.3 ± 2.59 43.3 ± 11.2 44.5 ± 4.80 0.521 Alb, albumin; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CRP, C-reactive protein; Hb, hemoglobin; Hct, hematocrit; LDH, lactate dehydrogenase; Plt, platelet; T-Bil, total bilirubin; TP, total protein; WBC, white blood cell. The data are presented as the mean ± SE for continuous variables and as the number of patients (%) for categorical variables. The P values were obtained using an ANOVA or #Pearson's chi-squared test.

Male, n (%) 70 (58.3) 10 (55.6) 28 (65.1) 0.689# Age, month 33.4 ± 2.78† 41.1 ± 7.22 11.9 ± 4.63 <0.001 T-Bil, mg/dl 1.24 ± 0.12 0.76 ± 0.47 0.65 ± 0.21 0.060 AST, IU/L 150.1 ± 23.6 45.4 ± 97.1 42.3 ± 43.8 0.089 ALT, IU/L 138.0 ± 16.4† 28.0 ± 67.2 28.2 ± 30.3 0.006 LDH, IU/L 349.8 ± 13.6 305.9 ± 54.6 283.0 ± 24.8 0.074 TP, g/dl 6.50 ± 0.07 6.46 ± 0.20 6.49 ± 0.10 0.980 Alb, g/dl 3.77 ± 0.05† 4.03 ± 0.17 4.16 ± 0.08 <0.001 Sodium, mEq/L 134.5 ± 0.25∗∗ 138.3 ± 2.05 137.2 ± 0.47 <0.001 CRP, mg/dl 9.05 ± 0.47 2.36 ± 1.65 6.51 ± 0.78 <0.001 WBC, ×103/μL 15.0 ± 0.55† 13.2 ± 2.18 19.6 ± 1.01 <0.001 Neutrophil, % 71.9 ± 1.40∗ 42.8 ± 5.74 59.5 ± 2.57 <0.001 Hb, g/dl 11.4 ± 0.12 11.8 ± 0.42 11.4 ± 0.18 0.680 Hct, % 34.2 ± 0.27 34.5 ± 1.15 33.2 ± 0.51 0.194 Plt, ×104/μL 38.3 ± 2.59 43.3 ± 11.2 44.5 ± 4.80 0.521 Alb, albumin; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CRP, C-reactive protein; Hb, hemoglobin; Hct, hematocrit; LDH, lactate dehydrogenase; Plt, platelet; T-Bil, total bilirubin; TP, total protein; WBC, white blood cell. The data are presented as the mean ± SE for continuous variables and as the number of patients (%) for categorical variables. The P values were obtained using an ANOVA or #Pearson's chi-squared test. ∗P < 0.001 versus VI and UTI, ∗∗P < 0.01 versus VI and UTI, and †P < 0.01 versus UTI were obtained using Tukey's HSD test. Table 2 A comparison of U-LDH and its isozyme patterns among KD, VI, and UTI groups. KD VI UTI P (n = 120) (n = 18) (n = 43) U-LDH, IU/L 35.4 ± 4.8∗ 17.0 ± 6.2 66.0 ± 8.0†† 0.002

The P values were obtained using an ANOVA or #Pearson's chi-squared test. ∗P < 0.001 versus VI and UTI, ∗∗P < 0.01 versus VI and UTI, and †P < 0.01 versus UTI were obtained using Tukey's HSD test. Table 2 A comparison of U-LDH and its isozyme patterns among KD, VI, and UTI groups. KD VI UTI P (n = 120) (n = 18) (n = 43) U-LDH, IU/L 35.4 ± 4.8∗ 17.0 ± 6.2 66.0 ± 8.0†† 0.002 U-LDH1, IU/L 12.5 ± 0.9∗† 5.5 ± 2.4 7.4 ± 1.5 0.002 U-LDH2, IU/L 8.1 ± 0.6∗ 3.9 ± 1.5 7.1 ± 0.9 0.039 U-LDH3, IU/L 4.2 ± 0.7 2.0 ± 2.0 9.5 ± 1.2†† <0.001 U-LDH4, IU/L 4.0 ± 1.2 1.6 ± 3.3 14.4 ± 2.0†† <0.001 U-LDH5, IU/L 6.6 ± 2.3 3.9 ± 6.4 27.4 ± 3.9†† <0.001 The P values were obtained using an ANOVA. ∗P < 0.05 versus VI, †P < 0.05 versus UTI, and ††P < 0.01 versus KD and VI were obtained using Tukey's HSD test. Table 3 A comparison of U-LDH and its isozyme patterns between IVIG responders and nonresponders groups with KD. IVIG responders IVIG nonresponders (n = 89) (n = 31) Kobayashi score 3.5 ± 0.2 5.6 ± 0.4∗ U-LDH, IU/L 32.0 ± 2.8 45.1 ± 4.7∗ U-LDH1, IU/L 10.9 ± 1.1 17.4 ± 1.9∗ U-LDH2, IU/L 7.1 ± 0.6 10.8 ± 1.1∗ U-LDH3, IU/L 3.9 ± 0.4 5.0 ± 0.7 U-LDH4, IU/L 3.8 ± 0.6 4.4 ± 1.0 U-LDH5, IU/L 6.4 ± 0.8 7.5 ± 1.4 ∗P < 0.05 versus IVIG responders was obtained using Student's t-test.

1. Introduction Bronchiolitis is an inflammatory disease of the smallest airways (bronchioles) and is the leading cause of respiratory distress of small children [1]. It is a clinical diagnosis, characterized by cough and respiratory distress associated with wheeze, preceded by runny nose with or without fever in young children below 2 years of age particularly between 2 and 6 months of age [1]. It is predominantly a viral disease. Respiratory Syncytial Virus (RSV) is responsible for more than 50% cases. Other agents include parainfluenza virus, adenovirus, rhinovirus, and mycoplasma [2]. There is no evidence of bacterial cause for bronchiolitis and bronchiolitis is rarely followed by bacterial superinfection [2]. Based on severity of clinical features, bronchiolitis is classified into mild, moderate, and severe [1]. Severe bronchiolitis is characterized by being unable to drink or take feed, severe respiratory distress (chest indrawing, nasal flaring, grunting, and cyanosis), and severe hypoxemia (restlessness, inconsolable cry, and SO2 <95%) [1]. About 21% of under 5 children attending different hospitals of Bangladesh have bronchiolitis [3]. Worldwide, 150 million new cases occur annually; 11–20 million (7–13%) of these are severe enough to require hospital admission. 95% of all cases occur in developing countries [4].

1. Introduction Bronchiolitis is an inflammatory disease of the smallest airways (bronchioles) and is the leading cause of respiratory distress of small children [1]. It is a clinical diagnosis, characterized by cough and respiratory distress associated with wheeze, preceded by runny nose with or without fever in young children below 2 years of age particularly between 2 and 6 months of age [1]. It is predominantly a viral disease. Respiratory Syncytial Virus (RSV) is responsible for more than 50% cases. Other agents include parainfluenza virus, adenovirus, rhinovirus, and mycoplasma [2]. There is no evidence of bacterial cause for bronchiolitis and bronchiolitis is rarely followed by bacterial superinfection [2]. Based on severity of clinical features, bronchiolitis is classified into mild, moderate, and severe [1]. Severe bronchiolitis is characterized by being unable to drink or take feed, severe respiratory distress (chest indrawing, nasal flaring, grunting, and cyanosis), and severe hypoxemia (restlessness, inconsolable cry, and SO2 <95%) [1]. About 21% of under 5 children attending different hospitals of Bangladesh have bronchiolitis [3]. Worldwide, 150 million new cases occur annually; 11–20 million (7–13%) of these are severe enough to require hospital admission. 95% of all cases occur in developing countries [4]. Risk factors implicated in the development of severe bronchiolitis include young age, male sex, parental smoking, low socioeconomic background, using of wood burning stoves, and nonbreastfeeding child [1, 5, 6]. Environmental tobacco smoke (ETS) is an important and established risk factor for both susceptibility and severity of bronchiolitis [7].

he development of severe bronchiolitis include young age, male sex, parental smoking, low socioeconomic background, using of wood burning stoves, and nonbreastfeeding child [1, 5, 6]. Environmental tobacco smoke (ETS) is an important and established risk factor for both susceptibility and severity of bronchiolitis [7]. Currently, there are 1.3 billion smokers in the world [8]. In developed countries, about 35% of men and 22% of women are daily smokers. In developing countries, about 50% of men and 9% of women are daily smokers [8]. In Bangladesh, smoking prevalence is 50% among men and 3% among women [8]. Passive smoking in the family is a major influence in the risk of lower respiratory infections in infants especially on bronchiolitis [9]. Exposure to environmental tobacco smoke had significant association with severe bronchiolitis and prolonged hospitalization [10]. From different studies and observations, it is seen that parental smoking has significant effects in the incidence and severity of acute bronchiolitis. But very few studies are available in our country in this regard. The present study was designed to determine the role of parental smoking in the development of severe bronchiolitis. 2. Materials and Methods 2.1. Place and Duration of Study The study was conducted in the Department of Paediatrics, Sylhet MAG Osmani Medical College Hospital, Bangladesh, from July 2013 to December 2015. 2.2. Design It is a case-control study. 2.3. Objective Its objective is to determine the role of parental smoking in the development of severe bronchiolitis.

2. Materials and Methods 2.1. Place and Duration of Study The study was conducted in the Department of Paediatrics, Sylhet MAG Osmani Medical College Hospital, Bangladesh, from July 2013 to December 2015. 2.2. Design It is a case-control study. 2.3. Objective Its objective is to determine the role of parental smoking in the development of severe bronchiolitis. 2.4. Diagnosis After taking a detailed history, clinical examinations were done and severity was assessed according to classification criteria. Chest radiograph was done for the evidence of air trapping in both lungs (hypertranslucency, increased interstitial markings, and hyperinflation). Complete blood counts were done in all the patients. Viral testing (polymerase chain reaction, rapid immunofluorescence, or viral culture) was not done. To differentiate bronchiolitis from pneumonia and asthma, we considered clinical features, white blood cell and differential counts, chest radiograph, and response to bronchodilator. 2.5. Treatments All patients were treated by humidified oxygen with high flow nasal cannula, parenteral fluids, and nebulized salbutamol with or without intravenous corticosteroids. No antibiotics were used.

2.4. Diagnosis After taking a detailed history, clinical examinations were done and severity was assessed according to classification criteria. Chest radiograph was done for the evidence of air trapping in both lungs (hypertranslucency, increased interstitial markings, and hyperinflation). Complete blood counts were done in all the patients. Viral testing (polymerase chain reaction, rapid immunofluorescence, or viral culture) was not done. To differentiate bronchiolitis from pneumonia and asthma, we considered clinical features, white blood cell and differential counts, chest radiograph, and response to bronchodilator. 2.5. Treatments All patients were treated by humidified oxygen with high flow nasal cannula, parenteral fluids, and nebulized salbutamol with or without intravenous corticosteroids. No antibiotics were used. 2.6. Sample and Sampling Technique Sixty-four patients admitted into the ward with severe bronchiolitis were enrolled as cases and sixty-four suitably matched apparently healthy children attending EPI (Extended Programme on Immunization) center and outpatient department presenting with nonrespiratory illness were enrolled as controls. Sample size was calculated using Guilford and Frucher formula. The technique was systematic random sampling. Every second case satisfying the inclusion and exclusion criteria was enrolled in the study.

me on Immunization) center and outpatient department presenting with nonrespiratory illness were enrolled as controls. Sample size was calculated using Guilford and Frucher formula. The technique was systematic random sampling. Every second case satisfying the inclusion and exclusion criteria was enrolled in the study. 2.7. Demographic Variables For describing socioeconomic status, we arbitrarily described low (monthly household income <10000 taka), lower-middle (10000–20000 taka), and middle (>20000 taka). The age group was subdivided into three groups: 2–6, 7–12, and 13–24 months. 2.8. Data Collection and Analysis Data were collected in a predesigned case record form. Data were processed and analyzed with the help of statistical package for social sciences [SPSS version 20].

2.7. Demographic Variables For describing socioeconomic status, we arbitrarily described low (monthly household income <10000 taka), lower-middle (10000–20000 taka), and middle (>20000 taka). The age group was subdivided into three groups: 2–6, 7–12, and 13–24 months. 2.8. Data Collection and Analysis Data were collected in a predesigned case record form. Data were processed and analyzed with the help of statistical package for social sciences [SPSS version 20]. 3. Results Sixty-four cases and sixty-four controls were enrolled for this study. The age of the patients ranged from 2 to 24 months with the mean age of 7. 53 (±4.75) months. Most frequent age group was 2 to 6 months (50%) followed by 7–12 months (39%) and 13–24 months (11%). Forty (62.5%) cases were male and 24 (37.5%) cases were female. Male-female ratio was 1.7 : 1. Most of the cases (60.95%) came from low socioeconomic background. More than half of the cases (53.13%) were not exclusively breastfed babies. Minimum length of hospital stay was 3 days and maximum was 10 days; mean hospital stay was 6.41 (±2.82) days. Baseline characteristics between cases and controls did not vary significantly regarding mean age (p = 0.452), sex (p = 0.370), socioeconomic status (p = 0.474), breastfeeding (p = 0.859), and use of wood burning stoves (p = 0.474) (Table 1).

was 3 days and maximum was 10 days; mean hospital stay was 6.41 (±2.82) days. Baseline characteristics between cases and controls did not vary significantly regarding mean age (p = 0.452), sex (p = 0.370), socioeconomic status (p = 0.474), breastfeeding (p = 0.859), and use of wood burning stoves (p = 0.474) (Table 1). Among 64 cases, 38 had history of exposure to parental smoking. All exposed cases (38) had history of only paternal smoking and there was no history of maternal smoking or both. Among 64 controls, 22 had history of exposure to parental smoking. The value was highly significant: p value = 0.005, odds ratio 2.8 (95% CI from 1.36 to 5.72). So parental smoking carried 2.8 times risk of developing severe bronchiolitis (Table 2). 4. Discussion The age of the patients ranged from 2 to 24 months: most frequent age group was 2–6 months (50%) with the mean age of 7. 53 (±4.75) months. This finding was quite consistent with a study done at the same institute where the mean age of the patients was 5.5 (±3.83) months [11]. Rida [12] found age group 1–6 months as the most frequent (60%) affected group in bronchiolitis. Kabir et al. [6] showed most of the children within 2–12 months (71.5%). Bradley et al. found age as a significant factor in severity of infection; the younger the infant, the more severe the infection [13].

3) months [11]. Rida [12] found age group 1–6 months as the most frequent (60%) affected group in bronchiolitis. Kabir et al. [6] showed most of the children within 2–12 months (71.5%). Bradley et al. found age as a significant factor in severity of infection; the younger the infant, the more severe the infection [13]. Sex distribution of the patients (male : female; 1.7 : 1) was almost similar to the study of Bashar et al. [11] who found male to female ratio 1.8 : 1. Kabir et al. [6] showed male: female ratio 2.7 : 1. Denicola [14] found males 1.6 times more likely to be hospitalized with bronchiolitis than females, male to female ratio was 1.5 : 1, and death was 1.5 times more likely in males. Semple et al. [15] found male sex significantly associated with severity of the disease. In a Canadian study, male sex was regarded as a strong and independent risk factor for Respiratory Syncytial Virus (RSV) related hospitalization [16]. The reason seems to be of anatomical nature that boys have shorter and narrower airways and are more likely to develop bronchial obstruction in case of RSV infection [17].

a Canadian study, male sex was regarded as a strong and independent risk factor for Respiratory Syncytial Virus (RSV) related hospitalization [16]. The reason seems to be of anatomical nature that boys have shorter and narrower airways and are more likely to develop bronchial obstruction in case of RSV infection [17]. In the present study, mean hospital stay was 6.41 (±2.82) days. Kabir et al. [6] found mean length of hospital stay 4.14 (±1.79) days. Bradley et al. [13] found mean hospital stay 2.5 (±2.5) days. Carroll et al. [5] found median length of hospital stay 3 days and infants with maternal smoking had higher risk of having hospitalization >3 days (hazard ratio: 1.38; 95% CI: 1.12–1.71). The prolonged duration of hospital stay found in the current study may be due to peripheral situation of the study institute where both consultant physician and parents of the patients wanted to be more ensured about recovery of the disease.

isk of having hospitalization >3 days (hazard ratio: 1.38; 95% CI: 1.12–1.71). The prolonged duration of hospital stay found in the current study may be due to peripheral situation of the study institute where both consultant physician and parents of the patients wanted to be more ensured about recovery of the disease. The present study found exposure to parental smoking in 38 (59%) cases and 26 (41%) controls (p = 0.005, odds ratio 2.8). Jones et al. [9] found in their meta-analysis that smoking by either parent or other household members increased the risk of bronchiolitis by an odds ratio of 2.51 (95% CI 1.96–3.21). Chatzimichael et al. [10] found that exposure to environmental tobacco smoke (ETS) carried 2.2 times risk of having severe bronchiolitis (odds ratio = 2.2, 95% CI 1.1–3.6). Semple et al. [15] found the infants from tobacco smoking households at increased risk of severe bronchiolitis needing supplemental oxygen (p < 0.001) and mechanical ventilation (p < 0.001). Sritippayawan et al. [18] showed environmental tobacco smoke was associated with increased risk of desaturation of the patients (p = 0.01). Rida [12] found exposure to passive smoking 2.3 times increased the risk of developing bronchiolitis.

eeding supplemental oxygen (p < 0.001) and mechanical ventilation (p < 0.001). Sritippayawan et al. [18] showed environmental tobacco smoke was associated with increased risk of desaturation of the patients (p = 0.01). Rida [12] found exposure to passive smoking 2.3 times increased the risk of developing bronchiolitis. In the present study, all exposed cases of severe bronchiolitis (38 out of 64) had history of only paternal smoking and there was no history of maternal smoking or both. Jones et al. [9] found odds ratio 1.22 (95% CI 1.10 to 1.35) for paternal smoking and 1.62 (95% CI 1.38 to 1.89) for both parental smoking. Strachan and Cook [19] described a causal relationship between parental smoking and acute lower respiratory illness where odds ratios were 1.57 (95% CI 1.42 to 1.74) for smoking by either parent and 1.72 (95% CI 1.55 to 1.91) for maternal smoking. Schvartsman et al. [20] found that children were more affected by maternal smoking (p = 0.00016) than paternal one (p = 0.015). Jurado et al. [21] described a greater influence of exposure to maternal smoking than paternal smoking in the development of respiratory symptoms in young children. Limitation of this study was not to estimate urinary “cotinine” level. Cotinine, a major metabolite of nicotine, has been used as a biological marker of smoke absorption to strengthen the evidence of exposure to tobacco smoke

In the present study, all exposed cases of severe bronchiolitis (38 out of 64) had history of only paternal smoking and there was no history of maternal smoking or both. Jones et al. [9] found odds ratio 1.22 (95% CI 1.10 to 1.35) for paternal smoking and 1.62 (95% CI 1.38 to 1.89) for both parental smoking. Strachan and Cook [19] described a causal relationship between parental smoking and acute lower respiratory illness where odds ratios were 1.57 (95% CI 1.42 to 1.74) for smoking by either parent and 1.72 (95% CI 1.55 to 1.91) for maternal smoking. Schvartsman et al. [20] found that children were more affected by maternal smoking (p = 0.00016) than paternal one (p = 0.015). Jurado et al. [21] described a greater influence of exposure to maternal smoking than paternal smoking in the development of respiratory symptoms in young children. Limitation of this study was not to estimate urinary “cotinine” level. Cotinine, a major metabolite of nicotine, has been used as a biological marker of smoke absorption to strengthen the evidence of exposure to tobacco smoke 5. Conclusion Exposure to parental smoking, particularly paternal smoking, causes a statistically significant (p = 0.005, odds ratio = 2.8) increase in the risk of developing severe bronchiolitis in the first year of life. Protecting young children from parental smoking should be the important approach to prevent the morbidity and mortality caused by severe bronchiolitis. Competing Interests The authors declare that they have no competing interests.

5. Conclusion Exposure to parental smoking, particularly paternal smoking, causes a statistically significant (p = 0.005, odds ratio = 2.8) increase in the risk of developing severe bronchiolitis in the first year of life. Protecting young children from parental smoking should be the important approach to prevent the morbidity and mortality caused by severe bronchiolitis. Competing Interests The authors declare that they have no competing interests. Authors' Contributions Dr. Rubina Farzana (First author) did data collection and drafted the manuscript. Dr. Mujibul Hoque (Second author) designed and reviewed the manuscript. Dr. Mohammad Shah Kamal (Third author) did statistical analysis. Dr. Md. Moseh Uddin Choudhury (Fourth author) reviewed the manuscript. Table 1 Baseline characteristics of the cases and controls (n = 64). Characteristics Cases Controls p value (1) Total number 64 64 (2) Age in months     0.452∗ Mean (±SD) 7.53 (±4.75) 8.78 (±5.64) (3) Sex     0.370† Male 40 (62.5%) 35 (54.69%) Female 24 (37.5%) 29 (45.31%) (4) Socioeconomic status     0.474† Low 39 (60.94%) 35 (54.69%) Lower-middle 25 (39.06%) 29 (45.31%) (5) Breast feeding     0.859† Exclusively breastfed 30 (46.88%) 29 (45.31%) Not exclusively breastfed 34 (53.13%) 35 (54.69%) (6) Wood-burning stoves users     0.474† Yes 39 (60.94%) 35(54.69%) No 25 (39.06%) 29 (45.31%) ∗Unpaired “t” test was employed to analyze the data. †Chi-Square (χ2) test was employed to analyze the data. Table 2 Exposure to parental smoking. Parental smoking Cases Controls P value Odds ratio Yes 38 (59%) 22 (34%) 0.005 2.8 No 26 (41%) 42 (66%)

1. Introduction Acute poisoning in the paediatric age group is an important cause of preventable mortality and morbidity. Globally, every year 3000 children who are less than 14 years of age die following acute poisoning [1]. Children between 1 and 5 years of age have the highest risk for unintentional poisoning [2]. The type of poisons could be one of household poisons, plants, medicines, pesticides, or miscellaneous agents, and plants are identified as an important cause of paediatric poisoning reported to poison information centers [3]. Patterns of plant poisoning vary globally with varying botanical, geographical, and sociocultural characteristics among different populations [4].

, plants, medicines, pesticides, or miscellaneous agents, and plants are identified as an important cause of paediatric poisoning reported to poison information centers [3]. Patterns of plant poisoning vary globally with varying botanical, geographical, and sociocultural characteristics among different populations [4]. Plants remain to be an important cause of mortality among adults in Sri Lanka. Two studies from Northern Sri Lanka reported Thevetia peruviana (yellow Oleander) as the commonest plant poison (17% of cases) and it was associated with a case fatality rate of 6-7% [5]. Similar observations were made in subsequent studies from Southern Sri Lanka [6]. Further studies on adult patients from North-Central province showed the unfavorable trend having 32–36% of patients with self-poisoning ingesting T. peruviana with a case fatality rate approximating 4% [7]. Studies on plant poisoning involving Sri Lankan children however are limited. A study from Western Sri Lanka [8] identified plants as an important cause of unintentional poisoning in paediatric age group two decades ago. A prospective study (1984–2001) based on predominantly an urban population in Sri Lanka reported that plants accounted for 10% of poisoning events with a case fatality rate of 0.8% [9]. Up to date there has been no evidence from either prospective or retrospective studies with regard to plant poisoning among children from rural Sri Lanka. This multicenter study has described the patterns of plant poisoning in North-Central province of Sri Lanka (NCP) which accommodates a predominant rural community.

. Up to date there has been no evidence from either prospective or retrospective studies with regard to plant poisoning among children from rural Sri Lanka. This multicenter study has described the patterns of plant poisoning in North-Central province of Sri Lanka (NCP) which accommodates a predominant rural community. 2. Methods This multicenter prospective study included 36 hospitals in the North-Central province of Sri Lanka (Anuradhapura Teaching hospital (THA), Polonnaruwa District General hospital (THP), and 34 base/district/rural hospitals of the province under RDHS, Regional Director of Health Services). The study was conducted in four major arms: (1) a two-year prospective study (2012–2014) at Anuradhapura Teaching Hospital, (2) a two-year prospective study (2012–2014) at Polonnaruwa general hospital, (3) one-year prospective study involving 34 hospitals under RDHS of NCP (2013-2014), and (4) a five-year retrospective study at Anuradhapura Teaching Hospital (2007–2012). The major part of data collections was carried out at Anuradhapura hospital prospectively by the same investigator using a pretested, multistructured, interviewer administered questionnaire that comprehensively assessed clinical profiles, plant poison related factors, clinical management, complications, and outcome following acute paediatric poisoning.

was carried out at Anuradhapura hospital prospectively by the same investigator using a pretested, multistructured, interviewer administered questionnaire that comprehensively assessed clinical profiles, plant poison related factors, clinical management, complications, and outcome following acute paediatric poisoning. Both children with intentional and unintentional poisoning were included in the assessment. Children aged 9 months to 12 years were considered for the analysis. Acute poisoning due to nonplant poisons (household poisons/medicines/pesticides), food poisons, snake envenomation, allergic reactions, and adverse drug reactions which can be considered in the purview of toxicology was omitted in the study. Also children with doubtful poisoning where there was no clear aetiology were excluded from the study. All three prospective studies identified clinical presentations, reasons for delayed management in addition to demographic data. A prospective controlled risk factor study included all children who presented with plant poisoning to Anuradhapura Teaching Hospital over the two-year study period (2012–2014). The controls were selected from the same hospital and children who presented with acute medical illnesses were recruited as controls. The acute medical illnesses considered included viral fever, acute upper respiratory tract infection, and urticaria. All other acute conditions including nonspecific symptoms without a definitive diagnosis were excluded. All children were matched for age and gender on a case by case basis.

s were recruited as controls. The acute medical illnesses considered included viral fever, acute upper respiratory tract infection, and urticaria. All other acute conditions including nonspecific symptoms without a definitive diagnosis were excluded. All children were matched for age and gender on a case by case basis. Data collections in all components of the current study were subjected to independent audit and close monitoring by South Asian Clinical Toxicology Research Collaboration (SACTRC) and the investigators of the study themselves. Ethical clearance for the study was issued by ethical review committees, faculty of medicine, University of Kelaniya, and Rajarata University of Sri Lanka. 3. Results There were 325 incidents of plant poisoning reported in all arms of the study. Male children outnumbered female children in all studies and amounted to 186 (57%). Sixty-seven percent of children were less than five years of age. However, plant poisoning was uncommon among children who were less than two years of age (4.6% of total) compared to older children. The majority of poisoning events were secondary to unintentional ingestion of the poison (314/325, 96.6%). Mortality rate was 1.2% (4 cases) and all four cases (100%) followed ingestion of lethal dose of Oleander. Sixty-six percent of children (216/325) were transferred from a local hospital (under RDHS) to a tertiary care hospital following the poisoning event. Table 1 has compared the demographic characteristics, patterns of poisoning, and transfer rates of children in different arms of the study.

n of lethal dose of Oleander. Sixty-six percent of children (216/325) were transferred from a local hospital (under RDHS) to a tertiary care hospital following the poisoning event. Table 1 has compared the demographic characteristics, patterns of poisoning, and transfer rates of children in different arms of the study. Jatropha curcas (Barbados nut/physic nut/poison nut) was the commonest plant poison ingested by children and it was consistently seen in all studies. Poisoning with Thevetia peruviana (Oleander) and Abrus precatorius (rosary pea/paternoster pea/wild licorice) was commonly seen after Jatropha curcas. Poisoning with those three plants comprised 83.4% of total plant poisonings. Patterns of plant poisoning were similar and consistent in all studies. Table 2 has presented all children with poisoning based on the plant ingested. Commonest route of poisoning was ingestion (323/325, 99.4%). One child had symptoms following contact of the poison with mucus membranes. Another child indirectly ingested the poison indirectly through breast milk following her mother intentionally ingesting Oleander.

Jatropha curcas (Barbados nut/physic nut/poison nut) was the commonest plant poison ingested by children and it was consistently seen in all studies. Poisoning with Thevetia peruviana (Oleander) and Abrus precatorius (rosary pea/paternoster pea/wild licorice) was commonly seen after Jatropha curcas. Poisoning with those three plants comprised 83.4% of total plant poisonings. Patterns of plant poisoning were similar and consistent in all studies. Table 2 has presented all children with poisoning based on the plant ingested. Commonest route of poisoning was ingestion (323/325, 99.4%). One child had symptoms following contact of the poison with mucus membranes. Another child indirectly ingested the poison indirectly through breast milk following her mother intentionally ingesting Oleander. 3.1. Comparison of Clinical Manifestations and Reasons for Delayed Presentations to Primary Care Hospital 160 children recruited to studies at THA, THP, and RDHS were available for the analysis. Gastrointestinal symptoms (125 children, 78.1%) were the predominant symptoms following plant toxin ingestion and they were consistently seen in all three studies. Most gastrointestinal symptoms occurred following Jatropha circus intoxication. All children with cardiovascular symptoms were intoxicated with Oleander. Table 3 illustrates the variability in clinical manifestations in detail.

mptoms following plant toxin ingestion and they were consistently seen in all three studies. Most gastrointestinal symptoms occurred following Jatropha circus intoxication. All children with cardiovascular symptoms were intoxicated with Oleander. Table 3 illustrates the variability in clinical manifestations in detail. Fifty-two children (32.5%) presented to primary care hospital at least two hours after the ingestion of the poison. Commonest reason for delayed presentation was lack of concern regarding urgency of the situation, 33 children (20.6%). Detailed analysis of reasons for delayed presentation to primary care unit is presented in Table 4. 3.2. Detailed Evaluation of Patterns and Risk Factors of Plant Poisoning among Children at THA Sixty-five children presented to THA following plant poisoning over the two-year study period. Mean age of children was 4.3 years (range: 11 months–12 years). Most parents had received secondary education, 50 fathers (77%) and 54 mothers (83.1%). The majority of fathers were engaged in farming (21, 32.3%), defense service (11, 16.9%), and manual labour (11, 16.9%). Most mothers were housewives (45, 69.2%). Most of the poisoning events occurred in home garden (43, 66.1%) followed by cultivation area (12, 18.4%) and inside of home (5, 7.7%). Harmful first-aid measures were practiced in 19 children (29.2%). The commonest measure was forceful ingestion of water (8, 12.3%) and it was followed by forceful finger insertion (3, 4.6%), forceful milk (2, 3.1%), and coconut milk ingestion (2, 3.1%).

3.2. Detailed Evaluation of Patterns and Risk Factors of Plant Poisoning among Children at THA Sixty-five children presented to THA following plant poisoning over the two-year study period. Mean age of children was 4.3 years (range: 11 months–12 years). Most parents had received secondary education, 50 fathers (77%) and 54 mothers (83.1%). The majority of fathers were engaged in farming (21, 32.3%), defense service (11, 16.9%), and manual labour (11, 16.9%). Most mothers were housewives (45, 69.2%). Most of the poisoning events occurred in home garden (43, 66.1%) followed by cultivation area (12, 18.4%) and inside of home (5, 7.7%). Harmful first-aid measures were practiced in 19 children (29.2%). The commonest measure was forceful ingestion of water (8, 12.3%) and it was followed by forceful finger insertion (3, 4.6%), forceful milk (2, 3.1%), and coconut milk ingestion (2, 3.1%). The majority of children had onset of symptoms within one hour from the time of poisoning event (35, 53.8%). Though most children (45, 69.2%) were brought to primary care unit within two hours from the poisoning event, twenty children (31.8%) presented at least 2 hours after the poison was ingested (range: 2 to 8 hours). Emesis induction was offered to 32 children (49.2%). Two children (3.1%) required prescription of antidotes and management in an intensive care unit. Reported medical complications included cardiac arrhythmia (2) (3.1%), severe dehydration (1) (1.5%), hematemesis (1) (1.5%), and seizures (1) (1.5%).

nge: 2 to 8 hours). Emesis induction was offered to 32 children (49.2%). Two children (3.1%) required prescription of antidotes and management in an intensive care unit. Reported medical complications included cardiac arrhythmia (2) (3.1%), severe dehydration (1) (1.5%), hematemesis (1) (1.5%), and seizures (1) (1.5%). Risk factor evaluation showed four proposed risk factors which were associated with significantly elevated risk (p < 0.001) of plant poisoning among children. They were the presence of poisoning pants in home garden, inadequate supervision of the child, past history of poisoning, and parents' subjective feeling of lack of family support to look after children. Five proposed risk factors did not reveal a significant association with plant poisoning (working mother, children with deprived schooling, young mother, poorly educated mother, farming parents, and parents' subjective perception of having economic problems). Table 5 has compared the presence of proposed risk factors in the two groups.

k factors did not reveal a significant association with plant poisoning (working mother, children with deprived schooling, young mother, poorly educated mother, farming parents, and parents' subjective perception of having economic problems). Table 5 has compared the presence of proposed risk factors in the two groups. 4. Discussion Plant poisoning in children is one of the common presentations to emergency departments. However, the risk factors and circumstances of these poisoning events are largely underexplored in international literature [10] and most studies have concentrated on data reported to poison information centers. The pattern of plant poisoning varies considerably based on age, sex, and sociodemographic factors among different geographic regions [11]. Young children comprise a unique group in that they have the natural curiosity about their surroundings and tendency for oral exploration keeping them highly vulnerable to unintentional poisoning [12]. The current study identified male children more vulnerable than female children for plant poisoning. Furthermore, it also revealed that the majority of children were between 2 and 5 years of age. Unintentional poisoning accounted for more than 90% of poisoning events and almost all cases had taken the poison through ingestion. Similar observations are made in previously published studies from Sri Lanka [9], India [10], Thailand [13], and Taiwan [14].

t the majority of children were between 2 and 5 years of age. Unintentional poisoning accounted for more than 90% of poisoning events and almost all cases had taken the poison through ingestion. Similar observations are made in previously published studies from Sri Lanka [9], India [10], Thailand [13], and Taiwan [14]. A key observation in the current study was the higher transfer rate of children. The study identified that 66.4% of total population were transferred between two hospitals and 67.2% of children who attended local hospitals under RDHS were subsequently transferred to a tertiary hospital for further management. Previously published adult studies in the same region of Sri Lanka observed a transfer rate of 50% [15] which is lower than current figures. Despite higher transfer rates, case fatality and complication rates in current study were significantly lower than adult studies [7]. Transferring of patients significantly adds to healthcare expenditure and the average patient cost per transfer was US$ 14.03 in one study eight years before in the same region [16]. These facts reveal the value of increased awareness among healthcare workers in rural territories regarding the nature and outcome of plant poisoning in the paediatric age group which is mostly benign. Effective triage and limitation of transfers only to needy children would likely cut down healthcare expenditure, duration of hospital stay, and effect on families of children with plant poisoning.

rural territories regarding the nature and outcome of plant poisoning in the paediatric age group which is mostly benign. Effective triage and limitation of transfers only to needy children would likely cut down healthcare expenditure, duration of hospital stay, and effect on families of children with plant poisoning. We observed a higher percentage of deliberate ingestions compared to a previously published study in urban Sri Lanka [9]. In that study, the common plant poisons were Jatropha circus, Ricinus communis, and Jatropha multifida (physic nut/nettle spurge) accounting for 43% of total plant poisonings. In the current study based on NCP, Jatropha circus, Oleander, and Abrus precatorius were identified in 83.4% of all plant poisonings. All deliberate poisonings happened by ingestion of Oleander. These poisonous plants were usually found in fences of houses, by sides of walking paths and canals. Some were in fact grown as house plants by parents. Higher proportion of deliberate poisoning with Oleander raises the question as to how they learn these behaviours given the fact that NCP has higher numbers of suicides among adults [1].

nts were usually found in fences of houses, by sides of walking paths and canals. Some were in fact grown as house plants by parents. Higher proportion of deliberate poisoning with Oleander raises the question as to how they learn these behaviours given the fact that NCP has higher numbers of suicides among adults [1]. It was also revealed that 78.5% of children had ingested plant poisons within their own home gardens or neighborhoods. This figure of 78.5% is much higher compared to previous Sri Lankan studies conducted in more urban areas [9] and underpins hazardous nature of home gardens and neighborhoods of remote villages which accommodate poisonous plants without being noticed by house owners. Higher rates of transfers from many rural territories within the province compared to other poison types [17] also highlight widespread distribution of environmental risk factors within the community.

gardens and neighborhoods of remote villages which accommodate poisonous plants without being noticed by house owners. Higher rates of transfers from many rural territories within the province compared to other poison types [17] also highlight widespread distribution of environmental risk factors within the community. Home garden was the commonest place for poisoning with plant seeds in current study. A study from urban Sri Lanka observed Alocasia (elephant's ear plant) as the commonly encountered plant poison in home garden [18]. In our study, commonly encountered plant poisons in home gardens were Jatropha circus, Abrus precatorius, and Oleander. Home gardens in remote areas of NCP are less cleared allowing growth of these plants and most parents were unaware of the potential toxicity. Jatropha circus was identified as the commonest poison in current study and Jatropha circus induced gastrointestinal symptoms were the commonest clinical presentation. The same observation has been made in studies from other parts of Sri Lanka [9] and Asia [13]. Patterns of poisoning and as well as subsequent outcome are always related to the underlying sociocultural circumstances. Observation of scientifically unproven yet culturally based first-aid practices by some parents as observed in this study can be associated with detrimental effects. Therefore providing knowledge to at-risk communities regarding such issues is helpful in bringing down childhood poisoning related morbidity and mortality.

bservation of scientifically unproven yet culturally based first-aid practices by some parents as observed in this study can be associated with detrimental effects. Therefore providing knowledge to at-risk communities regarding such issues is helpful in bringing down childhood poisoning related morbidity and mortality. Delayed presentation to primary care hospital following the poisoning event has a potentially strong negative impact on effective management and patient outcomes [19]. In the current study, the commonest reason for delayed presentation had been lack of concern regarding the urgency for seeking medical care. Child remaining asymptomatic after ingestion, lack of identity of the ingested substance as a poison, and small ingested amount of poison had kept a higher threshold for some parents to seek urgent medical attention. The duration of hospital stay and severity of complications have been shown to have a direct correlation with lag time in reaching the hospital following the poisoning event [20]. These facts reveal the value of community awareness in seeking early primary care in improving patient outcomes.

k urgent medical attention. The duration of hospital stay and severity of complications have been shown to have a direct correlation with lag time in reaching the hospital following the poisoning event [20]. These facts reveal the value of community awareness in seeking early primary care in improving patient outcomes. The current study identified four factors including presence of poisoning plants in home garden, inadequate supervision, past history of poisoning, and parents' subjective feeling of lack of family support to look after children as being associated specifically with significantly elevated risk of plant poisoning in children. Schmertmann et al. reported inadequate supervision as a risk factor for unintentional poisoning in the paediatric age group [21]. Similarly, past history of poisoning has been identified as a risk factor for acute poisoning [22]. However, evidence regarding risk factors for plant poisoning by properly controlled studies is scant in currently available literature and further studies are needed. As majority of plant poisoning occurred within home premises in the current study, a holistic approach which targets home environment would help in managing the burden of poisoning.

risk factors for plant poisoning by properly controlled studies is scant in currently available literature and further studies are needed. As majority of plant poisoning occurred within home premises in the current study, a holistic approach which targets home environment would help in managing the burden of poisoning. 5. Conclusion The study revealed harmful effects of traditional first-aid practices which are detrimental to health of the child. Children become victims of plant poisoning mostly secondary to presence of poisoning plants in home garden, inadequate supervision by care givers, and previous poisoning. As these risk factors are significantly associated with plant poisoning, the effect of community education to enhance vigilance and assurance of safe environment should be evaluated. The study also identified the importance of awareness among healthcare workers regarding the mostly benign nature of plant poisoning in children in view of reducing expenditure on patient management. Acknowledgments The authors of this study acknowledge Dr. Suneth Agampodi, Head of Department of Community Medicine, and Dr. Lalith Senarathna, Senior Lecturer, Faculty of Applied Sciences, Rajarata University of Sri Lanka, for providing technical advice in data analysis, and Dr. Thilini Hemachandra and Dr. Chamila Dissanayaka of Anuradhapura Teaching Hospital, Sri Lanka, for providing support in entering data into statistical databases.

Dr. Lalith Senarathna, Senior Lecturer, Faculty of Applied Sciences, Rajarata University of Sri Lanka, for providing technical advice in data analysis, and Dr. Thilini Hemachandra and Dr. Chamila Dissanayaka of Anuradhapura Teaching Hospital, Sri Lanka, for providing support in entering data into statistical databases. Ethical Approval The study was granted ethical approval by ethical review committees of Faculties of Medicine, University of Kelaniya, and Rajarata University of Sri Lanka. Consent Parents of all participants gave written consent for participation of their children in the study and publication of results. Conflicts of Interest The authors declare that they have no conflicts of interest. Authors' Contributions M. B. Kavinda Chandimal Dayasiri designed the study, carried out data collection following appropriate methodology, analysed data, and wrote the manuscript. Shaluka F. Jayamanne and Chamilka Y. Jayasinghe designed the study, analysed data, and supervised manuscript writing process. Disclosure No external funding was obtained for this study. Table 1 Demographic characteristics, patterns of poisoning, and transfer rates of children with plant poisoning.

Authors' Contributions M. B. Kavinda Chandimal Dayasiri designed the study, carried out data collection following appropriate methodology, analysed data, and wrote the manuscript. Shaluka F. Jayamanne and Chamilka Y. Jayasinghe designed the study, analysed data, and supervised manuscript writing process. Disclosure No external funding was obtained for this study. Table 1 Demographic characteristics, patterns of poisoning, and transfer rates of children with plant poisoning. Variable Retrospective study (n = 165) THA Study (n = 65) Polonnaruwa study (n = 58) Peripheral study (n = 37) Total (n = 325) (1) Male : female 90 : 59 4.5% : 45.5% 38 : 27 58.4% : 1.6% 38 : 20 65.5% : 34.5% 20 : 17 54% : 46% 186 : 139 57% : 43% (2) <5 years : >5 years 99 : 66 60% : 40% 52 : 13 80% : 20% 26 : 32 44.8% : 55.2% 31 : 6 84.8% : 15.2% 208 : 117 64% : 36% (3) Unintentional : intentional 162 : 3 98% : 2% 60 : 5 92% : 8% 57 : 1 98.3% : 1.7% 35 : 2 95.6% : 5.4% 314 : 11 96.6% : 3.4% (4) Mortality 2 (1.2%) 1 (1.5%) 1 (1.7%) — 4 (1.2%) (5) Commonest poison Jatropha circus Jatropha circus Jatropha circus Jatropha circus Jatropha circus (6) Transfer rate 108 (65.4%) 49 (75.3%) 34 (58.2%) 25 (67.6%) 216 (66.4%) Table 2 Patterns of poisoning with poisonous plants among children in North-Central province.

Variable Retrospective study (n = 165) THA Study (n = 65) Polonnaruwa study (n = 58) Peripheral study (n = 37) Total (n = 325) (1) Male : female 90 : 59 4.5% : 45.5% 38 : 27 58.4% : 1.6% 38 : 20 65.5% : 34.5% 20 : 17 54% : 46% 186 : 139 57% : 43% (2) <5 years : >5 years 99 : 66 60% : 40% 52 : 13 80% : 20% 26 : 32 44.8% : 55.2% 31 : 6 84.8% : 15.2% 208 : 117 64% : 36% (3) Unintentional : intentional 162 : 3 98% : 2% 60 : 5 92% : 8% 57 : 1 98.3% : 1.7% 35 : 2 95.6% : 5.4% 314 : 11 96.6% : 3.4% (4) Mortality 2 (1.2%) 1 (1.5%) 1 (1.7%) — 4 (1.2%) (5) Commonest poison Jatropha circus Jatropha circus Jatropha circus Jatropha circus Jatropha circus (6) Transfer rate 108 (65.4%) 49 (75.3%) 34 (58.2%) 25 (67.6%) 216 (66.4%) Table 2 Patterns of poisoning with poisonous plants among children in North-Central province. Poisonous Plant Retrospective study (n = 165) Prospective study (n = 65) Polonnaruwa study (n = 58) Peripheral study (n = 37) Total (n = 325) (1) Jatropha curcas 81 (49.1%) 22 (33.8%) 24 (41.3%) 16 (43.2 %) 143 (44%) (2) Thevetia peruviana 36 (21.8%) 10 (15.4%) 14 (24.1%) 8 (21.6%) 68 (20.9%) (3) Abrus precatorius 27 (16.4%) 15 (23%) 11 (19%) 7 (18.9%) 60 (18.5%) (4) Ricinus communis 9 (5.4%) 5 (7.7%) 1 (1.7%) 2 (5.4%) 17 (5.2%) (5) Jatropha multifida 6 (3.6%) 4 (6.2%) 2 (3.3%) — (0%) 12 (3.7%) (6) Caladium 1 (0.6%) 3 (4.6%) 2 (3.3%) 2 (5.4%) 8 (2.5%) (7) Datura stramonium 1 (0.6%) 2 (3.1 %) 1 (1.7%) 1 (2.7%) 5 (1.5%) (8) Gloriosa superba 1 (0.6%) 2 (3.1%) 1 (1.7%) — (0%) 4 (1.2%) (9) Other 3 (1.8%) 2 (3.1%) 2 (3.3%) 1 (2.7%) 8 (2.4 %) Table 3 Clinical manifestations of plant poisoning among children in rural Sri Lanka.

m 1 (0.6%) 3 (4.6%) 2 (3.3%) 2 (5.4%) 8 (2.5%) (7) Datura stramonium 1 (0.6%) 2 (3.1 %) 1 (1.7%) 1 (2.7%) 5 (1.5%) (8) Gloriosa superba 1 (0.6%) 2 (3.1%) 1 (1.7%) — (0%) 4 (1.2%) (9) Other 3 (1.8%) 2 (3.1%) 2 (3.3%) 1 (2.7%) 8 (2.4 %) Table 3 Clinical manifestations of plant poisoning among children in rural Sri Lanka. Reasons for delayed presentation THA THP RDHS Total (1) Gastrointestinal symptoms 59 (90.1%) 41 (70.6%) 25 (67.5%) 125 (78.1%) (2) Respiratory symptoms 25 (38.4%) 22 (37.9%) 16 (43.2%) 63 (39.3%) (3) Cardiovascular symptoms 3 (4.6%) 6 (10.3%) 2 (6.4%) 11 (6.8%) (4) Neurological symptoms 2 (3%) 1 (1.7%) — 3 (1.8%) Table 4 Reasons for delayed presentation to primary care unit among children in rural Sri Lanka. Reasons for delayed presentation THA THP RDHS Total (1) Lack of concern regarding need for urgent care 12 (18.4%) 11 (18.9%) 10 (27%) 33 (20.6%) (2) Lack of transport facilities for emergency management 8 (12.3%) 7 (12%) 12 (32.4%) 25 (15.6%) (3) Lack of knowledge regarding possible complications 6 (9.2%) 6 (10.3%) 4 (10.8%) 16 (10%) (4) Lack of financial resources 6 (9.2%) 4 (6.8%) 2 (5.4%) 12 (7.5%) (5) Poisoning event unrevealed until symptoms occurred 2 (3.1%) 1 (1.7%) 1 (2.7%) 4 (2.5%) (6) Delayed attention by medical team 1 (1.5%) — — 1 (0.6%) Table 5 Analysis of proposed risk factors in case-control study.

(9.2%) 6 (10.3%) 4 (10.8%) 16 (10%) (4) Lack of financial resources 6 (9.2%) 4 (6.8%) 2 (5.4%) 12 (7.5%) (5) Poisoning event unrevealed until symptoms occurred 2 (3.1%) 1 (1.7%) 1 (2.7%) 4 (2.5%) (6) Delayed attention by medical team 1 (1.5%) — — 1 (0.6%) Table 5 Analysis of proposed risk factors in case-control study. Proposed risk factor Cases Controls Chi square value p value (1) Poisonous plants in home garden 54 6 71.31 <0.001 (2) Inadequate supervision of the child 49 13 39.96 <0.001 (3) Past history of poisoning 24 2 23.26 <0.001 (4) Lack of family support 33 12 14.98 <0.001 (5) Mother working during the daytime 17 15 0.16 0.68 (6) Lack of schooling/education in mother 2 1 0.34 0.55 (7) Young mother (<19 years) 7 9 0.28 0.59 (8) Primary level education in mother 10 7 0.60 0.43 (9) Parents from farming community 22 19 0.32 0.57 (10) Economic problems 21 20 0.03 0.85

1. Introduction Literature data show that children with chronic pain often encounter significant delays between their initial pain complaints and the time they eventually consult a pain specialist [1–3]. Zernikow et al. have shown that children consulted on average 3 physicians prior to being referred to a pain specialist with most children having visited a doctor 1 to 5 times and about 13% of them having had a very high number of previous treatments and consultations (≥6) [3]. Konijnenberg et al. have reported a 12-month median duration of pain related symptoms and a median of 2 different physicians' consultations before children were referred to an academic pediatric specialist [4]. Reviews of trends in medicalization of children with chronic pain in the United States showed a longer duration of symptoms (18–24 months) and a higher level of medicalization, including a higher number of professionals seen (>8) compared to European studies [5, 6]. The reasons for these findings are still unclear. The current literature does not address how often an accurate diagnosis is made by general practitioners and nonpain specialist and whether the psychological and social components of pain are taken in consideration in the care plan. It is also unclear how the evaluation by a pain specialist can affect patients' outcomes, particularly in the presence of data showing that adherence to pain physicians' recommendations is inconsistent in part due to a negative attitude towards specific recommendations, in particular psychological interventions [7, 8].

It is also unclear how the evaluation by a pain specialist can affect patients' outcomes, particularly in the presence of data showing that adherence to pain physicians' recommendations is inconsistent in part due to a negative attitude towards specific recommendations, in particular psychological interventions [7, 8]. The aim of this retrospective review was twofold. We analyzed the referral history for pediatric patients seen at our pain clinic, looking at the time interval between initial pain complaints and pain clinic consultation, including the number of physicians who consulted these patients and the referral diagnosis. We then analyzed the results of a multimodal approach to manage patients' symptoms and examined patients' compliance with the recommended treatments. 2. Methods The study was reviewed and approved by the Institutional Review Board. We reviewed the medical record of patients seen in the pain clinic of an academic institution between October 2014 and June 2016. The inclusion criteria were children aged 3–20 years diagnosed with chronic headaches, neuropathic pain, and muscle-skeletal and chronic abdominal pain and patients whose prior medical record and note from the referring physicians were available. We excluded patients with cancer pain or acute onset pain defined as any pain lasting less than 12 weeks [9] and whose prior medical records were not available.

neuropathic pain, and muscle-skeletal and chronic abdominal pain and patients whose prior medical record and note from the referring physicians were available. We excluded patients with cancer pain or acute onset pain defined as any pain lasting less than 12 weeks [9] and whose prior medical records were not available. The pain clinic was staffed by a physician (anesthesiologist), a psychiatrist, a nurse practitioner, and a psychologist. A physical therapist and an occupational therapist were available for consultation as needed. The duration of the initial encounter ranged between 90 and 120 minutes. During this time, Complete past medical, surgical, and family history were taken. This was followed by a psychological evaluation and a physical examination. Patients with neuromuscular deficits were also evaluated by physical therapists. The following data were extracted from the patients' electronic medical record: sociodemographic data, race and ethnicity, diagnoses, history of health care contacts including a list of consulting physicians, therapeutic procedures including surgeries, and ability of patient to practice sports and attend classes at school on regular basis. We defined patients as unable to attend regular school classes when they were homeschooled because of pain symptoms and enrolled in online school or home-hospital programs and those who had missed more than 10% of school days during the previous school year, based on the California Department of Education policy [10].

The duration of the initial encounter ranged between 90 and 120 minutes. During this time, Complete past medical, surgical, and family history were taken. This was followed by a psychological evaluation and a physical examination. Patients with neuromuscular deficits were also evaluated by physical therapists. The following data were extracted from the patients' electronic medical record: sociodemographic data, race and ethnicity, diagnoses, history of health care contacts including a list of consulting physicians, therapeutic procedures including surgeries, and ability of patient to practice sports and attend classes at school on regular basis. We defined patients as unable to attend regular school classes when they were homeschooled because of pain symptoms and enrolled in online school or home-hospital programs and those who had missed more than 10% of school days during the previous school year, based on the California Department of Education policy [10]. Patients were asked about the time interval between the onset of pain symptoms and the referral to this pediatric pain clinic. A review of the patients' prior medical records and referral documentation confirmed the data's accuracy. We documented the type and number of providers consulted because of pain and we recorded whether patients were referred by a practitioner working at this institution or a physician practicing in the community.

nic. A review of the patients' prior medical records and referral documentation confirmed the data's accuracy. We documented the type and number of providers consulted because of pain and we recorded whether patients were referred by a practitioner working at this institution or a physician practicing in the community. Every patient was evaluated by a psychologist following a clinical interview assessing mental disorder classes and subcategories according to the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, Text Revision (DSM-IV-TR) [11] including depression, anxiety, behavior disorders, developmental disabilities, somatoform, substance use, and eating disorders. These interviews were conducted at the time of the patients' initial evaluation as well as during follow-up visits. The recommendations made to patients and families included medications, psychotherapy, and Complementary and Alternative Medicine (CAM) interventions including acupuncture, yoga, biofeedback, massage therapy, and occupational therapy interventions. Nerve blocks and local or joint steroid infiltrations were also recommended when clinically indicated. Patients with Complex Regional Pain Syndrome and those unable to ambulate for chronic pain were admitted in the hospital rehabilitation unit for intense physical and occupational therapy. Patients with history of opioid abuse were referred to the hospital substance abuse program specialists.

when clinically indicated. Patients with Complex Regional Pain Syndrome and those unable to ambulate for chronic pain were admitted in the hospital rehabilitation unit for intense physical and occupational therapy. Patients with history of opioid abuse were referred to the hospital substance abuse program specialists. Patients were asked to follow up with the clinic within a month after the initiation of the recommended treatments. The following appointments were scheduled based on the response to the treatments. Phone calls were made to remind patients of their clinical appointments. Treatments including psychotherapy and physical and occupational therapy as well as CAM interventions were done at this institution or in the community based on patients' insurance requirements and their physical address. We defined clinical improvement in our population as the ability of patients to go back to school full time as well as being able to resume sport activities that were interrupted because of pain [12, 13].

Patients were asked to follow up with the clinic within a month after the initiation of the recommended treatments. The following appointments were scheduled based on the response to the treatments. Phone calls were made to remind patients of their clinical appointments. Treatments including psychotherapy and physical and occupational therapy as well as CAM interventions were done at this institution or in the community based on patients' insurance requirements and their physical address. We defined clinical improvement in our population as the ability of patients to go back to school full time as well as being able to resume sport activities that were interrupted because of pain [12, 13]. 2.1. Statistical Analysis STATA/IC 14 software was used for data analysis (StataCorp LP, College Station, TX, USA). Chi-square analysis was conducted to compare categorical data after verification of their normal distribution using the Silk-Wilk test. A multivariable least square regression model was used to analyze the independent relationship between duration of symptoms and potential different predictors of delayed referral to our pain clinic including distance from our hospital, type of insurance, referral diagnosis, and race. A similar analysis was conducted to determine whether any of variables could predict the number of doctors' visits done prior to our evaluation. p values < 0.05 were considered statistically significant.

d referral to our pain clinic including distance from our hospital, type of insurance, referral diagnosis, and race. A similar analysis was conducted to determine whether any of variables could predict the number of doctors' visits done prior to our evaluation. p values < 0.05 were considered statistically significant. 3. Results 3.1. Population Characteristics Seventy-five patients were included in the review. Demographic data are shown in Table 1. A greater number of female than male patients were referred to the pain clinic (p < 0.01), during the study period, there were fewer Hispanic patients compared to non-Hispanic patients (p < 0.001), and only 5% of the families did not speak English and needed a translator during the visit (p < 0.001). Thirty-nine children (52%) were not able to attend regular classes because of pain and 35 (47%) were unable to continue playing sports because of pain (Table 1). Patients had been reporting pain for an average of 34 ± 55 months. The duration of symptoms was twice as long in patients who had a Health Maintenance Organization (HMO) insurance plan and Medi-Cal compared to patients who had a Preferred Provider Organization (PPO) or Kaiser insurance plan (42 ± 44 versus 21 ± 18 months, resp., p < 0.05). The longest duration of symptoms prior to the referral to our pain clinic was observed in patients with headaches (61 ± 58 months), followed by abdominal pain (39 ± 33 months) and back pain (37 ± 55 months). Patients diagnosed with dystonia and fibromyalgia had the shortest referral time (8 ± 5 months).

sp., p < 0.05). The longest duration of symptoms prior to the referral to our pain clinic was observed in patients with headaches (61 ± 58 months), followed by abdominal pain (39 ± 33 months) and back pain (37 ± 55 months). Patients diagnosed with dystonia and fibromyalgia had the shortest referral time (8 ± 5 months). The referral diagnoses are listed in Table 2. The most common reason for referral was chronic muscle-skeletal pain, followed by abdominal pain, back pain, and headaches. The list of specialties of the referring physicians is shown in Table 3. There was no statistical difference in delay to refer to this pain clinic among the different referring physicians (range 25 to 41 months). Patients consulted an average of 3 specialists in addition to their pediatrician before being referred to the pain clinic (range 1–8 specialists), and only 4 patients never saw their pediatrician (5%). The most commonly consulted specialists were as follows: orthopedic surgeons (41%), rheumatologists (27%), and neurologists and gastroenterologists (24%). Ten percent of the patients had at least one emergency room visit since the pain symptoms had started with 2 visits (range 1–4) on average prior to the referral to our clinic. There was no correlation between the number of specialists consulted and the delay to refer patients to this clinic. The multivariate regression analysis identified distance to our clinic and type of health insurance as the variables that predicted a delay in referring patients to our pain clinic (Table 4).

our clinic. There was no correlation between the number of specialists consulted and the delay to refer patients to this clinic. The multivariate regression analysis identified distance to our clinic and type of health insurance as the variables that predicted a delay in referring patients to our pain clinic (Table 4). Sixteen patients (21%) had consulted a psychologist prior to visiting our pain clinic. Four patients were seen for anxiety, 5 for depressive symptoms, and 6 either after parents divorced or after episodes of trauma (i.e., bullying, physical assault). In only one of these patients psychotherapy had been recommended to help manage chronic pain symptoms. Eleven patients (15%) underwent an operation because of pain complaints prior to the referral to the pain clinic. The list of procedures includes 2 appendectomies, 1 Nissen fundoplication and 1 cholecystectomy done for complaints of chronic abdominal pain, 2 foot surgeries for pain thought to be related to feet deformities, and 1 resection of a small leg osteochondroma for chronic leg pain. All of these patients continued to report pain after the aforementioned operations similar in intensity and quality to that experienced prior to their operation.

ronic abdominal pain, 2 foot surgeries for pain thought to be related to feet deformities, and 1 resection of a small leg osteochondroma for chronic leg pain. All of these patients continued to report pain after the aforementioned operations similar in intensity and quality to that experienced prior to their operation. 3.2. Outcomes The types of interventions recommended by our team are shown in Table 5. We confirmed the referral diagnosis in 22 patients (29%). In 34 patients (45%) we identified psychological factors as a major contributor to pain and disability: 19 patients were diagnosed with anxiety, 13 were diagnosed with depression, 1 was diagnosed with psychosis, and 1 was diagnosed with somatization disorder. We confirmed the diagnosis of CRPS in 4 of the 6 patients referred for CRPS of the lower extremity. These patients were admitted in our chronic pain inpatient rehabilitation program, while the remaining 2 patients were diagnosed with generalized anxiety disorder and referred for psychotherapy. We prescribed SSRI antidepressants to 7 patients. Six of them improved during the follow-up period. The patient who failed the recommended treatment also had a diagnosis of opioid abuse. We recommended psychotherapy in conjunction with CAM therapies in 24 patients. Three of them did not come back to the clinic, 14 followed the recommendations, and all of them improved during the follow-up period. Seven patients chose not to consult a psychologist; subsequently, none of these patients reported improvements in their pain symptoms.

y in conjunction with CAM therapies in 24 patients. Three of them did not come back to the clinic, 14 followed the recommendations, and all of them improved during the follow-up period. Seven patients chose not to consult a psychologist; subsequently, none of these patients reported improvements in their pain symptoms. Our team misdiagnosed a patient. He was initially diagnosed by surgeons with chronic functional abdominal pain. Because of negative physical findings we recommend CAM. However, after further investigations conducted by his surgeon, he was found to have an intra-abdominal neuroblastoma 6 months later.

y in conjunction with CAM therapies in 24 patients. Three of them did not come back to the clinic, 14 followed the recommendations, and all of them improved during the follow-up period. Seven patients chose not to consult a psychologist; subsequently, none of these patients reported improvements in their pain symptoms. Our team misdiagnosed a patient. He was initially diagnosed by surgeons with chronic functional abdominal pain. Because of negative physical findings we recommend CAM. However, after further investigations conducted by his surgeon, he was found to have an intra-abdominal neuroblastoma 6 months later. The length of the follow-up was 11 ± 10 months. Twelve patients (16%) decided not to come back to this pain clinic and were lost to follow-up. The average number of follow-up visits was 3 (range 1–20). Of those who returned to the clinic for follow-up, 17 patients (27%) did not follow up with the recommended treatments. Of those who adhered to the treatment plan, 43 patients (68%) were functional and had returned to their daily activities either pain-free or able to cope with their symptoms. In particular, there were a significant higher number of children who were able to go back to regular classes during the follow-up period compared to the time prior to our evaluation and recommendations (85% versus 50%; p < 0.0001). Similarly, 62% of our patients were practicing a sport activity after having followed our treatment plan compared to 20% prior to it (p < 0.0001). It should be noticed that 36% of the children who were not interested in practicing any sport at the time of the initial clinic visit started practicing some sport as part of our overall recommendations.

were practicing a sport activity after having followed our treatment plan compared to 20% prior to it (p < 0.0001). It should be noticed that 36% of the children who were not interested in practicing any sport at the time of the initial clinic visit started practicing some sport as part of our overall recommendations. The interval between the initial evaluation and clinical improvement was 4 ± 3.5 months. The pain symptoms in this group of patients had lasted on average 37 months prior to being seen in our clinic. Twenty patients (32%) reported no improvement in their symptoms. This lack of improvement has been attributed to one of the following: wrong diagnosis in 1 patient made by this team, drug abuse and dependence in 2 patients, lack of insurance for 2 patients, no compliance with the recommended treatment in 7 patients, and major depression or generalized anxiety poorly controlled by psychotherapy and medications in 7 patients. 4. Discussion This review highlights two important aspects of the care delivered to children with chronic pain: the delay to refer these patients to a pediatric pain specialist and the failure to recognize psychological disorders as an important comorbid condition in chronic pain in children and adolescents.

The interval between the initial evaluation and clinical improvement was 4 ± 3.5 months. The pain symptoms in this group of patients had lasted on average 37 months prior to being seen in our clinic. Twenty patients (32%) reported no improvement in their symptoms. This lack of improvement has been attributed to one of the following: wrong diagnosis in 1 patient made by this team, drug abuse and dependence in 2 patients, lack of insurance for 2 patients, no compliance with the recommended treatment in 7 patients, and major depression or generalized anxiety poorly controlled by psychotherapy and medications in 7 patients. 4. Discussion This review highlights two important aspects of the care delivered to children with chronic pain: the delay to refer these patients to a pediatric pain specialist and the failure to recognize psychological disorders as an important comorbid condition in chronic pain in children and adolescents. We found that the delay to refer children with chronic pain to our clinic was significantly longer than what has been reported in other studies [5, 6]. The reason for this difference is unclear. Our data indicate that the type of insurance patients carry may be a contributing factor to a delayed first contact with a pain specialist. In particular, patients with Medi-Cal or HMO type of insurance coverage waited twice as long as compared to patients who had PPO plan. The referral process in the United States health care system can be quite complex and it is triggered in the majority of cases by a general practitioner or a pediatrician. The request for a specialist consultation is then reviewed by insurance companies that may approve or deny it. Although the time for preauthorization approval is similar for Medi-Cal as well as other insurance plans (10 versus 3–5 days, resp.) [14], insurance companies can deny a specific referral. We did not have any information on the insurance companies' review decision processes. A 2003 study on private insurance patients covered by two California medical groups showed that the denial rate for physician services ranged between 11 and 24% [15].

ys, resp.) [14], insurance companies can deny a specific referral. We did not have any information on the insurance companies' review decision processes. A 2003 study on private insurance patients covered by two California medical groups showed that the denial rate for physician services ranged between 11 and 24% [15]. We have confirmed literature data [16] suggesting that distance to specialized health care can be a barrier to access resulting in a late and inaccurate diagnosis and management of children with chronic pain.

ys, resp.) [14], insurance companies can deny a specific referral. We did not have any information on the insurance companies' review decision processes. A 2003 study on private insurance patients covered by two California medical groups showed that the denial rate for physician services ranged between 11 and 24% [15]. We have confirmed literature data [16] suggesting that distance to specialized health care can be a barrier to access resulting in a late and inaccurate diagnosis and management of children with chronic pain. Lack of specialized care and low household income are other major barriers to access specialized care in general [17, 18]. The presence of limited resources available in communities to manage children with chronic pain has been shown to be a significant barrier in several countries [19–21]. There are only two pediatric pain comprehensive programs in the Los Angeles County (population 2,325,047 according to the 2015 census) [22] which can explain the delay to access our services: the average waiting time for an evaluation at our clinic is 7 weeks. Our data seems also to support reports showing that low-income household may face additional barriers in accessing pain specialists. We were not able to assess our families' income and this may limit our conclusions. However, the type of health care insurance is often a reliable indirect indicator of household income in the United States, with families in the lower income brackets usually insured by governmental or HMO plans, which have often limited options with respect to specialized health care providers. The majority of our patients (72%) were enrolled in such plans (Table 1).

n a reliable indirect indicator of household income in the United States, with families in the lower income brackets usually insured by governmental or HMO plans, which have often limited options with respect to specialized health care providers. The majority of our patients (72%) were enrolled in such plans (Table 1). Although the type of insurance and distance were significantly correlated to the delayed referral to the clinic, the model accounted only for 10% of the findings. When looking at other potential reasons for the delayed referral it is important to consider the pediatricians', specialists', and families' biases towards the search for an organic cause of pain in children. It is difficult to quantify how much inappropriate referral practices can be attributed to the belief that there are definitive tests for every specific disease and that causal determinism can be explained with certitude in medicine [23]. It is quite concerning that the quest for more diagnostic tests and specialists could lead to operations as shown in our series (15% of the patients) which did not resolve the patients pain complaints. Moreover, our data confirm previous reports that support waiting as little as five weeks before establishing an appropriate set of interventions to manage chronic pain results in declining health related quality of life [24].

shown in our series (15% of the patients) which did not resolve the patients pain complaints. Moreover, our data confirm previous reports that support waiting as little as five weeks before establishing an appropriate set of interventions to manage chronic pain results in declining health related quality of life [24]. With respect to health care utilization, our data are consistent with other reports. While a study found that higher income, more severe pain scores, and activity limitations were significant predictors of the number of doctors' visits [25], we could not identify a significant predictor among the variables we evaluated.

With respect to health care utilization, our data are consistent with other reports. While a study found that higher income, more severe pain scores, and activity limitations were significant predictors of the number of doctors' visits [25], we could not identify a significant predictor among the variables we evaluated. Even though the biopsychosocial model of chronic pain is widely accepted by chronic pain practitioners, it is often difficult for families to accept that psychological and social factors intricately interrelate to affect pain perception, coping with symptoms, engagement with providers, and, ultimately, clinical outcomes. We identified anxiety and/or depression in 45% of our patients. Longitudinal studies have clearly indicated that depression and anxiety are risk factors for developing chronic pain [26, 27]. At the same time chronic pain can trigger depression, fear, and anxiety with significant behavioral consequences, functional limitation, and perceived disability [28, 29]. The high prevalence of psychological disorders in our patients, similar to that reported in other studies [30, 31], confirms the importance of integrating a psychologist in the team of practitioners treating children with chronic pain. Some of our patients were already receiving psychological support from a psychologist before being referred to our clinic. However, this was an isolate intervention without the support of multiple integrated services further emphasizing the importance of a multidisciplinary approach to chronic pain.

with chronic pain. Some of our patients were already receiving psychological support from a psychologist before being referred to our clinic. However, this was an isolate intervention without the support of multiple integrated services further emphasizing the importance of a multidisciplinary approach to chronic pain. Not only is it important to recognize the appropriate time to refer a child with chronic pain to a pain specialist but also it is important to identify what kind of pain practice the child should be referred to. There are different types of pain clinics, those where symptom control is the priority and others that focus on improving patients' functional activities. It is rare that both approaches are used [32].

to a pain specialist but also it is important to identify what kind of pain practice the child should be referred to. There are different types of pain clinics, those where symptom control is the priority and others that focus on improving patients' functional activities. It is rare that both approaches are used [32]. Adherence with physicians' recommendations is often a major obstacle to clinical improvement. The number of patients who did not come back to the clinic or did not adhere to the treatment plan (36%) was slightly lower than what has been reported in other studies focused on chronic pain (47%) [33]. Surprisingly, Hechler et al. reported a significant improvement in patients who did not attend their clinic as recommended, opposite to our findings [33]. We failed to see any improvement in 32% of our patients. This lack of improvement could be attributed to relatively short follow-ups (11 months), the high percentage of patients with severe psychiatric disorders and drug abuse (12%), conditions that may have require prolonged interventions, and the limited compliance with the recommended treatments. Patients who complied with our recommended treatments returned to school and were able to socialize again with their peers within a short period of time (4 months) in relation to the length of their symptoms. These data are in line with those from other reports where a multidisciplinary approach by pain specialists has been utilized to address chronic pain in adolescents [34, 35].

to school and were able to socialize again with their peers within a short period of time (4 months) in relation to the length of their symptoms. These data are in line with those from other reports where a multidisciplinary approach by pain specialists has been utilized to address chronic pain in adolescents [34, 35]. There are multiple reasons why we observed a poor compliance as well as a relatively high number of patients who did not follow up with the recommended treatments. Patients may have faced problems obtaining insurance coverage for some of the CAM interventions we recommended. Opposite to data obtained in the adult population where a recent study has shown that 30–40% of adults in the United States use some form of CAM intervention every year [36], a significantly lower number of children and adolescents (12%) utilize CAM interventions [37]. The reasons for this different behavior could again be explained by parents' understanding and attitude towards chronic pain in children and their search for a medically based solution. Adolescents' limited acceptance of psychotherapy and poor adherence to treatment protocols are also a well-documented phenomenon that may have contributed to our findings [38].

uld again be explained by parents' understanding and attitude towards chronic pain in children and their search for a medically based solution. Adolescents' limited acceptance of psychotherapy and poor adherence to treatment protocols are also a well-documented phenomenon that may have contributed to our findings [38]. This study has several limitations. This was a retrospective analysis and we could not report on the recommended outcomes proposed by the PedIMPACT [39] focus group because of missing data. It should be recognized that lack of time, lack of money, and lack of manpower are obvious barriers to the routine implementation of the PedIMPACT outcomes measurements in clinical practice [40]. We were though able to report on two criteria often utilized as indexes of functional improvement (return to school and sport activities) [12] in every patient we followed up. Although the type of insurance patients carried seemed to be a barrier to access our pain clinic we were not able to determine whether the denial of service was the most important determinant for the delayed referral. Some of our patients were already receiving psychological support from a psychologist before being referred to our clinic. Our data do not allow us to explain why previous psychological interventions were ineffective. We can only speculate whether this was due to the fact that the psychological support was an isolate intervention without the support of multiple integrated services. Although we reported the level of adherence to the recommended treatments, we do not have enough data to identify the exact cause of this behavior. Also, this review does not answer the question about the exact timing for the referral to a pain clinic. It seems that physicians should take into consideration a specialized evaluation and a multidisciplinary approach in the presence of stalled progress and persistent disability.

tify the exact cause of this behavior. Also, this review does not answer the question about the exact timing for the referral to a pain clinic. It seems that physicians should take into consideration a specialized evaluation and a multidisciplinary approach in the presence of stalled progress and persistent disability. In conclusion, our data support the notion that there is a significant delay in referring children with chronic pain to pediatric pain specialists. The psychological component of pain is often missed, and this results in multiple doctor and emergency room visits, multiple laboratory and radiologic tests, and unnecessary operations with consequences on the quality of life of children. A multidisciplinary evaluation and a broad range of interventions including CAM techniques, psychotherapy, and referral to appropriate pediatric specialists could significantly expedite the return to a functional life with well managed pain symptoms. Although some of the academic pediatric pain clinics housed in large medical centers utilize a multidisciplinary or interdisciplinary approach, our data support the need for providing mental and behavioral health services as part of a primary care model as psychologists play a key role in integrated health care by helping people modify their behavior to prevent and recover from health problems. Acknowledgments The funding source for this work is Department of Anesthesia and Critical Care Medicine, Children's Hospital Los Angeles. Abbreviations CAM:Complementary and Alternative Medicine.

In conclusion, our data support the notion that there is a significant delay in referring children with chronic pain to pediatric pain specialists. The psychological component of pain is often missed, and this results in multiple doctor and emergency room visits, multiple laboratory and radiologic tests, and unnecessary operations with consequences on the quality of life of children. A multidisciplinary evaluation and a broad range of interventions including CAM techniques, psychotherapy, and referral to appropriate pediatric specialists could significantly expedite the return to a functional life with well managed pain symptoms. Although some of the academic pediatric pain clinics housed in large medical centers utilize a multidisciplinary or interdisciplinary approach, our data support the need for providing mental and behavioral health services as part of a primary care model as psychologists play a key role in integrated health care by helping people modify their behavior to prevent and recover from health problems. Acknowledgments The funding source for this work is Department of Anesthesia and Critical Care Medicine, Children's Hospital Los Angeles. Abbreviations CAM:Complementary and Alternative Medicine. Conflicts of Interest The authors have no conflicts of interest. Table 1 Demographic data of the study population (CHLA: Children's Hospital Los Angeles; HMO: Health Maintenance Organization). N % p Speaking English Yes 71 95% 0.001 No 4 5% Age 13 ± 3 range (6–19) Sex M 28 29% 0.002 F 47 71% Ethnicity: Hispanic 22 29% 0.001 Caucasians 53 71% Duration of symptoms (months) 34 ± 37 (range 3–199)

Conflicts of Interest The authors have no conflicts of interest. Table 1 Demographic data of the study population (CHLA: Children's Hospital Los Angeles; HMO: Health Maintenance Organization). N % p Speaking English Yes 71 95% 0.001 No 4 5% Age 13 ± 3 range (6–19) Sex M 28 29% 0.002 F 47 71% Ethnicity: Hispanic 22 29% 0.001 Caucasians 53 71% Duration of symptoms (months) 34 ± 37 (range 3–199) School attendance: Missing classes-homeschooled 39 32% Practicing sports: No for pain 35 47% Distance from CHLA (miles) 33 ± 39 Referral CHLA 34 45% 0.253 Community physicians 41 55% Insurance Private 21 28% 0.186 Government 31 41% HMO 23 31% # of specialists consulted prior to CHLA visit 3 ± 2 range (1–8) Table 2 List of referral diagnoses (CRPS: Chronic Regional Pain Syndrome). Diagnosis N Percentage Musculoskeletal pain 27 36% Abdominal pain 15 20% Back pain 12 16% Headache 8 11% CRPS 5 7% Chest pain 3 4% Dystonia 2 3% Sickle cell disease 2 3% Table 3 List of referring physicians by specialties. Referring physicians N % Pediatrician 34 46% Orthopedic surgery 13 17% Rheumatology 9 12% Gastroenterology 7 9% Neurology 6 8% Surgery 2 3% Hematology 2 3% Endocrinology 1 1% Pulmonologist 1 1% Table 4 Multivariable least square regression analysis exploring the association between duration of symptoms and type of insurance, distance from our hospital, referral diagnosis, and race.

13 17% Rheumatology 9 12% Gastroenterology 7 9% Neurology 6 8% Surgery 2 3% Hematology 2 3% Endocrinology 1 1% Pulmonologist 1 1% Table 4 Multivariable least square regression analysis exploring the association between duration of symptoms and type of insurance, distance from our hospital, referral diagnosis, and race. Coefficient Std error t-value 95% CI p value Distance 0.31 0.14 2.25 .0351 .589 0.028 Type of insurance −0.55 0.25 −2.2 −1.043 −.049 0.032 Diagnosis 0.02 0.04 0.43 −.066 .102 0.67 Race 0.04 0.26 −0.17 −.482 .570 0.67 Table 5 Therapeutic interventions recommended by our team with outcomes (CAM: Complementary Alternative Medicine). New recommendations N Improvement Lost to follow-up CAM 14 8 (57%) 3 (21%) CAM + psychotherapy 25 14 (56%) 3 (8%) Admission in Chronic Pain Rehabilitation program 7 5 (71%) 0 Referral to surgical services 7 4 (57%) 1 (14%) Referral to medical services 3   3 (100%) Spinal cord stimulator, pump for intrathecal delivery of morphine 2 2 (100%) 0 Local or epidural infiltration with steroids 4 2 (50%) 1 (25%) Referral to physical therapy 3 1 (33%) 0 Antidepressant 7 6 (86%) 0 New opioid-anticonvulsant 3 1 (33%) 1 (33%) Total 75 43 (57%) 12 (16%)

1. Introduction HIE in full term neonates is not unusual since it occurs among 1–3 newborns for 1000 live births [1]. The immediate and the long-term consequences can be serious. It is, indeed, responsible for a high mortality, globally, estimated at 23% of 4 million annual neonatal deaths [2, 3] and source of neurologic disability estimated at 25% according to the most recent meta-analysis [4]. These poor outcomes are virtually associated with the lack of any effective neuroprotective treatment following perinatal asphyxia, the management of which remained, until recently, supportive. The incidence of HIE is significantly higher in developing countries this may present heavy social and economic costs. In Morocco perinatal asphyxia presents a large part of national public health policy; it is among the main causes of perinatal mortality but we do not have a published national epidemiological data. This high incidence is related mainly to socioeconomic factors: the lack of pregnancy follow-up, the lack of adequate infrastructure, the geographical distance of the delivery centers and consequently the persistence of home deliveries, the absence of structures adapted to the reception of the newborn at the delivery, and the lack of newborn transport policy.

socioeconomic factors: the lack of pregnancy follow-up, the lack of adequate infrastructure, the geographical distance of the delivery centers and consequently the persistence of home deliveries, the absence of structures adapted to the reception of the newborn at the delivery, and the lack of newborn transport policy. TH as whole body or selective head cooling has become a standard therapy for moderate-severe HIE to reduce neurological damage. Most recent meta-analyses documented the efficacy of TH in term infants with moderate to severe encephalopathy [5, 6]. The safety of controlled hypothermia is now well established. No serious adverse events were reported to date by the randomized controlled studies [7]. Since the clinical benefits of TH are well established, it is considered the standard of care in many developed countries. In the United States, 50% of the neonatal intensive care units were reported to provide therapeutic hypothermia [8]. In Europe, therapeutic hypothermia is already implemented in several countries as well, due in part to participation of centers in clinical trials [9–12]. In Morocco, Mohamed VI University Hospital is the only center until recently (end of 2015), to implement TH for the treatment of asphyxiated neonates with HIE [13]. This study aims to assess the feasibility of using this protocol in this unit, to identify the problems encountered in its implementation, and to assess the outcome of these newborns.

rsity Hospital is the only center until recently (end of 2015), to implement TH for the treatment of asphyxiated neonates with HIE [13]. This study aims to assess the feasibility of using this protocol in this unit, to identify the problems encountered in its implementation, and to assess the outcome of these newborns. 2. Population and Methods 2.1. Population This study was performed in the NICU at the Mohamed VI University Hospital of Marrakech, Morocco. This unit also receives children from all the south of Morocco. The hypothermia protocol was implemented in June 2012. 38 neonates out of 72 admitted for neonatal asphyxia in NICU from July 18, 2012, to May 15, 2014, were included in the study. These children were divided into two groups: the first group included 19 infants who were treated by hypothermia (protocol group) and a control group included 19 newborns with HIE but could not receive hypothermia.

2 admitted for neonatal asphyxia in NICU from July 18, 2012, to May 15, 2014, were included in the study. These children were divided into two groups: the first group included 19 infants who were treated by hypothermia (protocol group) and a control group included 19 newborns with HIE but could not receive hypothermia. 2.2. Type of Study and Data Collection This is a prospective study including neonates born in maternity of Mohamed VI Hospital (inborn) and neonates referred from other institutes (outborn). Data were collected from patients files, analyzing demographic parameters (gestational age, birth, and sex), perinatal-neonatal features (the origin of neonates, mode of delivery, acute intrapartum events, Apgar score at 1, 5, and 10 minutes, and the need of neonatal resuscitation), severity of HIE as assessed prior to cooling (Sarnat and Sarnat criteria), evolving information (hemorrhagic, infectious, renal complications, and death), and the result of the neurological examination at 6, 12, and 18 months. For infants in the protocol group we have registered the time of cooling initiation after birth, the rectal temperature monitoring, the adverse effects, and interventions during cooling.

on (hemorrhagic, infectious, renal complications, and death), and the result of the neurological examination at 6, 12, and 18 months. For infants in the protocol group we have registered the time of cooling initiation after birth, the rectal temperature monitoring, the adverse effects, and interventions during cooling. 2.3. Implementation of Hypothermia Protocol All babies were selected and treated according to the local NICU protocol which was consistent with those recommended by French Society of Neonatology [14, 15], Briefly, newborn infants born at ≥36 weeks of gestation with birth weight > 1800 g were eligible for treatment if they had evidence of acute perinatal asphyxia and of moderate or severe HIE according to the Sarnat and Sarnat criteria. An amplitude-integrated electroencephalogram (aEEG) assessment for abnormal findings (as proposed by Al Naqeeb et al.) [16] prior to treatment initiation was highly supported. The exclusion criteria were severe intrauterine growth retardation with birth weight less than 1800 g, imperforate anus, head injury causing major intracranial hemorrhage, and severe chromosomal or congenital anomalies.

findings (as proposed by Al Naqeeb et al.) [16] prior to treatment initiation was highly supported. The exclusion criteria were severe intrauterine growth retardation with birth weight less than 1800 g, imperforate anus, head injury causing major intracranial hemorrhage, and severe chromosomal or congenital anomalies. The selective cooling of the head was the hypothermia protocol employed in this work by using special apparatus (COOL CAP OLYMPIC). It should be started before the sixth hour of age. The objective of selective hypothermia was to reach a rectal temperature between 34 and 35°C for 72 h from the beginning of the procedure. At the admission, the newborn was installed in an infant warmer that was off unless the newborn had a temperature less than 34°C. Rectal temperature was checked every 15 min until obtaining a temperature of 34°C. Hypothermia was continued for 72 h while the rectal temperature was checked every 2 hours and the skin probe was held in place continuously. After 72 h of hypothermia, the newborn was gradually warmed from 0.2 to 0.4°C per hour (6 to 12 h). The newborn was under continuous monitoring with cardiopulmonary scope, a monitoring of diuresis, glucose monitoring every 6 to 8 hours, and monitoring of blood pressure every 2 hours (every hour during the warming). A close biological screening was carried out: daily chemistry panel, complete blood count, and coagulation profile.

was under continuous monitoring with cardiopulmonary scope, a monitoring of diuresis, glucose monitoring every 6 to 8 hours, and monitoring of blood pressure every 2 hours (every hour during the warming). A close biological screening was carried out: daily chemistry panel, complete blood count, and coagulation profile. 2.4. Statistical Analysis The results were expressed as number and percentage or by the average. Statistical analysis was carried out by SPSS 17. The difference between the two groups was studied either by the nonparametric Mann–Whitney test for quantitative variables or by Chi2 or Fisher test for qualitative variables. Statistical difference was considered significant if P < 0.05. 3. Results After the implementation of the protocol in this unit, 38 neonates among 72 hospitalized for neonatal asphyxia had the indication of TH and were included in the study. The remaining 34 were excluded from the study because they did not meet the TH inclusion criteria. The 38 neonates included in the study were divided into two groups; each one included 19 newborns: one group received hypothermia and the other was a control group. There were various reasons for the 19 infants who were not cooled, mainly logistical: admission beyond 6 hours of life in 41%, the lack of place in the unit and the nonavailability of the cooling machine in 10%, technical problems in the machine in 15.7%, and cooling contraindication in 10% (extensive cephalohematoma, pulmonary arterial hypertension) and 20% because the diagnosis was not made early.

al: admission beyond 6 hours of life in 41%, the lack of place in the unit and the nonavailability of the cooling machine in 10%, technical problems in the machine in 15.7%, and cooling contraindication in 10% (extensive cephalohematoma, pulmonary arterial hypertension) and 20% because the diagnosis was not made early. For the control group, we try to maintain hypothermia as long as possible, but it was difficult to maintain the temperature below 34°C for long time, so they were treated in normal temperature. 3.1. Population Characteristics Maternal and neonatal characteristics (Tables 1 and 2) were similar in both groups. The neonatal asphyxia was suspected when fetal heart rate was abnormal (bradycardia and decelerations) in both groups. The Apgar score was lower than 5 at 5 min of life in all neonates from the control group and 17 in the protocol group. The population under study consisted of full term eutrophic newborns, mostly inborn. The HIE was classified as Sarnat 2 in half of the cases in both groups and showed severe electroencephalographic criteria (Table 2). 3.2. Protocol Assessment Parents or caregivers of the asphyxiated neonates were informed about the important benefit of TH in the HIE. However, written consent was not mandatory for treatment initiation in this institution as cooling is considered a standard of care.

3.1. Population Characteristics Maternal and neonatal characteristics (Tables 1 and 2) were similar in both groups. The neonatal asphyxia was suspected when fetal heart rate was abnormal (bradycardia and decelerations) in both groups. The Apgar score was lower than 5 at 5 min of life in all neonates from the control group and 17 in the protocol group. The population under study consisted of full term eutrophic newborns, mostly inborn. The HIE was classified as Sarnat 2 in half of the cases in both groups and showed severe electroencephalographic criteria (Table 2). 3.2. Protocol Assessment Parents or caregivers of the asphyxiated neonates were informed about the important benefit of TH in the HIE. However, written consent was not mandatory for treatment initiation in this institution as cooling is considered a standard of care. Newborns were admitted in the unit at 3.4 ± 4.6 h of life. The beginning of the procedure was at 3.3 ± 1 hour from early life. The average rectal temperature at admission was 33.4 ± 0.6°C. Among all the temperature readings, a rectal temperature below 33°C was recorded 10 times including 2 times in the same child; during this excessive hypothermia, the average time to stabilize the temperature over 33.8°C was 120 min.

± 1 hour from early life. The average rectal temperature at admission was 33.4 ± 0.6°C. Among all the temperature readings, a rectal temperature below 33°C was recorded 10 times including 2 times in the same child; during this excessive hypothermia, the average time to stabilize the temperature over 33.8°C was 120 min. 3.3. Tolerance of Selective Hypothermia Neonatal diseases associated with HIE were similar between the two groups except for pulmonary arterial hypertension that was paradoxically more common in the control group (Table 3). No statistically significant difference was observed for bleeding disorders. Bradycardia less than 90 beats per minute was noted in the protocol group which is considered a physiological effect of hypothermia. One newborn baby from the protocol group had presented a deformation of the head which is reported as side effect of head cooling therapy. 3.4. Evolution of Newborns The mean duration of hospitalization was longer in the protocol group. Seven children died in the control group versus 3 in the protocol group (Table 4). Loss of views was more important in the control group 41% versus 15% in the protocol group; the follow-up rate at the age of 18 months was 81% in the protocol group versus 58% in the control group. 56% from the protocol group had normal neurological examination at 18 months of age, 20% presented different neurologic abnormalities (Table 5), and we recorded one case of death at the age of 16 months caused by status epilepticus.

at the age of 18 months was 81% in the protocol group versus 58% in the control group. 56% from the protocol group had normal neurological examination at 18 months of age, 20% presented different neurologic abnormalities (Table 5), and we recorded one case of death at the age of 16 months caused by status epilepticus. 4. Discussion In this study, we present our experience with TH when preformed for the management of asphyxiated neonates with moderate and severe HIE, especially after the results of major randomized controlled studies that have shown a beneficial effect of controlled hypothermia on survival and long-term neurological outcome for newborns who suffered from HIE using either selective hypothermia [17, 18] or whole body cooling [10, 19–21]. Our results are encouraging with respect to the feasibility and safety of head cooling and the beneficial effect in terms of survival and neurodevelopmental outcome.

vival and long-term neurological outcome for newborns who suffered from HIE using either selective hypothermia [17, 18] or whole body cooling [10, 19–21]. Our results are encouraging with respect to the feasibility and safety of head cooling and the beneficial effect in terms of survival and neurodevelopmental outcome. Two principal methods of cooling exist: selective head cooling and the total body cooling. No superiority of either modality is supported by the existing evidence [6, 22, 23]. However, the selective cooling is associated with a large gradient of intracerebral temperatures. The temperature difference measured at 2 cm depth of the cortical surface is typically 1.3 ± 1.1°C; it reached 7.5 ± 3.5°C during the selective cooling. The distribution of the intracerebral temperature is more homogeneous in the case of total body cooling (cortical area and deep brain gradient is 1.5 ± 1.2°C at baseline and 1.1 ± 0.9°C during the body cooling) [24, 25]. In this study, we used the selective head cooling for hypothermia.

°C during the selective cooling. The distribution of the intracerebral temperature is more homogeneous in the case of total body cooling (cortical area and deep brain gradient is 1.5 ± 1.2°C at baseline and 1.1 ± 0.9°C during the body cooling) [24, 25]. In this study, we used the selective head cooling for hypothermia. The best neuroprotective effect is obtained if the treatment is started before 6 hours of life as shown in animal studies, where there is still a “therapeutic window” where secondary neuronal injury could be prevented or reduced by brain cooling [26, 27]. In fact it cooling babies before 3 h of age is even suggested by Thoresen et al. to obtain the optimal neuroprotective effect [28] and in the TOBY study, hypothermia was more effective in children treated during the first four hours after birth [20]. The average time for starting hypothermia was 3.3 ± 1 h in this study.

n fact it cooling babies before 3 h of age is even suggested by Thoresen et al. to obtain the optimal neuroprotective effect [28] and in the TOBY study, hypothermia was more effective in children treated during the first four hours after birth [20]. The average time for starting hypothermia was 3.3 ± 1 h in this study. The need to start hypothermia before 6 hours of life has been a major limitation in this study since 41% of newborns referred in this center arrived too late, which explains the reason why the majority of cooled newborn neonates included in this study were inborn. This problem has already been discussed in other studies analyzing the feasibility of hypothermia in low- and middle-income countries [29] in which the therapeutic time-window for administering beneficial cooling may be already passed, due to delayed hospital admissions, prolonged or obstructed labor, and lack of neonatal transport facilities. It therefore seems necessary to improve the diffusion of this protocol within the perinatal network by promoting rapid transfer policy to level III centers of newborns with HIE.

l cooling may be already passed, due to delayed hospital admissions, prolonged or obstructed labor, and lack of neonatal transport facilities. It therefore seems necessary to improve the diffusion of this protocol within the perinatal network by promoting rapid transfer policy to level III centers of newborns with HIE. Starting the hypothermia protocol before 6 h of life assumes a rapid assessment of the severity of HIE and therefore early recognition of anoxic-ischemic nature. Close and repetitive clinical assessment (every 1-2 h during the first 6 h) is required for these patients to determine the stage of HIE and if the HIE stage progresses from stage I to stage II therapeutic hypothermia should be started immediately. The recommendations of neonatal societies were made to this effect [14, 15], to guide practitioners in better recognition of the diagnosis of HIE and its severity. However, detailed neurological examination of an asphyxiated newborn according to criteria defined by Sarnat and Sarnat, originally during the first 48 hours of life [30], may be defective in some urgent assessment circumstances such as painful or sedated newborn. Olsen et al. suggest a neurological assessment repeated every hour in newborns with perinatal asphyxia and starting cooling if the patient develops any 3 out of 6 neurological findings [31]. The classical EEG or aEEG not only can confirm the severity of HIE and monitor subclinical seizures, but also redirect diagnostics. The combination of clinical and electrophysiological criteria seems the most efficient method to confirm encephalopathy (94% specificity, 85% positive predictive value, and 92% negative predictive value) [32]. In our unit a written protocol is used by the practitioner in call explaining clearly the diagnosis criteria, yet 20% of newborns were misdiagnosed because the clinical picture was not straightforward from early on.

encephalopathy (94% specificity, 85% positive predictive value, and 92% negative predictive value) [32]. In our unit a written protocol is used by the practitioner in call explaining clearly the diagnosis criteria, yet 20% of newborns were misdiagnosed because the clinical picture was not straightforward from early on. Careful clinical and laboratory test is essential in newborns with HIE regardless of the mode of treatment, but TH requires additional parameters to be monitored throughout treatment such as umbilical arterial and venous catheterization for blood draw and urinary catheterization for urine output measurements. Full monitoring including heart rate, respiratory rate, blood pressure, core temperature, and SaO2 is required. The core temperature is recorded by esophageal or rectal probe. The axillary temperature measurements have been reported to give variable data and, therefore, should not be preferred over core temperature measurement methods [33]. The continuous monitoring of core temperature in this study was from a rectal probe.

red. The core temperature is recorded by esophageal or rectal probe. The axillary temperature measurements have been reported to give variable data and, therefore, should not be preferred over core temperature measurement methods [33]. The continuous monitoring of core temperature in this study was from a rectal probe. The head cooling protocol requires specialized equipment; the hypothermia is obtained by using a special cap with circulating cold water placed on the head of the neonate. All the newborns from the protocol group were already hypothermic on admission with an average rectal temperature of 33.4 ± 0.6°C. We report difficulties maintaining the temperature between 34 and 35°C trying not to exceed the thermal objective of 35°C. Several episodes of hypothermia below 33°C were recorded twice in 8 newborns, reaching sometimes up to 32°C; this required sometimes the hanging of hypothermia by removing the cooling cap; it took almost two hours of the attending medical team to reach a temperature of 34°C. In the most of times the desirable temperature is maintained manually by changing the cap temperature or the environmental temperature. Achieving and maintaining hypothermia required the nurses constant attention and vigilance to ensure that the temperature remained within the prescribed range. This was the main obstacle in the TH; since it has involved a significant work charge to the medical and paramedical team whose number is insufficient. This lack of human resources could divert from taking care of other babies with a better prognosis.

to ensure that the temperature remained within the prescribed range. This was the main obstacle in the TH; since it has involved a significant work charge to the medical and paramedical team whose number is insufficient. This lack of human resources could divert from taking care of other babies with a better prognosis. The core temperature monitoring should continue for several hours after normothermia to avoid overshooting the rewarming [34]. Side effects related to perinatal asphyxia were similar for both groups; we observed a low rate in pulmonary arterial hypertension in the protocol group compared to the control group which is contradicting with the literature data [35]. In both groups, there was no difference in renal injuries; although the study of Róka et al. [36] suggested a beneficial effect of hypothermia in other organs such as the kidney and liver by reducing the cell lysis secondary to the anoxic-ischemic attack, this renal protective effect was even found in the work published by Delnard et al. [37] where they noted a significant decrease in transient renal failure rates in treated children.

al effect of hypothermia in other organs such as the kidney and liver by reducing the cell lysis secondary to the anoxic-ischemic attack, this renal protective effect was even found in the work published by Delnard et al. [37] where they noted a significant decrease in transient renal failure rates in treated children. Hospital stay was longer in the protocol group which was also found in the randomized study of Shankaran et al. [19]. This could be explained by the occurrence of neonatal infection. The worsening of neonatal infection is a major concern of cooling therapy in low- and middle-income countries related to an extensive literature on the association of increased mortality with hypothermia, and a potential worsening of sepsis with cooling, however, was reported [29, 38]. This fact could not be definitely concluded in the meta-analysis on the safety and efficacy of cooling therapy in low- and middle-income countries [39]. Almost all newborns diagnosed with HIE are started on empirical antibiotic treatment if the etiology of asphyxia is not clear. Hence, the hypothermia is known to cause some degree of immunosuppression with decreased leukocyte number and impaired functions [40, 41], as also reported in the latest Cochrane meta-analysis [7].

st all newborns diagnosed with HIE are started on empirical antibiotic treatment if the etiology of asphyxia is not clear. Hence, the hypothermia is known to cause some degree of immunosuppression with decreased leukocyte number and impaired functions [40, 41], as also reported in the latest Cochrane meta-analysis [7]. We had a higher rate of newborn follow-up in the cooling group comparing to a control group, reaching 80%. In the noncooling group, we had a high rate of loss of views; this may be explained by a possible improvement in their status. A significant decrease in death rates and neurological morbidity at the age of 18 months in children who have moderate or severe HIE was found in newborns of protocol group. The limit of the study is the small number of included neonates. However, data presented here are derived from a single center and, therefore, only a limited number of neonates could have been assessed in a relative short time, particularly with respect to long-term outcome. On the other hand, the analysis of the results permits the detection of important clinical parameters which could allow further improvement in the clinical implementation of this novel therapeutic approach. The further studies with a larger population as well as training medical team are necessary to confirm these results.

he other hand, the analysis of the results permits the detection of important clinical parameters which could allow further improvement in the clinical implementation of this novel therapeutic approach. The further studies with a larger population as well as training medical team are necessary to confirm these results. 5. Conclusion Implementation of TH is facing a lot of problems in Morocco. The generalization of this practice needs to be guided by standardized protocols. Local protocols should be developed based on the existing international experience adapted to local context. Developing training programs, improving infrastructure including neonatal transport, and affording human resources are mandatory to guarantee the success of hypothermia in Morocco. Conflicts of Interest The authors declare that there are no conflicts of interest regarding the publication of this paper. Table 1 Maternal characteristics. Maternal characteristics Protocol group (n = 19) Control group (n = 19) P value Maternal age (average) 26,3 27,3 NS Parity (average) 1,3 2,1 NS Pregnancy complications, n (%)     NS Gestational diabetes 0 1 (5) Gestational hypertension 1 (5,2) 0 Delivery complications NS Abnormal fetal heart rate 13 (68,4) 11 (57) Meconium or stained amniotic fluid 9 (47,4) 14 (73) Peripartum pathologies NS Prolapsed cord 2 (10,6) 2 (10,6) Other cordonal pathologies 0 1 (5) Retention of the after-coming head 2 (10,6) 0 Shoulder dystocia 1 (5) 1 (5) NS: not significant. Table 2 Neonatal characteristics.

Delivery complications NS Abnormal fetal heart rate 13 (68,4) 11 (57) Meconium or stained amniotic fluid 9 (47,4) 14 (73) Peripartum pathologies NS Prolapsed cord 2 (10,6) 2 (10,6) Other cordonal pathologies 0 1 (5) Retention of the after-coming head 2 (10,6) 0 Shoulder dystocia 1 (5) 1 (5) NS: not significant. Table 2 Neonatal characteristics. Neonatal characteristics Protocol group (n = 19) Control group (n = 19) P value Female gender, n (%) 6 (31,6) 7 (36) NS Birth weight (g) (average) 3336 3300 NS Apgar score ≤ 5 at 5 min, n (%) 17 (89,5) 19 (100) NS Apgar score ≤ 5 at 10 min, n (%) 12 (63,1) 16 (84) NS Intubation in the delivery room     NS Intubation only 2 (10,6) 4 (21) Intubation and chest compression 3 (15,8) 2 (10,5) Features of aEEG, n (%)     NS Type 2 (i) With seizure 9 (47,3) 6 (32) (ii) Without seizure 1 (5,3) 2 (10,5) Type 3 (i) With seizure 7 (36,8) 4 (21) (ii) Without seizure 0 2 (10,5) Electric seizure 2 (10,6%) 3 (15,8) aEEG: amplitude-integrated electroencephalography (according to Al Naqeeb classification). Table 3 Complications during hospitalization. Protocol group (n = 19)  % Control group (n = 19)  % P value Bradycardia <90/min 17 (90) 2 (10) S Thrombocytopenia 5 (26,3) 7 (37) NS Hepatic cytolysis 7 (36,8) 11 (57) NS Acidosis 4 (21) 7 (37) NS Hyperkalemia 7 (36,8) 3 (15) NS Renal failure 8 (42) 13 (68) NS PAH 0 6 (31) S Head edema 1 0 — Nosocomial infection 5 (26,3) 4 (21) NS S: significant. PAH: pulmonary arterial hypertension. Table 4 Evolution during hospitalization.

Protocol group (n = 19)  % Control group (n = 19)  % P value Bradycardia <90/min 17 (90) 2 (10) S Thrombocytopenia 5 (26,3) 7 (37) NS Hepatic cytolysis 7 (36,8) 11 (57) NS Acidosis 4 (21) 7 (37) NS Hyperkalemia 7 (36,8) 3 (15) NS Renal failure 8 (42) 13 (68) NS PAH 0 6 (31) S Head edema 1 0 — Nosocomial infection 5 (26,3) 4 (21) NS S: significant. PAH: pulmonary arterial hypertension. Table 4 Evolution during hospitalization. Protocol group n (%) Control group n (%) Average of hospital stay 12,4 6,7 Death 3 (15) 7 (37) Normal neurological exam at discharge 8 (42) 5(26) Table 5 Data from the neurological assessment at the age of 18 months. Protocol group n (%) Control group n (%) Normal neurological exam 9 (56) 5 (41) Psychomotor delay 3 (15,7) 2 (10,5) Epilepsy 1 (5,2) 1 (5,2) Neurosensorial disorders 1 (5,2) 3 (25)

1. Introduction Choking on small parts is one of the leading, yet preventable causes of injury and death in infants and small children [1]. Children's products contain small parts or release small parts through detachment or breakage during normal play, which pose choking hazards and may be undetectable to parents and caregivers. Small parts have caused choking-related deaths in over 90 children between 2001 and 2012, as cited in the 28th Annual Survey of Toy Safety [2]. Limiting the prevalence of choking and related occurrences is an important public health goal. The current federal rule requiring the testing of small parts protects children under three years of age, which is the age group most likely to mouth objects [3]. The risk of choking can potentially be decreased by requiring that products for children under the age of three meet a new standard for size and breakability. The small parts test regulation is required by the Federal Hazardous Substances Act under the Code of Federal Regulation (CFR), premised to prevent death and injuries to children under the age of three from choking on, inhaling, or swallowing small objects they may put in their mouths [4]. The current testing method to prevent choking hazards employs an apparatus dubbed the small parts test fixture (SPTF), created by the US Consumer Product Safety Commission (CPSC). The SPTF measures whether a toy is too large to enter a child's esophagus and, thus, can be played with safely.

may put in their mouths [4]. The current testing method to prevent choking hazards employs an apparatus dubbed the small parts test fixture (SPTF), created by the US Consumer Product Safety Commission (CPSC). The SPTF measures whether a toy is too large to enter a child's esophagus and, thus, can be played with safely. If the object fits inside the cavity of the SPTF, it is too small and can potentially be lodged inside the throat of the child and cause choking. The SPTF measures 1 inch to 2.25 inches in height, slanted on a diagonal plane, and 1.25 inches in diameter at its current size. According to the American Academy of Pediatrics, this range would make choking on a small part highly improbable as it approximates the size of a fully expanded pharynx of an infant [5]. However, current evidence and statistics confirm that some products larger than the size of the SPTF have caused choking.

rrent size. According to the American Academy of Pediatrics, this range would make choking on a small part highly improbable as it approximates the size of a fully expanded pharynx of an infant [5]. However, current evidence and statistics confirm that some products larger than the size of the SPTF have caused choking. One of the first studies evaluating the current size of the SPTF and the injuries and deaths of children due to the exposure of small parts was released more than two decades ago. Meyers and Bond (1989) evaluated a compilation of 195 choking incidents reported by the CPSC between July 1973 and May 1983, roughly a ten-year span, and found that 57% of the children's products whose size was available had diameters larger than the 1.25-inch standard [6]. Even with the statistics provided by Meyers and Bond, no changes have been proposed to the testing method since that time. It has been documented in subsequent studies that objects larger than the current size of the SPTF have caused choking following this article as manifested by the numerous recalls and injuries listed on the CPSC website, individual parent reports/complaints, and other related children's safety research projects [7, 8]. Other case examples of the hidden dangers posed by children's products have been reviewed in Children and Injuries by Frost, Ed.D. [9].

ticle as manifested by the numerous recalls and injuries listed on the CPSC website, individual parent reports/complaints, and other related children's safety research projects [7, 8]. Other case examples of the hidden dangers posed by children's products have been reviewed in Children and Injuries by Frost, Ed.D. [9]. The objective of the present study is to provide the most up-to-date information on choking statistics and examples of children's products that are larger than the current fixture size that yielded a choking or related hazard. By providing these relevant statistics, it epitomizes that a larger test fixture could be the crucial next step in the prevention of choking occurrences. In 2006, Playskool, Inc., voluntarily recalled their Team Talkin' Tool Bench after receiving reports of two boys under two years old who suffocated on the toy set's oversized nails, measuring 0.75 inches longer than the current SPTF size [10]. In 2007, an eight-year-old boy died after choking on a toy dart that measured 2.5 inches long and 0.75 inches wide, again, longer than the current SPTF size [11]. An 11-month-old boy also choked on a toy nail, part of the Little Tikes' Toy Tool Set, after it became forcefully lodged in his throat and prompted the original recall in 2009 [12]. Two additional choking incidents were reported even after the original recall of this product [13]. These are just a few examples of choking occurrences that resulted in injury and even death from children's products that measured larger in size than the current SPTF.

roat and prompted the original recall in 2009 [12]. Two additional choking incidents were reported even after the original recall of this product [13]. These are just a few examples of choking occurrences that resulted in injury and even death from children's products that measured larger in size than the current SPTF. The current small part size may also not be adequately preventing children from choking due to the fact that, over the last five decades, children have grown taller and bigger in size. For example, according to Professor Mitch Blair, the average child's height has increased by 1 cm to 3 cm throughout every decade over the last 50 years. Thus, children on average have grown between 5 cm and 15 cm taller over the last five decades [14]. This means that one can assume that the child's organs, such as their esophagus, have grown exponentially with the height as well. Since the recommendation of the Academy of Pediatrics was made in 1987 regarding the size of an infant/toddler's esophagus to create the current SPTF size, there has been evidence that children have been growing taller since that time (roughly 30 years since the article was published) (see [6]). The current small part size has not adapted to the increase in size of children; thus a new and improved size (a larger one) may be suitable in preventing further choking incidents.

ere has been evidence that children have been growing taller since that time (roughly 30 years since the article was published) (see [6]). The current small part size has not adapted to the increase in size of children; thus a new and improved size (a larger one) may be suitable in preventing further choking incidents. The present study, assessed by Kids In Danger (KID), sought to gather up-to-date information about the sizes of the products involved in the choking-related recalls in the past twelve years and whether the small parts test fixture is adequate in preventing choking and other related hazards.

ere has been evidence that children have been growing taller since that time (roughly 30 years since the article was published) (see [6]). The current small part size has not adapted to the increase in size of children; thus a new and improved size (a larger one) may be suitable in preventing further choking incidents. The present study, assessed by Kids In Danger (KID), sought to gather up-to-date information about the sizes of the products involved in the choking-related recalls in the past twelve years and whether the small parts test fixture is adequate in preventing choking and other related hazards. 2. Method KID reviewed 303 recalled products publically listed by the CPSC that presented choking hazards to infants and young children due to small parts between January 1, 2003, and December 31, 2014 (This study is a thorough review of official recalls reported by the CPSC between January 1st, 2003, to December 31st, 2014. Individual incidents and complaints by parents are not included in this study.). Manufacturers tested these products and concluded that they were large enough and durable enough (i.e., would not break) for safe play prior to placement on the market. The products were eventually recalled due to breakage or even the original, intact product causing choking and related incidents. KID obtained the total number of recalled products and examined them by product categories and the kind of choking hazard caused by each product. Most importantly, KID focused on the sizes of the product or piece of the product that posed hazards and/or caused choking in proportion to the current size of the SPTF. Lastly, KID also reviewed the total number of incidents caused by the recalled products. Choking “incidents” include any interaction with the mouth, including mouthing, gagging, choking, ingestion, aspiration, or coughing of the product by the child. KID noted the various types of incidents that were caused by the products involved within the dataset, which are listed below. The authors of this article organized the data stated above in a spreadsheet (i.e., type of product and what kind of choking hazard the specific product caused), thus allowing ease of access to product details.

ious types of incidents that were caused by the products involved within the dataset, which are listed below. The authors of this article organized the data stated above in a spreadsheet (i.e., type of product and what kind of choking hazard the specific product caused), thus allowing ease of access to product details. 3. Results Analyses show that 48.8% of the products within the dataset fall under the toys category (This study did not include outdoor toys such as bikes, battery-powered life-size electric vehicles, and trampolines.), 27.7% of the products fall under the clothing category (This study did not include glasses frames or sunglasses.), 18.9% of the products fall under the nursery category (This study did not include strollers, baby bouncers, exersaucers, furniture (cribs, sofas, and bean bag chairs), blankets, or high chairs.), and the final 4.6% of the products fall under the books and art supplies category. This categorizes the kinds of products that were listed on the CPSC website as having the potential to cause choking or have caused choking and related incidents. With the current size of the SPTF, a notable quantity of the products within the dataset did not fit inside the cavity of the tube. Research shows that 17.1% of the products in the dataset were larger than the current fixture size and still posed a choking threat or caused choking and related occurrences in some instances. The additional 82.8% of the products in the dataset were smaller than the SPTF.

aset did not fit inside the cavity of the tube. Research shows that 17.1% of the products in the dataset were larger than the current fixture size and still posed a choking threat or caused choking and related occurrences in some instances. The additional 82.8% of the products in the dataset were smaller than the SPTF. As described in Section 2, the next matter that KID analyzed was the number of incidents that occurred due to the reported products causing choking and related incidents. Analyses show that a total of 211 incidents were caused by these recalled products that occurred between January 1st, 2003, through December 31st, 2014. These incidents were caused by 63 of the 303 products, or 21% of the total products. Fifty-six of these 63 products were marketed for children under the age of three. Table 1 categorizes the incidents by type. Four of the choking incidents above resulted in death; three of the four deaths were caused by an object that was larger than the dimensions of the SPTF. Two of the deaths resulted in two children who choked on toy nails, as part of Playskool, Inc.'s Team Talkin' Toy Tool Bench, which measured 3 inches long. These toy nails were 0.75 inches longer than the current SPTF length.

e of the four deaths were caused by an object that was larger than the dimensions of the SPTF. Two of the deaths resulted in two children who choked on toy nails, as part of Playskool, Inc.'s Team Talkin' Toy Tool Bench, which measured 3 inches long. These toy nails were 0.75 inches longer than the current SPTF length. 4. Discussion Statistics from the present study indicate that despite meeting the testing requirements via third-party testing prior to distribution, certain children's products still posed choking hazards and caused choking and even death in some instances. More research on the SPTF is necessary in order to determine whether the fixture should be enlarged. For instance, data from health care providers in the field and even complaints from parents/caregivers might indicate an even larger proportion of choking incidents than reported by the CPSC alone. With the data shown above, however, the current size of the SPTF may be one cause of not adequately preventing children from choking on children's products. Although there may be other factors involved, the SPTF was created with the sole purpose of preventing choking in children. Some products within the dataset that were larger than the SPTF have posed choking hazards and even caused choking and related injuries including death between 2003 and 2014, showing that its current size may not effectively prevent choking.

the SPTF was created with the sole purpose of preventing choking in children. Some products within the dataset that were larger than the SPTF have posed choking hazards and even caused choking and related injuries including death between 2003 and 2014, showing that its current size may not effectively prevent choking. Overall, the current SPTF size has thus far not completely ruled out all choking and related hazards within children's products as exemplified through the numerous hazards, injuries, and deaths presented in this study. A larger test fixture would be the first step in generating the capacity to reduce the number of choking incidents in infants and toddlers. This is because it would abolish manufacturers from including parts even larger than the current size of a “small part” within their products. By doing so, it takes into account that small children can still potentially choke on small parts, but, by increasing the size of the SPTF, it can reduce the risk of choking and related hazards. A number of products within the dataset contained parts that were larger than the SPTF and still posed a choking threat. Therefore, a larger SPTF could theoretically rule out more potential risks and possibilities of choking. Supplementary Material The original dataset includes 303 recalled products listed on the CPSC website which lists the date of original recall, the manufacturer, the number of reported injuries, and other pertinent information about each recalled children's product. Additional Points

Overall, the current SPTF size has thus far not completely ruled out all choking and related hazards within children's products as exemplified through the numerous hazards, injuries, and deaths presented in this study. A larger test fixture would be the first step in generating the capacity to reduce the number of choking incidents in infants and toddlers. This is because it would abolish manufacturers from including parts even larger than the current size of a “small part” within their products. By doing so, it takes into account that small children can still potentially choke on small parts, but, by increasing the size of the SPTF, it can reduce the risk of choking and related hazards. A number of products within the dataset contained parts that were larger than the SPTF and still posed a choking threat. Therefore, a larger SPTF could theoretically rule out more potential risks and possibilities of choking. Supplementary Material The original dataset includes 303 recalled products listed on the CPSC website which lists the date of original recall, the manufacturer, the number of reported injuries, and other pertinent information about each recalled children's product. Additional Points Statement of Knowledge. The SPTF has been implemented by the CPSC to measure for small parts and identify their presence on product labeling for the prevention of choking. Meyers and Bond (1989) discussed the inability of the SPTF to accurately prevent choking because objects larger than the SPTF have caused choking in children. No changes have been made by the CPSC to the small parts test fixture since the 1989 publication. Statement of Addition to the Public. This study provides up to date statistics showing that the current SPTF size does not adequately prevent choking. Three of the four deaths in this study were caused by an object that was larger than the current SPTF size. This study shows that enlarging the SPTF can help prevent choking in the future by eliminating slightly larger parts from children's products.

owing that the current SPTF size does not adequately prevent choking. Three of the four deaths in this study were caused by an object that was larger than the current SPTF size. This study shows that enlarging the SPTF can help prevent choking in the future by eliminating slightly larger parts from children's products. Conflicts of Interest The authors declare that they have no conflicts of interest. Table 1 Type and occurrence of choking incidents. Percent (%) Choking 46.91 Mouthing 27.96 Gagging 21.33 Ingestion 2.37 Coughing 0.94 Aspiration 0.50

1. Background Acute poisoning in the pediatric age group is an important cause of preventable mortality and morbidity. The circumstances of poisoning and related risk factors vary widely across different geographic regions globally due to variable accessibility and availability and varied environmental factors. Mortality due to acute unintentional poisoning among children under 4 years of age varies from 0.3 to 7 per 100,000 people in different countries of the world [1]. Accidental poisoning in the pediatric age group is rising day by day in developing countries. Easy availability of household chemicals, medicines, and pesticides predisposes to accidental poisoning. These unintentional self-poisonings have become major health hazards causing severe toxic effects in children with long-term morbidity. Acute poisoning is an important clinical problem in Sri Lanka and it has a significant economic impact on the health service of the country. The data on financial costs of managing children with unintentional poisoning in rural Sri Lanka, however, are currently unavailable. The significance of this problem lies mainly in the factors predisposing to acute unintentional poisoning. These factors are diverse and include both situational factors (geographic location, social and economic barriers, and culture) and person related factors (personality, lifestyle, parenting style, and education level of parents).

of this problem lies mainly in the factors predisposing to acute unintentional poisoning. These factors are diverse and include both situational factors (geographic location, social and economic barriers, and culture) and person related factors (personality, lifestyle, parenting style, and education level of parents). An Asian study [2] showed maternal employment and previous history of poisoning as significant risk factors for unintentional poisoning among children, with unavailability of poisons being a protective factor. The same study reported that poor maternal education, inadequate supervision of children, substance abuse, and mental illness in family members are risk factors that increase the incidence of poisoning in children. The study identified safe storage and health education on prevention of substance abuse as effective interventions for reducing the unintentional poisoning risk among children in the studied community.

nce abuse, and mental illness in family members are risk factors that increase the incidence of poisoning in children. The study identified safe storage and health education on prevention of substance abuse as effective interventions for reducing the unintentional poisoning risk among children in the studied community. The literature on risk factors for acute poisoning among children is sparse in the South Asian region. Ahmed et al. (2011) [3] comprehensively evaluated multiple risk factors of poisoning among children in Pakistan with the focus of their study being on population attributable risk factors for acute poisoning. Agarwal et al. (2016) proposed the need for raising awareness among the public in poisoning prevention based on their epidemiological study in India [4]. Childhood poisoning statistics in Sri Lanka are limited. Though there are a number of studies available for characterizing poisoning patterns among adult populations in Sri Lanka, comprehensive risk factor studies for the pediatric age group are unavailable. A 15-year prospective study (1985–2000) of poisoning patterns among children at Lady Ridgeway Hospital for Children described 2100 children with unintentional poisoning [5]. The study was based predominantly on an urban population. Poisoning trends have changed over the years in the pediatric age group with varied socioeconomic, political, and cultural developments and there have been no studies published in Sri Lanka for more than a decade. The risk factors operating on poisoning patterns are also likely to have changed along with changing trends in childhood poisoning. Availability of such information would indefinitely benefit poison management centers in Sri Lanka in planning preventive interventions, educating the community, and allocating scarce resources more efficiently. In this background, the current study aimed to identify child, caregiver, environment, and poisoning substance related risk factors for unintentional poisoning via a comprehensive analysis of multiple risk factors.

ning preventive interventions, educating the community, and allocating scarce resources more efficiently. In this background, the current study aimed to identify child, caregiver, environment, and poisoning substance related risk factors for unintentional poisoning via a comprehensive analysis of multiple risk factors. 2. Methods 2.1. Study Setting This hospital based case-control study was conducted over a period of two years (from February 2012 to January 2014) at Anuradhapura Teaching Hospital. Anuradhapura Teaching Hospital is the largest hospital in the north-central province in which the majority of people belong to a rural community. 2.2. Participants This study involved all inpatient children who presented with acute unintentional poisoning and who were between 1 and 5 years of age. Children were recruited as “cases” after their poisoning events were confirmed by caregivers following the initial evaluation at the hospital's emergency department and subsequently at general pediatric wards. Children with doubtful poisoning and with no clear etiology were excluded from the study. Children who had intentional poisoning were also excluded. Children with past history of poisoning were excluded from both groups. Children with food poisoning, snake envenomation, allergic reactions, and adverse drug reactions which can be considered in the purview of toxicology were also excluded from the study.

udy. Children who had intentional poisoning were also excluded. Children with past history of poisoning were excluded from both groups. Children with food poisoning, snake envenomation, allergic reactions, and adverse drug reactions which can be considered in the purview of toxicology were also excluded from the study. Three hundred children were recruited as “cases” over the two-year study period. The “control” group was selected from the same pediatric wards over the same study period. Children, who presented with acute medical illnesses and without any history of chronic medical illnesses or accidental or deliberate poisoning, were interviewed as controls to compare the prevalence of risk factors among the two groups. The acute medical illnesses considered included viral fever, acute upper respiratory tract infection, and urticaria. All other acute conditions including nonspecific symptoms without a definitive diagnosis were excluded. All “cases” were matched for age and gender on individual patient basis. Both groups comprised three hundred children adding to a total of six hundred children. Minimum sample size required for this matched case-control study was 248 pairs (α: 0.05, β: 0.1, PA: 0.8, and PD: 0.1).

t a definitive diagnosis were excluded. All “cases” were matched for age and gender on individual patient basis. Both groups comprised three hundred children adding to a total of six hundred children. Minimum sample size required for this matched case-control study was 248 pairs (α: 0.05, β: 0.1, PA: 0.8, and PD: 0.1). 2.3. Data Collection Data were collected from the caregivers of children recruited to the “case” and “control” groups. Mothers were interviewed in most encounters and fathers or other caregivers (grandparents/other related caregivers) were interviewed only when mothers were not available to participate in the study. Data collection from all six hundred caregivers was done by the principal investigator himself to minimize interviewer bias. Interviews with the caregivers were conducted on the same day of admission to minimize possible recall bias. Data were collected using a pretested structured questionnaire which comprised questions to identify demographic data, type and circumstances of poisoning, and risk factors for acute poisoning (Appendix) and qualitative evaluation via focused group discussions. The study instruments were pretested by administration of the questionnaire to fifty caregivers (twenty-five in each group) in the same study setting over a two-month period prior to commencement of the study. Risk factors were categorized under four domains, environmental, psychosocial, and family related factors and personal characteristics, and twenty-three risk factors were proposed. Extensive literature survey was done to identify previously reported risk factors in other geographic regions. The investigators proposed and designed a risk factor questionnaire themselves and the questionnaire was administered following careful pretesting and expert review.

nd twenty-three risk factors were proposed. Extensive literature survey was done to identify previously reported risk factors in other geographic regions. The investigators proposed and designed a risk factor questionnaire themselves and the questionnaire was administered following careful pretesting and expert review. 2.4. Outcome and Exposures The outcome of interest was medically attended acute poisoning from medicines, household chemical agents, garden plants, and pesticides present at the child's home or home garden resulting in hospital admission. Suspicious and doubtful poisoning events were excluded. Twenty-three proposed risk factors were considered in terms of “exposures.” The proposed 23 risk factors were broadly categorized to environmental, psychosocial, and family related factors and personal characteristics. Table 1 illustrates the proposed risk factors. Each risk factor was defined prior to inclusion of those risk factors in the questionnaire.

2.4. Outcome and Exposures The outcome of interest was medically attended acute poisoning from medicines, household chemical agents, garden plants, and pesticides present at the child's home or home garden resulting in hospital admission. Suspicious and doubtful poisoning events were excluded. Twenty-three proposed risk factors were considered in terms of “exposures.” The proposed 23 risk factors were broadly categorized to environmental, psychosocial, and family related factors and personal characteristics. Table 1 illustrates the proposed risk factors. Each risk factor was defined prior to inclusion of those risk factors in the questionnaire. The presence of childhood personality abnormalities was identified by parents' subjective judgment of the child's personality as being one or more of the following: shy, timid, aggressive, avoidant, antisocial, overdependent, or any psychiatric illness related personality disorder. Child behavioral abnormalities were defined for the study as one of abnormal behaviors including nightmares, night terrors, nail biting, stammering, abnormal eating or sleeping habits, hyperactivity, impulsivity, and attention seeking behavior. The presence of unsafe storage (medicines, household chemicals, and pesticides) was identified when those compounds were not stored in a lockable container or an inaccessible location to the child. Caregivers' judgment was considered in determining the presence of economic and marital problems, sibling related problems, inadequate house space, and lack of family and social support. Harmful alcohol use was considered as an adverse exposure. A young mother was defined as a person who was nineteen years old or less at the time of assessment. The presence of a psychological illness in a parent was defined for the study as being diagnosed with a psychiatric illness or having sufficient clinical criteria for a diagnosis of a psychiatric illness based on DSM-V (Diagnostic and Statistical Manual of Mental Disorders). Inadequate supervision was defined for the study as the lack of consistent presence of a principal caregiver (either mother or father) during the child's stay at the home premises. It was considered as an environmental risk factor based on cultural circumstances given that the home environment and family structure were determining the status of supervision. Incorrect parenting style was defined for the study as one of the nonauthoritative parenting styles including neglectful, permissive, and authoritarian parenting styles.

ronmental risk factor based on cultural circumstances given that the home environment and family structure were determining the status of supervision. Incorrect parenting style was defined for the study as one of the nonauthoritative parenting styles including neglectful, permissive, and authoritarian parenting styles. Developmental delay was defined as a delay of more than six months in achieving milestones in one of four domains of child development up to five years as identified in the Child Health Development Record (CHDR) published by the World Health Organization (WHO). The four domains included gross motor, fine motor and vision, speech, language and hearing, and social and behavioral development. Investigators understood that CHDR provides only crude assessments; however, it was more convenient to use and familiar to the parents of children recruited in the study. Sophisticated developmental assessment tools were not feasible with the study due to limitation of resources and time factor and lack of cooperation from participants.

rstood that CHDR provides only crude assessments; however, it was more convenient to use and familiar to the parents of children recruited in the study. Sophisticated developmental assessment tools were not feasible with the study due to limitation of resources and time factor and lack of cooperation from participants. In order to perform an in-depth analysis of the proposed predisposing risk factors of acute accidental poisoning and to ensure that all the proposed risk factors meet the study definitions, a qualitative study was conducted by the principal investigator himself, recruiting parents of all participant children concurrent with the administration of the pretested, multistructured questionnaire. Each risk factor was evaluated in qualitative terms and in relation to study definitions before the response was recorded in the data collection questionnaire. Data in the qualitative study was recorded as field notes and had emphasis on study definitions. Data collection was done prospectively over two years via focused group discussions.

evaluated in qualitative terms and in relation to study definitions before the response was recorded in the data collection questionnaire. Data in the qualitative study was recorded as field notes and had emphasis on study definitions. Data collection was done prospectively over two years via focused group discussions. 2.5. Statistical Methods and Analysis All data were analyzed using SPSS version 19.0. All twenty-three proposed risk factors were used to create a binary logistic regression model adjusted for age and gender. In this model, each factor was initially evaluated using univariate analysis for significance levels. Controls were kept as the dependent variables and all proposed risk factors were submitted as categorical covariates. Stepwise backward conditional method was applied in the model. Probability for stepwise method was set as entry 0.5 and removal 0.20. Odds ratios were calculated for each risk factor along with 95% confidence intervals (CI). Independent risk factors were identified in the same model by multivariate analysis. Odds ratios were calculated for each of the risk factors along with 95% confidence intervals for each ratio similar to univariate analysis.

.20. Odds ratios were calculated for each risk factor along with 95% confidence intervals (CI). Independent risk factors were identified in the same model by multivariate analysis. Odds ratios were calculated for each of the risk factors along with 95% confidence intervals for each ratio similar to univariate analysis. 3. Results Six hundred participants comprising 300 children in each of the “case” and “control” groups were available for analysis. All poisoning events occurred by ingestion of a poison. Household poisoning substances were the commonest type of poison (n = 102, 34%), followed by medicines (n = 88, 29.3%), plant poisons (n = 50, 16.7%), pesticides (n = 22, 7.3%), and miscellaneous poisons (n = 38, 12.7%). The commonest poison was kerosene oil (n = 70, 23.3%). Paracetamol (n = 22, 7.3%), Ricinus communis (n = 21, 7%), organophosphate pesticides (n = 11, 3.7%), and Abrus precatorius (n = 10, 3.3%) were other commoner poisons. 234 (78%) poisoning events occurred in either home or home garden. 144 children (48%) were transferred from a regional hospital to Anuradhapura Teaching Hospital. All children were between 1 and 5 years of age and had similar sex distributions. The age and sex distributions of the two groups are illustrated in Tables 2 and 3.

3. Results Six hundred participants comprising 300 children in each of the “case” and “control” groups were available for analysis. All poisoning events occurred by ingestion of a poison. Household poisoning substances were the commonest type of poison (n = 102, 34%), followed by medicines (n = 88, 29.3%), plant poisons (n = 50, 16.7%), pesticides (n = 22, 7.3%), and miscellaneous poisons (n = 38, 12.7%). The commonest poison was kerosene oil (n = 70, 23.3%). Paracetamol (n = 22, 7.3%), Ricinus communis (n = 21, 7%), organophosphate pesticides (n = 11, 3.7%), and Abrus precatorius (n = 10, 3.3%) were other commoner poisons. 234 (78%) poisoning events occurred in either home or home garden. 144 children (48%) were transferred from a regional hospital to Anuradhapura Teaching Hospital. All children were between 1 and 5 years of age and had similar sex distributions. The age and sex distributions of the two groups are illustrated in Tables 2 and 3. 3.1. Univariate Analysis of Individual Risk Factors Nine out of twenty-three risk factors showed a significant effect at p < 0.001 (CI: 99%) in univariate analysis. Those factors included unsafe storage of medicines, unsafe storage of household chemicals, inadequate supervision of the child, mother employment during the daytime, nonauthoritative parenting styles, primary level education in the mother, psychological illnesses in parents, poisonous plants in the home garden, and parental concern of lack of family support. Four risk factors showed a significant effect at 0.001 < p < 0.05 (CI: 95%). They included inadequate house space, unsafe use/storage of agrochemicals, developmental delay in the child, and young maternal age (<19 years old). Eight risk factors including history of personality and behavioral disorders in the child, marital problems among parents, mother employment outside the country, parental concern regarding sibling disharmony, lack of schooling/educational opportunities for the child, single parent status, and lack of social support showed no significant association with confidence intervals of respective odds ratios lying on both sides of 1.00. The risk factors observed in univariate analysis are illustrated in Table 4.

sibling disharmony, lack of schooling/educational opportunities for the child, single parent status, and lack of social support showed no significant association with confidence intervals of respective odds ratios lying on both sides of 1.00. The risk factors observed in univariate analysis are illustrated in Table 4. 3.2. Multivariate Analysis in the Binary Regression Model Table 5 illustrates the independent risk factors observed in multivariate analysis. In step 8 of the backward conditional approach, eight risk factors were left in the model showing significance of at least p < 0.05. The risk factors which showed significance of p < 0.05 but >0.001 (confidence level = 95%) included unsafe storage of household chemicals (0.001), nonauthoritative parenting style (0.005), developmental delay in the child (0.012), primary level education in the mother (0.039), and poisonous plants in the home garden (0.001). Three proposed risk factors showed significance of p < 0.001 (CI: 99%) and they were inadequate supervision of the child, mother being employed during daytime, and parental concern of lack of family support to look after the child. The lower and upper limits indicating 95% confidence intervals in respective odds ratios in all nine risk factors lied above 1.00. 4. Discussion The current study evaluated a broad range of potential risk factors for unintentional poisoning in children aged 1–5 years. The risk factors were broadly categorized under environmental, psychosocial, and family related factors and personal characteristics.

In step 8 of the backward conditional approach, eight risk factors were left in the model showing significance of at least p < 0.05. The risk factors which showed significance of p < 0.05 but >0.001 (confidence level = 95%) included unsafe storage of household chemicals (0.001), nonauthoritative parenting style (0.005), developmental delay in the child (0.012), primary level education in the mother (0.039), and poisonous plants in the home garden (0.001). Three proposed risk factors showed significance of p < 0.001 (CI: 99%) and they were inadequate supervision of the child, mother being employed during daytime, and parental concern of lack of family support to look after the child. The lower and upper limits indicating 95% confidence intervals in respective odds ratios in all nine risk factors lied above 1.00. 4. Discussion The current study evaluated a broad range of potential risk factors for unintentional poisoning in children aged 1–5 years. The risk factors were broadly categorized under environmental, psychosocial, and family related factors and personal characteristics. Unsafe storage of medicines and household chemicals [6], low parental education, low socioeconomic status, larger family size (≥4 children), and history of previous poisoning are previously reported risk factors for acute unintentional childhood poisoning [7, 8]. A recent study concluded that poor child-caregiver relationship is an important risk factor for unintentional poisoning [9]. The current study identified eight independent risk factors for acute unintentional poisoning by multivariate analysis in the binary logistic regression model. The identified risk factors were unsafe storage of medicines, unsafe storage of household chemicals, inadequate supervision of the child, mother employment during the daytime, nonauthoritative parenting styles, primary level education in the mother, poisonous plants in the home garden, and parental concern of lack of family support. Among these factors, the strongest risk factors were inadequate supervision of the child, mother being employed during daytime, and parental concern of lack of family support. Five additional risk factors showed a significant effect in univariate analysis. They were inadequate house space, unsafe use/storage of agrochemicals, psychological illnesses in parents, developmental delay in the child, and young maternal age (<19 years old). Maternal psychiatric illness has been associated with a significantly elevated risk of unintentional poisoning in children in studies from developed countries [9]. Young maternal age [10], unsafe use of pesticides [11], and overcrowding [11] have also been shown to be associated with an increased risk of childhood unintentional injuries. Thus, the current study reinforces previously reported risk factors for unintentional child injuries in different geographic locations.

ies [9]. Young maternal age [10], unsafe use of pesticides [11], and overcrowding [11] have also been shown to be associated with an increased risk of childhood unintentional injuries. Thus, the current study reinforces previously reported risk factors for unintentional child injuries in different geographic locations. A European study [8] concluded that the absence of at least one parent was associated with an increased risk of unintentional poisoning. Easy accessibility to the poison increased the risk of toxic exposure in children. It was similarly observed in the current study with inadequate supervision being observed as the strongest risk factor. Unavailability of the mother during daytime and lack of family support in looking after the child were also two of the strongest risk factors recognized in the current study. Brayden et al. pointed out that unintentional poisoning occurs secondary to several factors including unsafe storage of poisons and curiosity of children [12]. The current study in its univariate analysis observed that unsafe storage of medicines, household chemicals, and pesticides was associated with a significantly elevated risk of unintentional ingestion of the respective poisons. The investigators however did not evaluate the child's curiosity due to lack of clear indicators to quantify that factor.

ariate analysis observed that unsafe storage of medicines, household chemicals, and pesticides was associated with a significantly elevated risk of unintentional ingestion of the respective poisons. The investigators however did not evaluate the child's curiosity due to lack of clear indicators to quantify that factor. The current study observed no direct or significant association between the economic problems and subsequent poisoning in compliance with other studies [13]. The fact that no significant difference was found between groups with respect to socioeconomic status of families suggests that the magnitude of this variable does not have statistical power. There are other factors that can be attributed to lower socioeconomic conditions such as lack of social and family support, poor maternal education, disruptive family environment, poor child-caregiver relationship, and unsafe storage of poisonous substances which are likely to coexist with poor socioeconomic status. Unsafe storage of household chemicals and medicines was three and two times more reported among cases than among controls and it was consistent with findings of studies from Pakistan [3], Brazil [14], USA [15], Malaysia, and Thailand [16]. One study [17] from the African continent has noted poor parental education as a risk factor for poisoning. Similarly, the current study appreciated poor maternal education (primary education or below) as one of the strongest risk factors in univariate analysis.

[3], Brazil [14], USA [15], Malaysia, and Thailand [16]. One study [17] from the African continent has noted poor parental education as a risk factor for poisoning. Similarly, the current study appreciated poor maternal education (primary education or below) as one of the strongest risk factors in univariate analysis. In the current study, inadequate supervision was observed six times more commonly among children with unintentional poisoning compared to the control group, and this is consistent with other Asian studies which reported a fivefold higher risk for unintentional poisoning [18]. European studies also concluded that adequate supervision was the most important factor which prevented childhood poisoning accidents [18], repeatedly highlighting the need for adequate supervision of children who are under five years of age as a measure of preventing unintentional poisoning accidents.

poisoning [18]. European studies also concluded that adequate supervision was the most important factor which prevented childhood poisoning accidents [18], repeatedly highlighting the need for adequate supervision of children who are under five years of age as a measure of preventing unintentional poisoning accidents. The study did not identify a significant association between childhood behavioral characteristics and acute unintentional childhood poisoning. This was consistent with other studies published from Europe [19]. We observed that less family support was significantly associated with high risks of acute poisoning. This is consistent with findings from other studies in South Asia that concluded that extended family support is protective against unintentional poisoning [3]. Studies show that children, who are poisoned, are more likely to belong to families with few social resources [20, 21]. However, we did not observe lack of social support as a significant risk factor leading to unintentional poisoning.

uded that extended family support is protective against unintentional poisoning [3]. Studies show that children, who are poisoned, are more likely to belong to families with few social resources [20, 21]. However, we did not observe lack of social support as a significant risk factor leading to unintentional poisoning. Other significant findings of the current risk factor study are the implications of the presence of poisoning plants in the home garden, nonauthoritative parenting styles, and developmental delay in the child as major and independent risk factors for acute unintentional poisoning. Evidence from properly controlled studies is scant in the currently available literature regarding these variables and needs further studies. As the majority of unintentional poisoning occurred within home premises in the current study, a holistic approach which targets the household environment would help in managing the burden of poisoning as suggested by other studies [22]. Systematic reviews have proved that provision of safety equipment, home safety education, and heightened community awareness increase safe storage of medicines and household chemicals [23]. Since unsafe storage is among the most significant risk factors, evaluation of the effectiveness of such interventions in this population is worthwhile. The results also show that it is important to build up necessary intersector collaboration to prevent modifiable risk factors through cost-effective interventions, community education, mobilization, and awareness to ensure a safer environment for the child.

veness of such interventions in this population is worthwhile. The results also show that it is important to build up necessary intersector collaboration to prevent modifiable risk factors through cost-effective interventions, community education, mobilization, and awareness to ensure a safer environment for the child. This study has several limitations in its methodology. The study was hospital based rather than community based. It is likely that the study has not addressed the poisoning events which were not brought to medical attention during the period of the study. Further, the study was conducted only at Anuradhapura Teaching Hospital which has a wider drainage area within the north-central province of Sri Lanka. Though most children with acute poisoning are transferred from local hospitals to the teaching hospital for further management, a fraction of acute poisoning cases are likely to have been not transferred, thus being not taken into account in the current study. The investigators studied twenty-three risk factors in the case-control study design. The questionnaire underwent expert review, piloting, and evaluation by a psychometrician. However, principal component analysis was not performed and internal consistency was not calculated. Thus, the study instrument may have deficiencies in its validity. The questions were carefully selected following expert review and each risk factor was defined for the study. Data collection from all six hundred participants in the current study was carried out by one investigator and interviews were administered as soon as patients were admitted to the pediatric wards. The authors believe that this likely minimized interviewer bias and recall bias in the questionnaire administration. The sample size was determined for the study based on regional data and most risk factors showed acceptable distributions of 95% confidence intervals of odds ratios; however, given the wide distribution of respective parameters in the variable “inadequate supervision of the child,” the authors suggest reanalysis of the same parameter, recruiting a larger sample to increase reliability.

and most risk factors showed acceptable distributions of 95% confidence intervals of odds ratios; however, given the wide distribution of respective parameters in the variable “inadequate supervision of the child,” the authors suggest reanalysis of the same parameter, recruiting a larger sample to increase reliability. 5. Conclusions Children become victims of acute unintentional poisoning mostly secondary to inadequate supervision by caregivers, unsafe storage of potentially poisonous substances, and unsafe environment. As these risk factors are significantly associated with unintentional poisoning, the effect of community education to enhance vigilance, safe storage, and assurance of safe environment should be evaluated. Acknowledgments The authors of this study would like to thank Dr. Suneth Agampodi, Head, Department of Community Medicine, and Dr. Lalith Senarathna, Senior Lecturer, Faculty of Applied Sciences, Rajarata University of Sri Lanka, for providing technical advice in data analysis. The authors also thank Dr. Thilini Hemachandra and Dr. Chamila Dissanayaka of Anuradhapura Teaching Hospital, Sri Lanka, for providing their support in entering of data into statistical databases. Appendix Quantitative Data Collection Study Instrument Risk Factors for Acute Unintentional Poisoning among Children in Rural Community of Sri Lanka Date of data collection: …/…/… Hospital: … Date of admission: …/…/… Ward: … Part 1 (1) Basic Demographic Data Name: … BHT Number: … Age: … Gender: male/female Residential address: … Medical officer of health (MOH) division: … Public health midwifery (PHM) division: …

Risk Factors for Acute Unintentional Poisoning among Children in Rural Community of Sri Lanka Date of data collection: …/…/… Hospital: … Date of admission: …/…/… Ward: … Part 1 (1) Basic Demographic Data Name: … BHT Number: … Age: … Gender: male/female Residential address: … Medical officer of health (MOH) division: … Public health midwifery (PHM) division: … Parent's education level: Father: … Mother: … Parents' occupation: Father: … Mother: … Ethnicity: … Religion: … Part 2 (2.1) Reason for Hospital Admission … (2.2) Type of Poison. Household poisons/medicines/poisonous plants/agrochemicals/miscellaneous (2.2.1) General/trade name of the poison: … (2.2.2) Chemical/scientific name of the poison (if available): … Part 3 (3.1) Quantity of poison: … (3.2) Location of the poisoning event: … (3.3) Route of poisoning: ingestion/inhalation/direct skin contact/other (3.4) Transferred hospital:…/not applicable Part 4 (4.0) Proposed risk factors Present Absent (1) Inadequate space in the house Present Absent (2) Nonauthoritative parenting styles Present Absent (3) Unsafe use/storage of agrochemicals Present Absent (4) Lack of schooling/education to the child Present Absent (5) Inadequate supervision of the child Present Absent (6) Unsafe storage of household chemicals Present Absent (7) Young mother (<19 years old) Present Absent (8) Marital problems among parents Present Absent (9) Sibling disharmony Present Absent (10) Mother employed outside the country Present Absent (11) Father using alcohol or illicit drugs Present Absent (12) Unsafe storage of medicines Present Absent (13) Mother working during the daytime Present Absent (14) Economic problems in the family Present Absent (15) Single parent status Present Absent (16) Personality abnormalities in the child Present Absent

(9) Sibling disharmony Present Absent (10) Mother employed outside the country Present Absent (11) Father using alcohol or illicit drugs Present Absent (12) Unsafe storage of medicines Present Absent (13) Mother working during the daytime Present Absent (14) Economic problems in the family Present Absent (15) Single parent status Present Absent (16) Personality abnormalities in the child Present Absent (17) Behavioral abnormalities in the child Present Absent (18) Primary level education in the mother Present Absent (19) Lack of family support Present Absent (20) Psychological illness in parents Present Absent (21) Lack of social support Present Absent (22) Developmental delay in the child Present Absent (23) Poisonous plants in the home garden Present Absent Ethical Approval The study was granted ethical approval by ethical review committees of the Faculty of Medicine, University of Kelaniya, and Rajarata University of Sri Lanka. Consent Parents of all participant children provided written consent for participation of their children in the study and publication of results. Conflicts of Interest The authors declare that they have no conflicts of interest. Authors' Contributions M. B. Kavinda Chandimal Dayasiri designed the study, carried out data collection following the appropriate methodology, analyzed data, and wrote the manuscript. Shaluka F. Jayamanne and Chamilka Y. Jayasinghe designed the study, analyzed data, and supervised the manuscript writing process. Table 1 Exposures of interest: the twenty-three proposed risk factors for acute poisoning among children aged 1–5 years.

Authors' Contributions M. B. Kavinda Chandimal Dayasiri designed the study, carried out data collection following the appropriate methodology, analyzed data, and wrote the manuscript. Shaluka F. Jayamanne and Chamilka Y. Jayasinghe designed the study, analyzed data, and supervised the manuscript writing process. Table 1 Exposures of interest: the twenty-three proposed risk factors for acute poisoning among children aged 1–5 years. Risk factor category Proposed risk factor (1) Environmental risk factors (1) Unsafe storage of medicines (2) Unsafe storage of household chemicals (3) Unsafe use/storage of agrochemicals (4) Inadequate space in the house (5) Inadequate supervision of the child (6) Poisonous plants in the neighborhood/home garden (2) Psychosocial risk factors (1) Psychological problems in parents (2) Lack of social support (3) Lack of schooling/education to the child (3) Family related factors (1) Father using alcohol or illicit drugs (2) Problems with the siblings (3) Incorrect parenting styles (4) Mother employed outside the country (5) Economic problems in the family (6) Mother working during the daytime (7) Poor education in the mother (<grade 8) (8) Young mother (<21 years old) (9) Marital problems among parents (10) Lack of family support (11) Single parent status (4) Personal characteristics (1) Developmental problems in the child (2) Personality abnormalities in the child (3) Behavioral abnormalities in the child Table 2 Age distributions of two groups: “children with acute unintentional poisoning” and “control group.”

(3) Family related factors (1) Father using alcohol or illicit drugs (2) Problems with the siblings (3) Incorrect parenting styles (4) Mother employed outside the country (5) Economic problems in the family (6) Mother working during the daytime (7) Poor education in the mother (<grade 8) (8) Young mother (<21 years old) (9) Marital problems among parents (10) Lack of family support (11) Single parent status (4) Personal characteristics (1) Developmental problems in the child (2) Personality abnormalities in the child (3) Behavioral abnormalities in the child Table 2 Age distributions of two groups: “children with acute unintentional poisoning” and “control group.” Age Children with unintentional poisoning Control group 1-2 years old 107 (35.7%) 107 (35.7%) 2–4 years old 153 (51.0%) 153 (51.0%) 4-5 years old 40 (13.3%) 40 (13.3%) Total 300 (100%) 300 (100%) Table 3 Sex distribution of the two groups with “acute poisoning” and “controls.” Sex Children with unintentional poisoning Control group Male 169 (56.3%) 169 (56.3%) Female 131 (43.7%) 131 (43.7%) Total 300 (100%) 300 (100%) Table 4 Univariate unadjusted analysis of risk factors in the binary logistic regression model.

Total 300 (100%) 300 (100%) Table 3 Sex distribution of the two groups with “acute poisoning” and “controls.” Sex Children with unintentional poisoning Control group Male 169 (56.3%) 169 (56.3%) Female 131 (43.7%) 131 (43.7%) Total 300 (100%) 300 (100%) Table 4 Univariate unadjusted analysis of risk factors in the binary logistic regression model. Proposed risk factor Cases Controls Odds ratio 95% CI (OR) p value Low High Environmental risk factors (1) Unsafe storage of medicines 142 (47.3%) 72 (24%) 2.85 2.04 4.00 <0.001 (2) Unsafe storage of household chemicals 168 (56%) 54 (18%) 5.80 4.00 8.40 <0.001 (3) Unsafe use/storage of agrochemicals 58 (19.3%) 38 (12.7%) 1.65 1.06 2.57 0.027 (4) Inadequate space in the house 102 (34%) 67 (22.4%) 1.79 1.24 2.57 0.002 (5) Inadequate supervision of the child 249 (83%) 39 (13%) 32.26 20.83 52.60 <0.001 (6) Poisonous plants in the home garden 95 (31.7%) 23 (7.7%) 5.58 3.42 9.09 <0.001 Psychosocial risk factors (1) Psychological illness in parents 11 (3.7%) 0 — — — <0.001 (2) Lack of social support 43 (14.3%) 42 (14%) 1.02 0.64 1.62 0.907 (3) Lack of schooling/education to the child 5 (1.7%) 7 (2.3%) 0.70 0.22 2.26 0.562 Family related risk factors (1) Father using alcohol or illicit drugs 61 (20.3%) 87 (29%) 0.62 0.43 0.91 0.014 (2) Sibling disharmony 12 (4%) 18 (6%) 0.65 0.31 1.38 0.260 (3) Nonauthoritative parenting styles 35 (11.7%) 2 (0.7%) 19.6 4.69 83.3 <0.001 (4) Mother employed outside the country 2 (0.7%) 5 (1.7%) 0.40 0.08 2.06 0.270 (5) Economic problems in the family 115 (38.3%) 155 (51.7%) 0.58 0.42 0.81 0.001 (6) Mother working during the daytime 68 (22.7%) 14 (4.7%) 5.98 3.28 10.87 <0.001 (7) Primary level education in the mother 51 (17%) 15 (5%) 3.89 2.13 7.09 <0.001 (8) Young mother (<19 years old) 37 (12.3%) 19 (6.3%) 2.08 1.16 3.70 0.013 (9) Marital problems among parents 23 (7.7%) 13 (4.3%) 1.83 0.91 3.69 0.090 (10) Lack of family support 122 (40.7%) 42 (14%) 4.20 2.82 6.28 <0.001 (11) Single parent status 2 (0.7%) 3 (1%) 0.67 0.11 4.00 0.656 Personal characteristics (1) Developmental delay in the child 18 (6%) 3 (1%) 6.31 1.84 21.68 0.003 (2) Personality abnormalities in the child 3 (1%) 2 (0.7%) 1.51 0.24 9.09 0.656 (3) Behavioral abnormalities in the child 6 (2%) 2 (0.7%) 3.03 0.61 15.10 0.175 Table 5 Multivariate analysis of independent risk factors for acute unintentional pediatric poisoning in the binary logistic regression model.

1.84 21.68 0.003 (2) Personality abnormalities in the child 3 (1%) 2 (0.7%) 1.51 0.24 9.09 0.656 (3) Behavioral abnormalities in the child 6 (2%) 2 (0.7%) 3.03 0.61 15.10 0.175 Table 5 Multivariate analysis of independent risk factors for acute unintentional pediatric poisoning in the binary logistic regression model. Proposed risk factor Odds ratio 95% CI (OR) p value Low High Environmental risk factors (1) Unsafe storage of household chemicals 2.70 1.49 4.90 0.001 (2) Inadequate supervision of the child 28.50 15.38 52.60 <0.001 (3) Poisonous plants in the home garden 3.67 1.70 7.93 0.001 Family related risk factors (1) Mother working during the daytime 7.14 2.89 17.54 <0.001 (2) Nonauthoritative parenting styles 12.34 2.14 71.40 0.005 (3) Primary level education in the mother 2.73 1.05 7.14 0.039 (4) Lack of family support 17.54 3.65 83.3 <0.001 Personal characteristics (1) Developmental delay in the child 7.93 1.57 40.00 0.012

1. Introduction With an estimated population of more than 30 million inhabitants, Afghanistan is composed of more than ten ethnic and tribal groups, most of whom have lived together in the country for centuries. These include the majority Pashtuns, who constitute almost one-half of the population, followed by a quarter of the population of Tajiks (27%) and sizeable communities of Uzbecs (9%) and Hazara (9%). Turkmen (3%), Aimaq (4%), Baluch (2%), and small communities of Brahui, Nuristani, Pashaie, Pamiri, Khirgiz, and Qizilbash are also represented. Each of these groups has developed its own forms of languages, culture, and religious beliefs over the course of Afghanistan's history. However, the centuries-long interaction between all these groups, though distinguishable by accent and clothing (as examples), has resulted in a cultural blending of various Afghan ethnic and tribal traditions. The country is almost exclusively Muslim with a majority Sunni population (80%) and an estimated 19% Shi'i population. Afghan, Persian, or Dari is the official language, spoken by about one-half of the population, with Pashtu, also an official language, spoken by some 35% of the population. Turkic languages are also spoken by some groups (11%), as well as another thirty minor languages that have been identified (e.g., Baluchi and Pashai). Many individuals speak more than one language.

spoken by about one-half of the population, with Pashtu, also an official language, spoken by some 35% of the population. Turkic languages are also spoken by some groups (11%), as well as another thirty minor languages that have been identified (e.g., Baluchi and Pashai). Many individuals speak more than one language. Afghanistan experienced a long period of relative peace until 1978. Up until that point, it was a relatively thriving and vibrant nation that provided women and children with many appropriate health and social services and freedoms. A short period of unrest and then establishment of a Soviet-allied regime followed, leading to invasion by Soviet troops in 1979. This invasion led to 10 years of armed resistance against Soviet troops, resulting in a mass exodus of approximately 25% of the total Afghan population, a majority of whom were women and children [1], followed by years of civil war among warlords for control of the country, with, first, the establishment of the Democratic Republic of Afghanistan, subsequently overthrown by the Islamic State of Afghanistan, eventually leading to Taliban rule of Afghanistan in 1997. The United States then launched “Operation Enduring Freedom” leading to the establishment of a democratic government in 2001. Since that time, Afghanistan has witnessed a fragile peace, threatened repeatedly by the Taliban and Al-Qaeda forces [2].

Afghanistan, eventually leading to Taliban rule of Afghanistan in 1997. The United States then launched “Operation Enduring Freedom” leading to the establishment of a democratic government in 2001. Since that time, Afghanistan has witnessed a fragile peace, threatened repeatedly by the Taliban and Al-Qaeda forces [2]. The result of this continuing warfare and violence in Afghanistan has been multifold. Afghanistan is now one of the poorest, most ravaged countries in the world. Figures for 2012 indicate that its population has an average life expectancy of 60.5 years, and its infant mortality rate is one of the highest in the world. The percentage of children moderately or severely underweight is 33%. Among adults living in Afghanistan, the most common stressful events experienced include lack of shelter (70%) and lack of food and water (56%). Anxiety, depressive symptoms, and posttraumatic stress disorder (PTSD) were identified in 72%, 68%, and 42% of respondents, respectively. Women were found to have worse mental health status relative to men [3]. Only 53.9% of the population has access to mobile phones and 5.5% have access to the Internet [4].

ter (56%). Anxiety, depressive symptoms, and posttraumatic stress disorder (PTSD) were identified in 72%, 68%, and 42% of respondents, respectively. Women were found to have worse mental health status relative to men [3]. Only 53.9% of the population has access to mobile phones and 5.5% have access to the Internet [4]. With more than 30 years of war and conflict, the educational system of Afghanistan is extremely strained. Figures from 2011 [5] indicate that attendance rates in primary and secondary school in Afghanistan are 66% and 18% for boys and 40% and 6% for girls, respectively. The literacy rate for individuals 15–24 years old is 49% for males and only 18% for females. Among all adults, the literacy rate is 43% for men and 13% for women [4]. Thus, there is a dire need for women and men of all ages to have basic education.

nistan are 66% and 18% for boys and 40% and 6% for girls, respectively. The literacy rate for individuals 15–24 years old is 49% for males and only 18% for females. Among all adults, the literacy rate is 43% for men and 13% for women [4]. Thus, there is a dire need for women and men of all ages to have basic education. A survey in 2010 [6] reported that approximately 1 million (8%) Afghans 15–64 years of age are dependent on psychoactive substances, a percentage twice that of the global average [6]. The Afghanistan National Urban Drug Use Survey found that 11.4% of the population tested positive for any psychoactive substance, with opioids prevalence at 5.6%, accounting for more than 50% of substance use in women and children but only a third of substance use in men (Cottler and Colleagues) [7]. A study conducted by Todd and colleagues [8] showed that 385 out of 483 IDUs completed one or more follow-up visits. All participants were male with a median age of 28 years and a median duration of injecting of 2 years. HCV and HIV risk are high and there is a need for increasing awareness of HCV transmission and overdose risk and to prepare clients for harm reduction needs during conflict and displacement and efforts are needed to engage community and police force. The Afghanistan National Drug Use Survey (ANDUS) 2015 [9] estimated between 2 and 2.4 million individuals in Afghanistan use psychoactive substances, which is 7.3% of the population (16.1% men, 9.5% women, and 0.8% children). Afghanistan has one of the highest opiates use rates worldwide. The national adult substance use rate is 12.6%, more than double the global psychoactive substance use rate of 5.2%. Substance use is estimated to affect almost one in three households in the country. The rate of substance use in rural areas is 2.5 times that in urban areas. Children in rural areas are far more likely to test positive for substance use. The substance-positive rate among rural children is estimated at 11.3% while for urban children the rate is estimated at 2.3%. Almost 91% of children who test positive have been affected by secondary contact with smoke from substances, typically opium, smoked by adults in the home or care environment. Opioids are the most commonly used psychoactive substance.

hildren is estimated at 11.3% while for urban children the rate is estimated at 2.3%. Almost 91% of children who test positive have been affected by secondary contact with smoke from substances, typically opium, smoked by adults in the home or care environment. Opioids are the most commonly used psychoactive substance. Gupta [10] surveyed 300 children 8–18 years of age in Kabul, finding that 72% had lost a family member in the past five years, among whom 40% had been a parent. Being a witness to acts of violence and death was almost universal in the sample. Almost 75% believe they would not live into adulthood, and more than 50% could be seen to have psychological problems that affected their childhood activities. Catani [11] and colleagues reported on schoolchildren 7–15 years of age in Kabul (122 girls, 165 boys). Traumatic events, particularly violence in the home, were experienced more frequently by boys than by girls. Probably posttraumatic stress disorder was found in 26% of boys and 14% of girls. Child labor was a frequent occurrence and was a risk factor for maltreatment of girls at home. Like their adult counterparts, girls are more likely than boys to suffer mental health problems and have depressive symptoms [12].

an by girls. Probably posttraumatic stress disorder was found in 26% of boys and 14% of girls. Child labor was a frequent occurrence and was a risk factor for maltreatment of girls at home. Like their adult counterparts, girls are more likely than boys to suffer mental health problems and have depressive symptoms [12]. The literature regarding the state of affairs in Afghanistan illustrates several important themes that serve as the basis for an intervention targeting children. These themes include opium use, lack of formal education, lack of basic health education, high rates exposure to stressful living situations, violence and trauma, and related comorbid mental health problems. The Child Intervention for Living Drug-free (CHILD) Protocol [formerly known as the Child Addiction Treatment (CAT) protocol] (described in detail, below) is a novel, culturally sensitive psychosocial intervention designed to address the multidimensional nature of problems faced by these children in a residential setting that also provided supportive services for the children. CHILD was not a research protocol; rather, it was an empirically based intervention developed to meet the needs of substance-using children in Afghanistan. However, it collected outreach, pretreatment, posttreatment, and follow-up data to inform the treatment staff of progress of each child during and after treatment. The purpose of the present study is twofold: (1) describing the implementation of the CHILD intervention protocol and (2) reporting preliminary outcome data in regard to its impact.

The literature regarding the state of affairs in Afghanistan illustrates several important themes that serve as the basis for an intervention targeting children. These themes include opium use, lack of formal education, lack of basic health education, high rates exposure to stressful living situations, violence and trauma, and related comorbid mental health problems. The Child Intervention for Living Drug-free (CHILD) Protocol [formerly known as the Child Addiction Treatment (CAT) protocol] (described in detail, below) is a novel, culturally sensitive psychosocial intervention designed to address the multidimensional nature of problems faced by these children in a residential setting that also provided supportive services for the children. CHILD was not a research protocol; rather, it was an empirically based intervention developed to meet the needs of substance-using children in Afghanistan. However, it collected outreach, pretreatment, posttreatment, and follow-up data to inform the treatment staff of progress of each child during and after treatment. The purpose of the present study is twofold: (1) describing the implementation of the CHILD intervention protocol and (2) reporting preliminary outcome data in regard to its impact. 2. Methods 2.1. Institutional Review Board (IRB) Approval The study protocol, including informed consent procedures, was reviewed and approved by the Johns Hopkins University Institutional Review Board as well as the Ministry of Public Health of Afghanistan's Institutional Review Board.

The purpose of the present study is twofold: (1) describing the implementation of the CHILD intervention protocol and (2) reporting preliminary outcome data in regard to its impact. 2. Methods 2.1. Institutional Review Board (IRB) Approval The study protocol, including informed consent procedures, was reviewed and approved by the Johns Hopkins University Institutional Review Board as well as the Ministry of Public Health of Afghanistan's Institutional Review Board. 2.2. Informed Consent All participants and/or their legal guardian(s) provided written informed consent to take part in the study. The aims of the study were explained and participants were informed that they could withdraw at any time without any further obligation to provide data, and they would continue in residential treatment but would not be provided with the CHILD treatment components. Consent was read to the parent(s)/guardian and if they could not write, they applied a thumb print to stamp the consent. In addition, during outreach the outreach team used assent (oral consent) because in outreach it was very difficult to obtain written consent based on the general Afghans culture beliefs and practices to such a request as well as the low literacy rate. 2.3. Participants Participants were 699 children (373 girls, 326 boys), 4–7 years of age, and 84 older children (1 girl, 83 boys) 8–18 years of age. No additional demographic information was collected from the children.

2.2. Informed Consent All participants and/or their legal guardian(s) provided written informed consent to take part in the study. The aims of the study were explained and participants were informed that they could withdraw at any time without any further obligation to provide data, and they would continue in residential treatment but would not be provided with the CHILD treatment components. Consent was read to the parent(s)/guardian and if they could not write, they applied a thumb print to stamp the consent. In addition, during outreach the outreach team used assent (oral consent) because in outreach it was very difficult to obtain written consent based on the general Afghans culture beliefs and practices to such a request as well as the low literacy rate. 2.3. Participants Participants were 699 children (373 girls, 326 boys), 4–7 years of age, and 84 older children (1 girl, 83 boys) 8–18 years of age. No additional demographic information was collected from the children. The data on the children reported in this paper is for the first 783 children who came in contact with the CHILD program and were among the first of several thousand children who were screened by the program and entered treatment. Moreover, of these participants, 144 younger children were screened and went directly to residential treatment, with no outpatient period, and 8 directly entered residential treatment without screening, while 30 older children entered outpatient treatment without screening. These variations to the screening protocol were largely based on treatment need.

4 younger children were screened and went directly to residential treatment, with no outpatient period, and 8 directly entered residential treatment without screening, while 30 older children entered outpatient treatment without screening. These variations to the screening protocol were largely based on treatment need. 2.4. Procedures The Child Intervention for Living Drug-free (CHILD) protocol was implemented in the Kabul, Herat, Balkh, Nangarhar, and Badakhshan provinces of Afghanistan. The project has three interconnected components: outreach services, outpatient services, and residential substance use treatment. There were 12 outreach teams, 10 outpatient centers, and 10 substance use treatment centers, with a total of 325 beds (120 beds for children, 70 beds for women, 110 beds for male adolescents, and 25 beds for female adolescents). Treatment duration for residential treatment was 45 days for children and 180 days for older children.

12 outreach teams, 10 outpatient centers, and 10 substance use treatment centers, with a total of 325 beds (120 beds for children, 70 beds for women, 110 beds for male adolescents, and 25 beds for female adolescents). Treatment duration for residential treatment was 45 days for children and 180 days for older children. Outreach activities involved trained staff circulating in the capital city of each province as well as nearby secure districts to identify children who were at risk of substance use or actively using substances. Outreach staff did not leave the secure areas for safety reasons. Outreach staff approached children under one of three different scenarios. In the first scenario, they approached a family with a child who had been identified as using illicit substances by a community leader; in the second, they approached children directly on the street; in the third, they approached children in orphanages. Each outreach team used a screener form (see Measures). A respondent who scored positive (one or more true responses) in any of the above five screening areas was referred to an outpatient center for assessment.

cond, they approached children directly on the street; in the third, they approached children in orphanages. Each outreach team used a screener form (see Measures). A respondent who scored positive (one or more true responses) in any of the above five screening areas was referred to an outpatient center for assessment. Trained staff in the outpatient centers assessed the children with a battery of assessment instruments (see Measures). A child positive for substance use was referred to a residential center for treatment, while a child negative for substance use but deemed at risk for such use would visit an outpatient center every day for eight weeks to receive a comprehensive psychosocial intervention that included education, life skills, and individual counseling. Outpatient centers also provided treatment for minor ailments as well as lunch and snacks. A child with severe medical or psychological problem was referred for appropriate services in the community.

ks to receive a comprehensive psychosocial intervention that included education, life skills, and individual counseling. Outpatient centers also provided treatment for minor ailments as well as lunch and snacks. A child with severe medical or psychological problem was referred for appropriate services in the community. Children who were deemed to be at risk for psychoactive substance use or were actively using such substances were referred to residential centers, where they were assessed prior to treatment with the same measures as used in outpatient center assessments to develop a treatment plan. If a child had a family member who was using psychoactive substance(s), then staff considered the child as at risk, while if the child himself or parent(s) were reporting that the child was actively using psychoactive substances then s/he was considered using a psychoactive substance. The length of time between outpatient and residential intake assessment was variable and dependent on when the child appeared at the residential treatment center but typically was between 2 and 7 days. Upon completion of treatment, the patients were again assessed to evaluate the treatment progress. After discharge from the residential treatment center the child or adolescent was followed in the community on a weekly basis by an outreach team. At the end of 12 weeks in the community, outreach staff conducted a reassessment to determine the need for further services or to determine that the weekly visits could be terminated.

harge from the residential treatment center the child or adolescent was followed in the community on a weekly basis by an outreach team. At the end of 12 weeks in the community, outreach staff conducted a reassessment to determine the need for further services or to determine that the weekly visits could be terminated. 2.5. Outreach Staff: Training and Experience Outreach staff were experienced social workers and psychologists trained both in how to carryout outreach activities and in the CHILD protocol. 2.6. Assessment Staff: Training and Experience Outpatient staff were psychologists, social workers, and a medical doctor, all of whom were trained in how to conducted assessment activities and in the CHILD protocol. 2.7. Interventionist Staff: Training and Experience Interventionist staff were members of a multidisciplinary team consisting of experienced psychologists, social workers, nurses, and medical doctors, all of whom were trained on the CHILD protocol. 2.8. Intervention: Child Intervention for Living Drug-Free (CHILD) Protocol CHILD is a comprehensive psychosocial intervention built on multiple prevention and treatment platforms and so employs motivational interviewing techniques, contingency management, skill-building education, traditional education, trauma-informed care, and art therapy techniques. It was developed for use with children either at risk for or actively using psychoactive substances.

multiple prevention and treatment platforms and so employs motivational interviewing techniques, contingency management, skill-building education, traditional education, trauma-informed care, and art therapy techniques. It was developed for use with children either at risk for or actively using psychoactive substances. The CHILD psychosocial program was provided 5 times a week to participants whom screening indicated were at risk for or actively using psychoactive substances while they were inpatient (45 days for children and 180 days for adolescents). Each group session lasted approximately 1 hour. An example schedule of inpatient components of treatment for the younger children can be found in Table 1. The older children schedule repeated this 6-week cycle three additional times, during which time the material covered included past material that was both reinforced and new material that was introduced. The components of the CHILD protocol were as follows. Age-Appropriate Basic Education. For younger children, the foundation of education is covered thrice weekly. Examples of topics covered include the alphabet, shapes, size relationships, opposites, writing of letters and numbers, and numbers and numeric relationships. For older children, literacy and basic mathematic functions, applying math to daily living, are covered.

n, the foundation of education is covered thrice weekly. Examples of topics covered include the alphabet, shapes, size relationships, opposites, writing of letters and numbers, and numbers and numeric relationships. For older children, literacy and basic mathematic functions, applying math to daily living, are covered. Drug Education. In an age-appropriate manner, children were taught that drugs change the way your body and mind work. Some medicines are helpful when used in the right way. Helpful and harmful drugs are explained. How drugs cause illness, impaired coordination, slowed growth, and emotional harm such as feelings of isolation or paranoia is discussed, with a focus on developing life skills such as refusal skills and decision making skills. The legal issues associated with drug and alcohol use are treated, because a conviction for a drug offense can lead to prison, loss of a job education. Talk about positive, drug-free alternatives forms the basis of the lesson, and they are explored with the counselor. Discussions are focused on providing information about the harmful use of drugs, without using judgmental terms and without creating a sense of fear about drug use. Nutrition. Using the UNICEF publications, the nutritional information is tailored to the age of the children. Topics covered include a discussion of what are healthy foods and the types of vitamins and minerals needed for healthy growth, and how to prepare healthy foods and simple meals is discussed.

Drug Education. In an age-appropriate manner, children were taught that drugs change the way your body and mind work. Some medicines are helpful when used in the right way. Helpful and harmful drugs are explained. How drugs cause illness, impaired coordination, slowed growth, and emotional harm such as feelings of isolation or paranoia is discussed, with a focus on developing life skills such as refusal skills and decision making skills. The legal issues associated with drug and alcohol use are treated, because a conviction for a drug offense can lead to prison, loss of a job education. Talk about positive, drug-free alternatives forms the basis of the lesson, and they are explored with the counselor. Discussions are focused on providing information about the harmful use of drugs, without using judgmental terms and without creating a sense of fear about drug use. Nutrition. Using the UNICEF publications, the nutritional information is tailored to the age of the children. Topics covered include a discussion of what are healthy foods and the types of vitamins and minerals needed for healthy growth, and how to prepare healthy foods and simple meals is discussed. Hygiene. In an age-appropriate manner, the importance of hand-washing, brushing teeth, and caring for the body is discussed. The use of toilets or latrines and the need to practice good hygiene, protect water sources, and safely dispose of waste water and refuse are presented. Children are taught how to make soap.

Nutrition. Using the UNICEF publications, the nutritional information is tailored to the age of the children. Topics covered include a discussion of what are healthy foods and the types of vitamins and minerals needed for healthy growth, and how to prepare healthy foods and simple meals is discussed. Hygiene. In an age-appropriate manner, the importance of hand-washing, brushing teeth, and caring for the body is discussed. The use of toilets or latrines and the need to practice good hygiene, protect water sources, and safely dispose of waste water and refuse are presented. Children are taught how to make soap. Personal Safety. In an age-appropriate manner, topics covered include basic living safety such as avoiding open cooking fires, bare light bulbs, live electrical wires, land mines, identifying hazards in the home, basic ways to stay healthy, and how to interact with adults and avoid and/or keep themselves safe during family violence situations. Trauma Coping Skills. Participants are taught to deal with intrusive thoughts and feelings, skills to reduce arousal (relaxing, concentrating, and sleeping), and skills to manage avoidance (fears/difficulties facing reminders). Role-playing and practices of skills are undertaken. Participants draw, write, and talk about the incidents. They are shown how to look to the future rather than the past (avoidance).

skills to reduce arousal (relaxing, concentrating, and sleeping), and skills to manage avoidance (fears/difficulties facing reminders). Role-playing and practices of skills are undertaken. Participants draw, write, and talk about the incidents. They are shown how to look to the future rather than the past (avoidance). Communication Skills. In an age-appropriate manner, effective communication techniques are taught. Items covered include interaction and communication with adults, conflict resolution, positive self-talk, culturally appropriate self-advocacy communication, and deescalation of angry situations. Structured Art Expression Techniques. In an age-appropriate manner, participants are given materials and guided to explore and express emotions. Topics include making self-masks, memory boxes, boxes representing themselves, painting pictures to express emotions, and creating clay objects to express stories about their life events.

rt Expression Techniques. In an age-appropriate manner, participants are given materials and guided to explore and express emotions. Topics include making self-masks, memory boxes, boxes representing themselves, painting pictures to express emotions, and creating clay objects to express stories about their life events. 2.9. Measures Four measures (SDQ, CRIES, ASCL, and SRQ-20) are already available in Afghanistan in Dari and Pashto versions; two (SCARED and QOL) were translated into Dari/Pashto and fully backtranslated into English by members of the research team who were fluent in both Dari/Pashtu and English and then modified as necessary to permit culturally sensitive administration of the items. Although all measures were originally developed as self-report instruments, because of the low literacy rate in the country and very low literacy rate among the substance-using child population all instruments were administered by clinic staff who had been trained in their administration. Administration was in the language chosen by the child. Measures were chosen to be age-appropriate, so that there was one set of measures for younger children and another set for older children. Only one measure was in common to both groups, other than the screening instrument. 2.9.1. Measures Administered to Both Younger and Older Children

2.9. Measures Four measures (SDQ, CRIES, ASCL, and SRQ-20) are already available in Afghanistan in Dari and Pashto versions; two (SCARED and QOL) were translated into Dari/Pashto and fully backtranslated into English by members of the research team who were fluent in both Dari/Pashtu and English and then modified as necessary to permit culturally sensitive administration of the items. Although all measures were originally developed as self-report instruments, because of the low literacy rate in the country and very low literacy rate among the substance-using child population all instruments were administered by clinic staff who had been trained in their administration. Administration was in the language chosen by the child. Measures were chosen to be age-appropriate, so that there was one set of measures for younger children and another set for older children. Only one measure was in common to both groups, other than the screening instrument. 2.9.1. Measures Administered to Both Younger and Older Children Screening Form (SF). The SF consisted of 16 3-point questions (e.g., 0 = “None”; 1 = “Some”; 2 = “A lot”) that screened for problems in five areas: behavioral, emotional, and social problems; psychological distress; trauma exposure; physical health/medical problems; and substance use. A respondent who scored above 0 in any of the above five areas was referred to an outpatient center for assessment.

1 = “Some”; 2 = “A lot”) that screened for problems in five areas: behavioral, emotional, and social problems; psychological distress; trauma exposure; physical health/medical problems; and substance use. A respondent who scored above 0 in any of the above five areas was referred to an outpatient center for assessment. Child Strengths and Difficulties Questionnaire (SDQ). The SDQ [13, 14] is an internationally validated 25-item questionnaire providing balanced coverage of behavioral, emotional, and social problems that asks the child about “your behavior over the past 6 months.” Items are scored 0, 1, or 2 (with some items reverse-scored). The SDQ has 5 scales of 5 items each: Emotional Symptoms (ES), Conduct Problems (CP), Hyperactivity (H), Peer Problems (PP), and Prosocial (P) scales. There is a separate score for each scale. There is also a total difficulties' score, which is ES + CP + H + PP scores (omitting the Prosocial subscale). 2.9.2. Measures Administered to Younger Children Child Revised Impact of Events Scale (CRIES). The CRIES [15–18] is a screening tool measuring risk for child posttraumatic stress symptoms. It has 8 questions that ask the child to respond on 4-point scale with “Not at all” = 0, to “Often” = 3. It has shown good internal consistency reliability (Cronbach's α = 0.82) and seven-day test-retest reliability (r = 0.78, p < .0001) in research in Afghanistan.

easuring risk for child posttraumatic stress symptoms. It has 8 questions that ask the child to respond on 4-point scale with “Not at all” = 0, to “Often” = 3. It has shown good internal consistency reliability (Cronbach's α = 0.82) and seven-day test-retest reliability (r = 0.78, p < .0001) in research in Afghanistan. Self-Report for Childhood Anxiety Related Emotional Disorders (SCARED). The SCARED [19] is a reliable and valid screening tool for determining anxiety disorders in children and adolescents. It has 41 items that ask the child to indicate feelings for the last 3 months. Items are scored “Not true” = 0, “Sometimes true” = 1, or “Very true” = 2. There are 5 subscales: a 13-item Panic Disorder (P), a 9-item Generalized Anxiety Disorder (GA), an 8-item Separation Anxiety Disorder (SeA), a 7-item Social Anxiety Disorder (SoA), and a 4-item Significant School Avoidance (SS). There is also a total SCARED score, which is P + GA + SeA + SoA + SS. 2.9.3. Measures Administered to Older Children Afghan Symptom Checklist (ASCL). The ASCL [20] is a 22-item check list that yields a single score with excellent reliability (α = .93) and good construct validity, correlating strongly with a measure of exposure to war-related violence and loss (r = .70). Self-Reporting Questionnaire-20 (SRQ-20). The SRQ-20 [19, 21] was developed by the WHO to screen for psychiatric disturbance in individuals in developing countries. It has adequate reliability and internal consistency. The SRQ-20 [21] has 20 items that ask about problems “bothering you the last 30 days.” Items are scored “No” = 0 or “Yes” = 1.

ire-20 (SRQ-20). The SRQ-20 [19, 21] was developed by the WHO to screen for psychiatric disturbance in individuals in developing countries. It has adequate reliability and internal consistency. The SRQ-20 [21] has 20 items that ask about problems “bothering you the last 30 days.” Items are scored “No” = 0 or “Yes” = 1. Quality of Life (QOL). This measure was developed specifically for the Child Addiction Treatment project in Afghanistan with questions drawn from other existing quality of life measures and tailored to the Afghan culture. It asks the respondent to indicate “In the past 3 months, how often have you experienced…” problem severity in 20 areas of functioning. Items are scored “Never or Almost Never” = 1, to “Always or Almost Always” = 5. It has 4 subscales: a 6-item Physical Health (PH), a 5-item Mental Health (MH), a 5-item Friends (F), and a 4-item Home (H) subscales. There is a separate score for each subscale. There is also a total QOL score, which is PH + MH + F + H scores. 2.10. Data Entry and Management A Microsoft Access database was written and developed to enter and store all data. Staff at the health facilities level were trained in how to enter data and use the database. All data were double-checked at the central hub data center in Kabul. 2.11. Statistical Analysis Because the length of residential treatment was different for the younger and older children, analyses were conducted separately for each group.

2.10. Data Entry and Management A Microsoft Access database was written and developed to enter and store all data. Staff at the health facilities level were trained in how to enter data and use the database. All data were double-checked at the central hub data center in Kabul. 2.11. Statistical Analysis Because the length of residential treatment was different for the younger and older children, analyses were conducted separately for each group. A general linear mixed model (GLMM) analysis was conducted on the scores on the scales and/or subscales of the above-named measures, as appropriate to the measure, with all scale and subscale scores assumed to follow a normal distribution in the population. For analysis of the child data, there were three effects in the model: gender as the fixed between-subjects factor, time (outreach, residential intake, posttreatment, and follow-up) as the fixed repeated factor, and their interaction; for the adolescent data, the one female observation was omitted, and so there was a single effect: time (outreach, residential intake, posttreatment, and follow-up) as the fixed repeated factor. A familywise error rate was used to set α = .00192 (.05/26), where the nominal Type I error rate was set at .05, and the family included both the 13 assessment scales/subscales for the younger children (CRIES total score, the 5 subscale scores and the SCARED total score, and the 5 subscale scores and the SDQ total score) and the 13 assessment scales/subscales for the older children (ASCL total score, SRQ total score, the 5 subscale scores and the SDQ total score, and the 4 subscale scores and the QOL total score). Post hoc testing of simple mean differences associated with significant effects utilized the Dunn-Sidak multiple comparison test to control the post hoc testing error rate to at most the familywise error rate [22]. All analyses were performed by SAS version 9.3 [23].

and the 4 subscale scores and the QOL total score). Post hoc testing of simple mean differences associated with significant effects utilized the Dunn-Sidak multiple comparison test to control the post hoc testing error rate to at most the familywise error rate [22]. All analyses were performed by SAS version 9.3 [23]. 3. Results 3.1. Feasibility and Acceptance Individual interviews with project staff and participants showed high rates of participation and high levels of satisfaction (both over 90%) from participants and staff. Although each of the education, life skills, and one-to-one sessions was scheduled for a 60-minute time slot, some sessions consistently ran over the allotted time because children were particularly engaged. Comments about the intervention were positive (e.g., I liked the art and games; I always learn something). Overall, staff members also reported a high level of satisfaction with the intervention and expressed the need for continued implementation.

y ran over the allotted time because children were particularly engaged. Comments about the intervention were positive (e.g., I liked the art and games; I always learn something). Overall, staff members also reported a high level of satisfaction with the intervention and expressed the need for continued implementation. 3.2. Follow-Up Rates The treatment completion rate (residential intake assessment to posttreatment assessment) was 88% for younger children and 98% for older children, both outstanding considering the war-torn intervention setting and the fact that treatment dropout rates reported by inpatient drug studies have ranged from 19% to 63% [24, 25] for US adult patients in intensive inpatient programs in shorter treatment (21–28 days). Follow-up rates from posttreatment to follow-up assessments were disappointing—14% for younger children and 51% for older children. The low follow-up rate was due to a lack of emphasis by the supervisory staff in follow-up data collection and the staff were not paid to complete such visits. It was voluntary for staff to visit children in their homes during the follow-up phase and that practice proved impractical. 3.3. Measure Reliability Internal consistency reliability Cronbach's α was outstanding for all scales (see Tables 2 and 3) and, for those scales with subscales, generally good to outstanding for the respective subscales, with the exception of the SDQ Peer Problems subscale for children.

3.2. Follow-Up Rates The treatment completion rate (residential intake assessment to posttreatment assessment) was 88% for younger children and 98% for older children, both outstanding considering the war-torn intervention setting and the fact that treatment dropout rates reported by inpatient drug studies have ranged from 19% to 63% [24, 25] for US adult patients in intensive inpatient programs in shorter treatment (21–28 days). Follow-up rates from posttreatment to follow-up assessments were disappointing—14% for younger children and 51% for older children. The low follow-up rate was due to a lack of emphasis by the supervisory staff in follow-up data collection and the staff were not paid to complete such visits. It was voluntary for staff to visit children in their homes during the follow-up phase and that practice proved impractical. 3.3. Measure Reliability Internal consistency reliability Cronbach's α was outstanding for all scales (see Tables 2 and 3) and, for those scales with subscales, generally good to outstanding for the respective subscales, with the exception of the SDQ Peer Problems subscale for children. 3.4. Change Over Time: Younger Children There were no significant Gender or Gender X Time effects (all p's > .16 and .35, resp.). The time main effect was significant for all outcomes (all p's < the per-comparison error rate of .00192) [see means (Standard Errors) in Table 2]. Post hoc testing indicated that there were significant mean decreases from residential treatment entry to residential treatment completion for all scales (the SDQ Prosocial subscale significantly increasing), with the smallest percentage change of 49%. The SCARED Generalized Anxiety Disorder and School Avoidance subscales and the SDQ Emotional Symptoms, Conduct Problems, and Hyperactivity subscales and the SDQ total mean scores significantly increased from residential treatment completion to community follow-up total score, with the SDQ Prosocial subscale mean significantly decreasing. However, with the exception of the CRIES and the SDQ Prosocial subscale means, all scales and subscale means were significantly lower at community follow-up than at residential treatment entry.

ential treatment completion to community follow-up total score, with the SDQ Prosocial subscale mean significantly decreasing. However, with the exception of the CRIES and the SDQ Prosocial subscale means, all scales and subscale means were significantly lower at community follow-up than at residential treatment entry. 3.5. Change Over Time: Older Children The time main effect was significant for all outcomes (all p's < the per-comparison error rate of .00192) except for the QOL Home subscale (p = .33) [see means (Standard Errors) in Table 3]. Post hoc testing indicated that change from residential treatment entry to residential treatment completion was significant for all outcomes, with reductions in means for all scales except that the SDQ Prosocial mean scores significantly increased during that period. As with the results for the children, these changes were generally quite large, with 8 mean differences exceeding 50% of the baseline mean and the smallest percentage change in means, for the QOL Friends subscale, of 24%. There were no significant changes from residential treatment completion to community follow-up, with the exception of the SDQ Prosocial subscale, for which the mean score was significantly lower at community follow-up. Moreover, all community follow-up means were significantly lower than their corresponding residential treatment entry means.

nt changes from residential treatment completion to community follow-up, with the exception of the SDQ Prosocial subscale, for which the mean score was significantly lower at community follow-up. Moreover, all community follow-up means were significantly lower than their corresponding residential treatment entry means. 3.6. Success Stories In order to provide greater context to the impact of the intervention, we asked 3 interventionist staff to provide brief summaries of their contact with a child, including information about the presenting problem, the impact of the CHILD intervention, and the outcome of treatment (The name of each child given below is a pseudonym, to protect the identity of the child.)

f the intervention, we asked 3 interventionist staff to provide brief summaries of their contact with a child, including information about the presenting problem, the impact of the CHILD intervention, and the outcome of treatment (The name of each child given below is a pseudonym, to protect the identity of the child.) 3.6.1. Ahmad Ahmad was an innocent 14-year-old boy, born into a poor family. His father was dependent on drugs and Ahmad was always dreaming of life without financial problems like other kids. Every night he was dreaming for a better tomorrow, which never came. Ahmad wishes if his father gets treatment but unfortunately his father was not ready for the treatment. Ahmad's mother requests him to work and contribute to the family. Ahmad searched for a job but could not get one. Then he joined a group of peers who were smoking cigarettes and using hashish, opium, heroin, and alcohol for the last two years. Ahmad also used these substances. To find money to fund his substance use and support his family, Ahmad began commercial sex work. However, he felt ashamed and depressed for what he was doing, yet he had no other opportunities for income. Thus, Ahmad was open to a helping hand to move him out of his situation. Our project outreach team found Ahmad and they talked with him about the children substance use treatment facilities in Kabul, which he indicated he was willing to enter. He received treatment in a residential facility where he was provided with the CHILD intervention. He very much liked the child modules of the CHILD interventions such as art therapy, personal safety, and drug education. At the conclusion of his treatment he stopped using psychoactive substances and he returned to his family and is more hopeful for future.

facility where he was provided with the CHILD intervention. He very much liked the child modules of the CHILD interventions such as art therapy, personal safety, and drug education. At the conclusion of his treatment he stopped using psychoactive substances and he returned to his family and is more hopeful for future. 3.6.2. Aziz Aziz was approximately 9 years old and was unaware that he would face problems due to drug use disorders. Aziz was the only son in his family and his parents were looking for and expecting a bright future for him. In fact, they placed all their hopes and dreams in him. But poverty had other ideas, not allowing Aziz's parents to see his bright future comes to pass and destroying their own future ambitions. Aziz was not able to tolerate his family problems and economic fortunes. He had five sisters, who seemed satisfied with what their father was able to bring them to the family. They seemed to totally overlook their future hopes, ambitions, and wishes. On the other hand, Aziz was neither satisfied nor pleased with the hardship and poverty of the family. He never valued his family life and was always thinking of ways to earn money and live a better life. When he turned seven, he left home and lived with his peers in the streets. Later he started working in bus stations as a conductor. Gradually he realized his hopes and dreams were gone.

ship and poverty of the family. He never valued his family life and was always thinking of ways to earn money and live a better life. When he turned seven, he left home and lived with his peers in the streets. Later he started working in bus stations as a conductor. Gradually he realized his hopes and dreams were gone. He joined another group of youngsters who were pickpockets. During this time, he was exposed to psychoactive such as heroin, hashish, and wine. He spent two years with this group using various illicit substances. Finally, the outreach team found him on the street and convinced him to stop using. He participated in regular meetings before going to outpatient services, where he received psychosocial counseling, opportunities for hygiene care, and social services for about one month. The outreach team contacted his family several times to get their consent for Aziz to participate in a residential drug treatment program. Finally both the family and Aziz gave consent, and he was placed in a residential program. He spent 45 days in the residential treatment program receiving psychosocial support through implementation of the CHILD protocols. The outreach team visited him regularly to provide support and prepare him for the transition back to the community.

mily and Aziz gave consent, and he was placed in a residential program. He spent 45 days in the residential treatment program receiving psychosocial support through implementation of the CHILD protocols. The outreach team visited him regularly to provide support and prepare him for the transition back to the community. After completing his residential program in residential setting, he is a totally substance-free boy. Both the outpatient and treatment programs helped him to reconnect with his family. Aziz and his family appreciated the importance of the CHILD program to Aziz and the family. Aziz is currently maintaining his substance-free status, he is happy, he continues to work, and he helps to support his family. Success for Aziz means having not only a substance-free but also a productive life.

with his family. Aziz and his family appreciated the importance of the CHILD program to Aziz and the family. Aziz is currently maintaining his substance-free status, he is happy, he continues to work, and he helps to support his family. Success for Aziz means having not only a substance-free but also a productive life. 3.6.3. Farid One day our outreach team saw a weak, skinny 13-year-old boy among other adults and children in a “drug hotspot.” He had heroin over cigarette paper and a lighter under the heroin paper. The outreach team arrived on the scene and found that he was sweaty, senseless, and laying down on the ground and the team feared that he might be on the verge of death. The outreach team transported him to an outpatient center. When he awoke he was very hungry, and he ate, took a bath, and changed into clean clothes, and he looked much better. He then started crying and weeping, indicating that this was the first time anyone had showed him such love and empathy. During his outpatient visit our staff found that he had no home and was living under a bridge during very cold winter. After his basic needs were met, he was referred to a treatment center. On intake, Farid indicated he was living in Kabul. He reported that his father said at 6 months of age his mother died, and so his aunt cared for him for three years. His father had remarried after one year of Farid's mother's death. Due to a weak economy condition his aunt and her family immigrated to Iran and returned Farid to his father and stepmother. Farid indicated he wanted to go to school but his stepmother forbade this. She would “always” beat, abuse, and insult him. She would give him a bucket to sell water. When Farid gave her “much” money she was happy; otherwise she beat him. He saw other children that they were laughing and playing and wearing new clothes. He became hopeless and wished his mother was alive to love and care for him. He had a stepbrother, and his stepmother loved him and bought him new clothes. Farid said he loved him too, but his stepmother would not let him play with his stepbrother. One day his stepmother told his father either she or Farid can live in this house. Farid's father cast him out of the house. Farid awoke crying at which point he met a man and he asked Farid to work in the man's hotel for three meals a day and spending money if he needed it. Farid washed dishes and cleaned tables and reported he was very happy with free time at night to go outside.

ive in this house. Farid's father cast him out of the house. Farid awoke crying at which point he met a man and he asked Farid to work in the man's hotel for three meals a day and spending money if he needed it. Farid washed dishes and cleaned tables and reported he was very happy with free time at night to go outside. Farid reported that one day he met three boys with red eyes who gave him a paste that was “smooth and with black color.” Farid used it and liked it. Some time later he met the boys who told him he would now need to pay for his opium. He would steal money to pay for his opium but was caught by the owner who dismissed him. Farid's life “went to darkness,” and he found himself living in dirty places and under bridges. After completion of CHILD treatment at children inpatient treatment facility, Farid was reunited with his father. Farid's father told Farid that he is happy that he found his son after such a long time. Farid was able to go home and play with his brother and to attend school. 4. Discussion Findings strongly suggest that use of the assessment measures with children in Afghanistan yielded reliable scores on the constructs being measured. This inference is supported by the generally excellent internal consistency of the scales and subscales, suggesting the measures were highly reliable, and the consistency in mean changes in scores, suggesting the measures were sensitive to change in the participants.

ed reliable scores on the constructs being measured. This inference is supported by the generally excellent internal consistency of the scales and subscales, suggesting the measures were highly reliable, and the consistency in mean changes in scores, suggesting the measures were sensitive to change in the participants. Results indicate that both younger and older children entering treatment had widespread psychological and social problems that could be considered serious. Scores on the Afghan Symptom Checklist suggest serious symptomatology relative to previous research with Afghan students 11–16 years of age, whose mean score on the ASCL was 1 standard deviation lower than the mean score of the present sample of older children at residential treatment entry [12]. The mean on the SRQ-20 at residential treatment entry among male older children, 13.6 out of a possible 20 maximum, exceeds cutoffs of 7–10 that have been found to be indicative of psychiatric comorbidity in samples from other countries. Moreover, this mean is more than 2 standard deviations higher than the mean reported by Panter-Brick et al. [12] for their sample of Afghan students 11–16 years of age. Finally, QOL scores, transforming each subscale and the total scale score into scale means rather than scale sums (as found in Table 3), would equate to “sometimes” (3 out of 5 on the Likert response scale) in terms of average problem ratings, strongly suggesting a pervasive negative quality of life.

1–16 years of age. Finally, QOL scores, transforming each subscale and the total scale score into scale means rather than scale sums (as found in Table 3), would equate to “sometimes” (3 out of 5 on the Likert response scale) in terms of average problem ratings, strongly suggesting a pervasive negative quality of life. Third, our findings suggest that CHILD had a positive impact on those children who were assessed at posttreatment, and, for the smaller subsamples of children assessed at follow-up, impact of treatment endured from posttreatment to follow-up assessment. The CHILD intervention can be viewed as having had a strong impact during the course of residential treatment. However, results for the CRIES suggest that, given the severity of trauma in Afghan society, 45-day residential treatment for younger children may not provide sufficient time to produce improvements in child posttraumatic stress symptoms. As such, it may be important that trauma coping skills for children be continued in the outpatient and aftercare phases following 45-day residential treatment. Nonetheless, it cannot be discounted that the positive impact of CHILD could be due to a more general positive response to treatment impact rather than the specific impact of the CHILD intervention. This positive response may be a result of being given the opportunity to live without exposure to psychoactive substances and reside in a place that is physically and emotionally safe. And, despite the small sample sizes at follow-up, there is a clear and consistent suggestion of enduring impact, at least for some children. There were several limitations to the present study. The major limitation was that the CHILD treatment was not administered as part of a research project. Rather, the CHILD protocol was designed to meet the treatment needs of children, based on a needs' assessment. An assessment component was built into the CHILD protocol to guide treatment planning. The analysis of these assessment measures shed light on the treatment needs of children in Afghanistan and the promise of the CHILD protocol at addressing their needs for treatment of substance use. A corollary limitation is that there are no data describing the children or their families, due to the fact that recording such data was deemed not cost-effective to the project.

he treatment needs of children in Afghanistan and the promise of the CHILD protocol at addressing their needs for treatment of substance use. A corollary limitation is that there are no data describing the children or their families, due to the fact that recording such data was deemed not cost-effective to the project. A related corollary is that the project was necessarily a single-group study that focused on change within a treated group, and there is neither a control group nor, given the instability and extreme poverty in the country, the ability to identify any comparison group. However, it should be noted that the CHILD treatment protocol was developed because treatment programs already in place in Afghanistan for children had largely been unsuccessful in bringing about change. 5. Conclusions Findings suggest that the CHILD intervention shows promise of producing significant change in deprived and traumatized children who are at risk for or are in need of treatment for psychoactive substance use. Results indicate that the CHILD intervention was both broadly impactful and perhaps enduring. A systematic, longer-term evaluation of the CHILD intervention compared to residential usual care is needed.

in deprived and traumatized children who are at risk for or are in need of treatment for psychoactive substance use. Results indicate that the CHILD intervention was both broadly impactful and perhaps enduring. A systematic, longer-term evaluation of the CHILD intervention compared to residential usual care is needed. Acknowledgments The authors thank the following organizations for their support: The United Nations Office on Drug Control for supporting the development of the protocols and establishing treatment services for children; the Colombo Plan for supporting treatment services to 97 treatment programs in Afghanistan, including women and children's programs, and for coordinating the Afghan National Drug Use Survey; and the US Department of State's Bureau of International Narcotics and Law Enforcement Affairs (INL) for its substantial contribution to addressing substance use in Afghanistan. Conflicts of Interest The authors declare that there are no conflicts of interest regarding the publication of this paper. Table 1 Example schedule of intervention components that younger children receive while in residential inpatient treatment for 45 days.

Acknowledgments The authors thank the following organizations for their support: The United Nations Office on Drug Control for supporting the development of the protocols and establishing treatment services for children; the Colombo Plan for supporting treatment services to 97 treatment programs in Afghanistan, including women and children's programs, and for coordinating the Afghan National Drug Use Survey; and the US Department of State's Bureau of International Narcotics and Law Enforcement Affairs (INL) for its substantial contribution to addressing substance use in Afghanistan. Conflicts of Interest The authors declare that there are no conflicts of interest regarding the publication of this paper. Table 1 Example schedule of intervention components that younger children receive while in residential inpatient treatment for 45 days. Week Sunday Monday Tuesday Wednesday Thursday 1 (i) Appropriate basic education (ii) Trauma coping skills (i) Nutrition (ii) Drug education (i) Appropriate basic education (ii) Structured art therapy (i) Hygiene (ii) Personal safety (i) Appropriate basic education (ii) Communication skills 2 (i) Appropriate basic education (ii) Trauma coping skills (i) Nutrition (ii) Drug education (i) Appropriate basic education (ii) Structured art therapy (i) Hygiene (ii) Personal safety (i) Appropriate basic education (ii) Communication skills 3 (i) Appropriate basic education (ii) Trauma coping skills (i) Nutrition (ii) Drug education (i) Appropriate basic education (ii) Structured art therapy (i) Hygiene (ii) Personal safety (i) Appropriate basic education (ii) Communication skills 4 (i) Appropriate basic education (ii) Trauma coping skills (i) Nutrition (ii) Drug education (i) Appropriate basic education (ii) Structured art therapy (i) Hygiene (ii) Personal safety (i) Appropriate basic education (ii) Communication skills 5 (i) Appropriate basic education (ii) Trauma coping skills (i) Nutrition (ii) Drug education (i) Appropriate basic education (ii) Structured art therapy (i) Hygiene (ii) Personal safety (i) Appropriate basic education (ii) Communication skills 6 (i) Appropriate basic education (ii) Trauma coping skills (i) Nutrition (ii) Drug education (i) Appropriate basic education (ii) Structured art therapy (i) Hygiene (ii) Personal safety (i) Appropriate basic education (ii) Communication skills

) Personal safety (i) Appropriate basic education (ii) Communication skills 6 (i) Appropriate basic education (ii) Trauma coping skills (i) Nutrition (ii) Drug education (i) Appropriate basic education (ii) Structured art therapy (i) Hygiene (ii) Personal safety (i) Appropriate basic education (ii) Communication skills Notes. The first three days of inpatient stay would involve baseline assessment, introduction to the residential rules and code of conduct, and the like. Older children would cycle through these same components three additional times during their residential stay—some basic material would repeat, as necessary (e.g., education, drug education) while most components would involve additional material. Table 2 Internal consistency reliability at residential treatment entry, estimated marginal means (M), Standard Errors (SE), and sample sizes [n] at outpatient assessment, residential treatment entry, residential treatment completion, and community follow-up for the younger child sample (N = 689). Evaluation measure α Outpatient assessment Residential treatment entry Residential treatment completion Community follow-up M (SE) [n] M (SE) [n] M (SE) [n] M (SE) [n] Child Revised Impact of Events Scale (CRIES) .95 a9.6 (.29) [544] b5.5 (.26) [689] c1.7 (.27) [607] a,d10.5 (.72) [83]

Table 2 Internal consistency reliability at residential treatment entry, estimated marginal means (M), Standard Errors (SE), and sample sizes [n] at outpatient assessment, residential treatment entry, residential treatment completion, and community follow-up for the younger child sample (N = 689). Evaluation measure α Outpatient assessment Residential treatment entry Residential treatment completion Community follow-up M (SE) [n] M (SE) [n] M (SE) [n] M (SE) [n] Child Revised Impact of Events Scale (CRIES) .95 a9.6 (.29) [544] b5.5 (.26) [689] c1.7 (.27) [607] a,d10.5 (.72) [83] Self-Report for Childhood Anxiety Related Emotional Disorders (SCARED) Panic Disorder .94 a14.2 (.28) [544] b15.6 (.25) [686] c2.8 (.26) [604] c4.1 (.71) [82] Generalized Anxiety Disorder .91 a8.6 (.20) [544] b9.6 (.18) [686] c2.0 (.19) [604] d4.4 (.52) [82] Separation Anxiety Disorder .88 a9.2 (.17) [544] a,b9.8 (.15) [686] c3.5 (.16) [604] c4.1 (.43) [82] Social Anxiety Disorder .83 a7.9 (.14) [544] b8.6 (.13) [686] c3.6 (.13) [604] c4.6 (.36) [82] School Avoidance .88 a3.1 (.11) [544] b4.2 (.09) [686] c0.6 (.10) [604] d1.6 (.27) [82] Total SCARED Score .95 a43.1 (.79) [544] b47.9 (.70) [686] c12.5 (.75) [604] c18.3 (2.03) [82]

Self-Report for Childhood Anxiety Related Emotional Disorders (SCARED) Panic Disorder .94 a14.2 (.28) [544] b15.6 (.25) [686] c2.8 (.26) [604] c4.1 (.71) [82] Generalized Anxiety Disorder .91 a8.6 (.20) [544] b9.6 (.18) [686] c2.0 (.19) [604] d4.4 (.52) [82] Separation Anxiety Disorder .88 a9.2 (.17) [544] a,b9.8 (.15) [686] c3.5 (.16) [604] c4.1 (.43) [82] Social Anxiety Disorder .83 a7.9 (.14) [544] b8.6 (.13) [686] c3.6 (.13) [604] c4.6 (.36) [82] School Avoidance .88 a3.1 (.11) [544] b4.2 (.09) [686] c0.6 (.10) [604] d1.6 (.27) [82] Total SCARED Score .95 a43.1 (.79) [544] b47.9 (.70) [686] c12.5 (.75) [604] c18.3 (2.03) [82] Child Strengths and Difficulties Questionnaire (SDQ) Emotional Symptoms .81 a5.4 (.11) [538] a,b5.7 (.10) [685] c1.3 (.11) [609] d2.6 (.29) [83] Conduct Problems .75 a4.9 (.10) [538] b5.3 (.09) [685] c1.5 (.10) [609] d2.9 (.27) [83] Hyperactivity .70 a5.02 (.10) [538] b6.0 (.09) [685] c2.4 (.10) [609] a4.3 (.26) [83] Peer Problems .56 a4.9 (.10) [538] a,b4.9 (.09) [685] c2.5 (.09) [609] c3.2 (.25) [83] Prosocial .74 a4.8 (.12) [538] b3.3 (.11) [685] c6.3 (.12) [609] b3.3 (.31) [83] Total SDQ Score .91 a20.3 (.36) [538] b21.9 (.32) [685] c7.8 (.34) [609] d12.9 (.92) [83]

Child Strengths and Difficulties Questionnaire (SDQ) Emotional Symptoms .81 a5.4 (.11) [538] a,b5.7 (.10) [685] c1.3 (.11) [609] d2.6 (.29) [83] Conduct Problems .75 a4.9 (.10) [538] b5.3 (.09) [685] c1.5 (.10) [609] d2.9 (.27) [83] Hyperactivity .70 a5.02 (.10) [538] b6.0 (.09) [685] c2.4 (.10) [609] a4.3 (.26) [83] Peer Problems .56 a4.9 (.10) [538] a,b4.9 (.09) [685] c2.5 (.09) [609] c3.2 (.25) [83] Prosocial .74 a4.8 (.12) [538] b3.3 (.11) [685] c6.3 (.12) [609] b3.3 (.31) [83] Total SDQ Score .91 a20.3 (.36) [538] b21.9 (.32) [685] c7.8 (.34) [609] d12.9 (.92) [83] Notes. α = Cronbach's measure of internal consistency reliability, which was calculated at residential treatment entry (at the largest sample sizes). Theoretical range of scores—CRIES: 0–24; SCARED Panic Disorder subscale: 0–26; SCARED Generalized Anxiety Disorder subscale: 0–18; SCARED Separation Anxiety subscale: 0–16; SCARED Social Anxiety subscale: 0–14; SCARED School Avoidance subscale: 0–8; SCARED Total scale: 0–82; SDQ subscales: 0–10; SDQ Total scale: 0–50. For all measures except the SDQ Prosocial subscale, higher scores indicate greater problem severity in the area measured (for the SDQ Prosocial scale, higher scores indicate a more prosocial orientation). Sample sizes do not equal 689 because not all children were given all measures at all times. A model that included the Gender X Time interaction effect for both SCARED Panic Disorder and SDQ Hyperactivity failed to reach a solution due to an infinite likelihood, so the model was refit without the interaction term; the means in this table were estimated from this latter model. Means that share the same superscript are not significantly different from each other using the Dunn-Sidak correction to conduct post hoc tests.

y failed to reach a solution due to an infinite likelihood, so the model was refit without the interaction term; the means in this table were estimated from this latter model. Means that share the same superscript are not significantly different from each other using the Dunn-Sidak correction to conduct post hoc tests. Table 3 Internal consistency reliability at residential treatment entry, estimated marginal means (M), Standard Errors (SE), and sample sizes [n] at outpatient assessment, residential treatment entry, residential treatment completion, and community follow-up for the older male sample (N = 84). Evaluation Measure α Outpatient entry Residential treatment entry Residential treatment completion Community follow-up M (SE) [n] M (SE) [n] M (SE) [n] M (SE) [n] Afghan Symptom Checklist (ASCL) .96 a61.4 (1.59) [83] a,b51.6 (1.59) [83] c25.5 (1.63) [79] c27.4 (2.23) [41] Self-Reporting Questionnaire-20 (SRQ-20) .93 a15.9 (.56) [84] a13.6 (.57) [81] c1.8 (.58) [79] c1.8 (.80) [41] Child Strengths and Difficulties Questionnaire (SDQ) Emotional Symptoms .85 a2.81 (.28) [84] b4.5 (.28) [82] c0.6 (.29) [78] c1.3 (.37) [41] Conduct Problems .92 a2.6 (.32) [84] b5.1 (.32) [82] c0.4 (.33) [78] a,c1.1 (.44) [41] Hyperactivity .90 a2.9 (31) [84] b5.3 (.32) [82] c0.5 (.32) [78] c1.2 (.43) [41] Peer Problems .78 a2.6 (.26) [84] b3.8 (.26) [82] a,c2.0 (.26) [78] c1.4 (.33) [41] Prosocial .74 a2.7 (.32) [84] a2.6 (.32) [82] c6.3 (.33) [78] d0.8 (.44) [41] Total SDQ Score .97 a10.9 (1.10) [84] b18.7 (1.11) [82] c3.6 (1.13) [78] c5.3 (1.46) [41]

Child Strengths and Difficulties Questionnaire (SDQ) Emotional Symptoms .85 a2.81 (.28) [84] b4.5 (.28) [82] c0.6 (.29) [78] c1.3 (.37) [41] Conduct Problems .92 a2.6 (.32) [84] b5.1 (.32) [82] c0.4 (.33) [78] a,c1.1 (.44) [41] Hyperactivity .90 a2.9 (31) [84] b5.3 (.32) [82] c0.5 (.32) [78] c1.2 (.43) [41] Peer Problems .78 a2.6 (.26) [84] b3.8 (.26) [82] a,c2.0 (.26) [78] c1.4 (.33) [41] Prosocial .74 a2.7 (.32) [84] a2.6 (.32) [82] c6.3 (.33) [78] d0.8 (.44) [41] Total SDQ Score .97 a10.9 (1.10) [84] b18.7 (1.11) [82] c3.6 (1.13) [78] c5.3 (1.46) [41] Quality of Life scale (QOL) Physical Health .93 a19.2 (.57) [84] a18.4 (.58) [82] c8.3 (.59) [78] c8.6 (.81) [41] Mental Health .84 a15.2 (.39) [84] a13.6 (.39) [82] c9.3 (.40) [78] c8.2 (.55) [41] Friends .82 a14.1 (.38) [84] a13.8 (.39) [82] c10.5 (.40) [78] c9.2 (.55) [41] Home .72 6.9 (.27) [84] 6.7 (.27) [82] 7.3 (.28) [78] 7.1 (.37) [41] Total QOL score .92 a55.4 (1.32) [84] a52.1 (1.34) [82] c35.3 (1.37) [78] c32.9 (1.89) [41]

Quality of Life scale (QOL) Physical Health .93 a19.2 (.57) [84] a18.4 (.58) [82] c8.3 (.59) [78] c8.6 (.81) [41] Mental Health .84 a15.2 (.39) [84] a13.6 (.39) [82] c9.3 (.40) [78] c8.2 (.55) [41] Friends .82 a14.1 (.38) [84] a13.8 (.39) [82] c10.5 (.40) [78] c9.2 (.55) [41] Home .72 6.9 (.27) [84] 6.7 (.27) [82] 7.3 (.28) [78] 7.1 (.37) [41] Total QOL score .92 a55.4 (1.32) [84] a52.1 (1.34) [82] c35.3 (1.37) [78] c32.9 (1.89) [41] Notes. α = Cronbach's measure of internal consistency reliability, which was calculated at residential treatment entry. Theoretical range of scores—ASCL: 22–110; SRQ-20: 0–20; SDQ subscales: 0–10; SDQ Total scale: 0–50; QOL Physical Health subscale: 6–30; QOL Mental Health and Friends subscales: 5–25; QOL Home 4–20; QOL Total score: 20–100. For all measures except the SDQ Prosocial subscale, higher scores indicate greater problem severity in the area measured (for the SDQ Prosocial scale, higher scores indicate a more prosocial orientation). Sample sizes do not equal 84 because not all boys were given all measures at all times. Means that share the same superscript are not significantly different from each other using the Dunn-Sidak correction to conduct post hoc tests. Post hoc testing among the means was not conducted for the QOL Home subscale, as the test of the time effect was not significant for this outcome.

1. Introduction Type 1 diabetes is an autoimmune disease triggered by the destruction of pancreatic beta cells. It is usually preceded by the emergence of autoimmunity, including anti-GAD antibodies (glutamine decarboxylase). Both genetic and environmental factors contribute to the development of the disease [1]. Several studies have demonstrated a link between proinflammatory activity and the presence or relative deficit of some fatty acids [2, 3]. Thus, a low level of omega-3 fatty acids, which occurs frequently in western diets, appears to promote an inflammatory response. In fact, higher intake of omega-3 fatty acids and a high concentration of these fatty acids in the erythrocyte membrane are associated with a lower risk of developing pancreatic beta-autoimmunity [4].

ow level of omega-3 fatty acids, which occurs frequently in western diets, appears to promote an inflammatory response. In fact, higher intake of omega-3 fatty acids and a high concentration of these fatty acids in the erythrocyte membrane are associated with a lower risk of developing pancreatic beta-autoimmunity [4]. It is also known that there is a relationship between inflammatory activity and the presence of type 1 diabetes-associated complications such as retinopathy or diabetic nephropathy. A diet rich in omega-3 fatty acids appears to be beneficial through the promotion of greater insulin sensitivity and improved glucose metabolism [5]. These fatty acids decrease inflammation by altering the transcription of genes involved in the inflammatory response (e.g., NF-kB, INF-γ) and by competing with the binding of omega-6 fatty acids to enzymes involved in the synthesis of proinflammatory eicosanoids [6]. Eicosanoids are biologically active lipid mediators that regulate inflammation [7] and include prostaglandins, prostacyclins, thromboxanes, lipoxins, and leukotrienes [8]. The presence of high levels of prostaglandin E2 appears to promote the loss of beta cells and inhibits beta cell proliferation. Prostaglandin E2 also reduces insulin secretion and insulin sensitivity through its binding to the EP3 receptor [9].

rostaglandins, prostacyclins, thromboxanes, lipoxins, and leukotrienes [8]. The presence of high levels of prostaglandin E2 appears to promote the loss of beta cells and inhibits beta cell proliferation. Prostaglandin E2 also reduces insulin secretion and insulin sensitivity through its binding to the EP3 receptor [9]. α-Linolenic acid is the primary omega-3 fatty acid present in western diets and is found especially in vegetables, namely, in soy [1]. Eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) are found in fat fish that is frequently consumed in the Mediterranean diet. Omega-6 fatty acids promote inflammation. The main acid of this type in the western diet is linoleic acid, which is present mainly in vegetable oils, and arachidonic acid, which is derived from meat. While the clinical follow-up of children and young people with type 1 diabetes involves the periodic evaluation of their lipid profile, namely, total cholesterol, HDL, LDL, and triglycerides, the concentrations of different types of fatty acids have not been investigated. To address this gap in knowledge, the authors aimed to study the differences between the fatty acid profiles in children with inaugural type 1 diabetes, diabetic children (at least 1 year after diagnosis), and healthy children.

While the clinical follow-up of children and young people with type 1 diabetes involves the periodic evaluation of their lipid profile, namely, total cholesterol, HDL, LDL, and triglycerides, the concentrations of different types of fatty acids have not been investigated. To address this gap in knowledge, the authors aimed to study the differences between the fatty acid profiles in children with inaugural type 1 diabetes, diabetic children (at least 1 year after diagnosis), and healthy children. 2. Methods 2.1. Human Sample Between 2012 and 2013, we obtained plasma fatty acids profiles from prepubescent children with inaugural diabetes and with a BMI < 85th percentile at the time of their diagnosis (N = 8). Ages ranged from 4.8 to 9.6 years (average: 8.47). We also included diabetic children who attended the Pediatric Endocrinology Unit of S. João Hospital (N = 34) who were between 2.8 and 10.4 years of age (average: 7.06). Children were excluded if they had a BMI above the 85th percentile, were diagnosed less than six months prior to enrolment, had other associated diseases, or were taking any other medication. Children with a Tanner stage ≥ 2 were also excluded. The control population included children who were scheduled to undergo planned surgical interventions, had no chronic diseases, and were not taking any medication. Children with a BMI > 85th percentile or who were pubescent were also excluded. These children were between 3.5 and 8.6 years of age (average: 6.1) (Table 1).

The control population included children who were scheduled to undergo planned surgical interventions, had no chronic diseases, and were not taking any medication. Children with a BMI > 85th percentile or who were pubescent were also excluded. These children were between 3.5 and 8.6 years of age (average: 6.1) (Table 1). The study was approved by the hospital's ethics committee. Authorization and informed consent were obtained from the parents of all the children participating in the study. 2.2. Chemicals and Reagents Sodium hydroxide and anhydrous sodium sulfate were obtained from Pronalab® (Lisbon, Portugal); boron trifluoride in methanol (14% in methanol) and butylated hydroxytoluene (BHT) (≥99%) were from Sigma-Aldrich® (St. Louis, USA); n-hexane (99%) was from Merck® (Darmstadt, Germany); methanol was from VWR Chemicals Prolabo® (Fontenay-sous-Bois, France); and sodium chloride (99.5%) was from Panreac® (Barcelona, Spain). 2.3. Determination of Plasma Fatty Acids Profiles by Gas Chromatography Coupled to a Flame Ionization Detector To analyze the plasma fatty acids profiles, the boiling point must first be lowered which is achieved by the derivatization of the fatty acids [10]. In this case, sodium methoxide (NaOMe) is used as a catalyst to form fatty acid methyl esters (FAMEs) [11] which have a lower boiling point.

to a Flame Ionization Detector To analyze the plasma fatty acids profiles, the boiling point must first be lowered which is achieved by the derivatization of the fatty acids [10]. In this case, sodium methoxide (NaOMe) is used as a catalyst to form fatty acid methyl esters (FAMEs) [11] which have a lower boiling point. The plasma samples were placed in tubes with 20 μg of internal standard (C13:0), and then 5 mL of NaOMe (0.5 M) was added, and the tube was vigorously shaken. The samples were heated to 100°C for 10 min and cooled for 5 min on ice. After cooling, 5 mL of boron trifluoride-methanol was added to the samples, which were again heated to 100°C for 30 min and cooled for 5 min on ice. Then, 600 μL of n-hexane with butylated hydroxytoluene (BHT) (0.02%) was added to prevent lipid oxidation. The tubes were vortexed, and 2 mL of sodium chloride was added; then, the tubes were centrifuged for 10 min at 2200 rpm. The top layer was retrieved and dried with anhydrous sodium sulfate. Then, 100 μL was removed and evaporated to dryness with nitrogen and finally rediluted in 75 μL of n-hexane [12].

lipid oxidation. The tubes were vortexed, and 2 mL of sodium chloride was added; then, the tubes were centrifuged for 10 min at 2200 rpm. The top layer was retrieved and dried with anhydrous sodium sulfate. Then, 100 μL was removed and evaporated to dryness with nitrogen and finally rediluted in 75 μL of n-hexane [12]. Gas chromatography analyses were performed using a Shimadzu GC-2010 equipped with a flame ionization detector and a Shimadzu AOC-20i autoinjector. The separation of FAMEs was carried out on an Agilent® J&W Cp-Sil 88 capillary column (50 m × 0.25 mm ID, 0.20 μm) from Santa Clara, USA. The operating conditions were as follows: the split–splitless injector was used in split mode with a split ratio of 1 : 50. The injection volume of the sample was 1.5 μL. The injector and detector temperatures were kept at 250°C and 270°C, respectively. The temperature program was as follows: initial temperature 120°C for 5 min, which was increased at 3°C/min to 220°C and held at this level for 10 min (total run time: 48 min); carrier gas: He, 30 mL/min; detector gas flows: H2, 40 mL/min; air, 400 mL/min. Data acquisition and processing were performed with Shimadzu software for GC systems.

m was as follows: initial temperature 120°C for 5 min, which was increased at 3°C/min to 220°C and held at this level for 10 min (total run time: 48 min); carrier gas: He, 30 mL/min; detector gas flows: H2, 40 mL/min; air, 400 mL/min. Data acquisition and processing were performed with Shimadzu software for GC systems. 2.4. Statistical Analysis Statistical analyses were performed with the Statistical Package for the Social Sciences (SPSS statistical software, version 21.0, IBM Corp.®, USA). The acquired data (clinical, biological, and fatty acids profiles) were divided into two different pairs: diabetic/not diabetic and inaugural diabetic/not inaugural diabetic. Clinical and biological data were reported as the mean and standard deviation (SD), as well as median and respective interquartile range (IQR), and the lipid profile data for these groups were reported as the median and respective interquartile range (IQR). The Mann–Whitney test was used to compare the median of the fatty acid percentage between the groups. All probabilities with P values < 0.05 were regarded as significant. 3. Results 3.1. Plasmatic Fatty Acids Profile Inaugural Diabetes versus Noninaugural Diabetes. Significant differences between these two groups of children were observed. There was a significant difference in the presence of saturated fatty acids, long-chain saturated fatty acids, palmitic acid, and palmitoleic acid, with higher levels at the time of diagnosis. Moreover, we observed higher levels of omega-6 fatty acids in diabetic children (Table 2).

wo groups of children were observed. There was a significant difference in the presence of saturated fatty acids, long-chain saturated fatty acids, palmitic acid, and palmitoleic acid, with higher levels at the time of diagnosis. Moreover, we observed higher levels of omega-6 fatty acids in diabetic children (Table 2). Inaugural Diabetes versus Controls. Children with a recent diagnosis of diabetes had higher levels of omega-3 fatty acids, alpha-linolenic acid, and DHA. Healthy children from the control population had a higher ratio of LA/ALA (linoleic acid/alpha-linolenic acid). Noninaugural Diabetes versus Controls. The EPA/AA ratio (eicosapentaenoic acid/arachidonic acid) and the amounts of saturated fatty acids, omega-9 fatty acids, cis-monounsaturated fatty acids, saturated long-chain fatty acids, palmitic acid, and palmitoleic acid were higher in the control population. The diabetic children had higher levels of omega-6 fatty acids, polyunsaturated fatty acids, cis-pentadecenoic acid, heptadecenoic acid, linoleic acid, linolenic acid, and arachidonic acid (Table 3). 4. Discussion Contrary to findings from other studies, the levels of omega-3 fatty acids were higher in the recent diabetes group relative to controls. However, other studies have found no relationship between the presence of these fatty acids and diabetes [13]. The amount of omega-6 fatty acids was actually higher in the control population. The content of fatty acids in the control population, which was made up of healthy children with a normal BMI, raises questions about the diet of these children.

no relationship between the presence of these fatty acids and diabetes [13]. The amount of omega-6 fatty acids was actually higher in the control population. The content of fatty acids in the control population, which was made up of healthy children with a normal BMI, raises questions about the diet of these children. We found that omega-6 levels were higher in the nonrecent diabetes group compared to the recent diabetes and to the control groups. In addition, linoleic acid levels were higher in the recent diabetes group. Kurotani et al. have already described an inverse relationship between the levels of C-peptide and this fatty acid [13]. An inhibitory effect of palmitic acid on the production and metabolic action of insulin has been described [14]. Curiously, our results showed higher levels of this fatty acid in the recent diabetes group. Several studies have also shown a relationship between elevated levels of palmitic and palmitoleic acids and low levels of linolenic and linoleic acids with insulin resistance markers [15–17].

An inhibitory effect of palmitic acid on the production and metabolic action of insulin has been described [14]. Curiously, our results showed higher levels of this fatty acid in the recent diabetes group. Several studies have also shown a relationship between elevated levels of palmitic and palmitoleic acids and low levels of linolenic and linoleic acids with insulin resistance markers [15–17]. It is interesting to note that the omega-6/omega-3 ratio did not fall within the recommended values (3/1) in either group (inaugural diabetes: 7.9/1; noninaugural diabetes: 9.2/1; controls: 9.3/1) [18]. The polyunsaturated fatty acids were more abundant in children with established diabetes, even when compared to controls. There was an inverse relationship with respect to saturated fatty acids and long-chain fatty acids, which were lower in the children with diabetes. Although many different fatty acids have statistically significant relationships between these three groups, some seem to detach due to very low levels of alpha-linolenic acid in either the control or the inaugural diabetes group. Furthermore, curiously, EPA and DHA were more abundant in the children with diabetes compared with the control population and the inaugural diabetes group. All of these differences were found in children with identical total triglycerides, though cholesterol levels were higher at the time diabetes was diagnosed. The overall lipid profile was not assessed in the controls (Table 4).

It is interesting to note that the omega-6/omega-3 ratio did not fall within the recommended values (3/1) in either group (inaugural diabetes: 7.9/1; noninaugural diabetes: 9.2/1; controls: 9.3/1) [18]. The polyunsaturated fatty acids were more abundant in children with established diabetes, even when compared to controls. There was an inverse relationship with respect to saturated fatty acids and long-chain fatty acids, which were lower in the children with diabetes. Although many different fatty acids have statistically significant relationships between these three groups, some seem to detach due to very low levels of alpha-linolenic acid in either the control or the inaugural diabetes group. Furthermore, curiously, EPA and DHA were more abundant in the children with diabetes compared with the control population and the inaugural diabetes group. All of these differences were found in children with identical total triglycerides, though cholesterol levels were higher at the time diabetes was diagnosed. The overall lipid profile was not assessed in the controls (Table 4). 5. Conclusion Our results showed that children with diabetes had higher levels of alpha-linolenic acid, EPA, and DHA, as well as mono- and polyunsaturated fatty acids. This observation is attributed to pharmacological therapy and nutritional management of these children. Indeed, knowing the crucial role of food in the treatment and control of type 1 diabetes is part of the follow-up of these patients, which is supposed to review and guide them from a nutritional point of view. Particular emphasis should be placed on the importance of maintaining a healthy diet, with an adequate component of fruits and vegetables as well as diversification of animal sources. Note that the fatty acid profile of diabetic children appears to be healthier than that of children in their inaugural episode as well as healthy children, leading the authors to highlight the importance of early nutritional intervention.

component of fruits and vegetables as well as diversification of animal sources. Note that the fatty acid profile of diabetic children appears to be healthier than that of children in their inaugural episode as well as healthy children, leading the authors to highlight the importance of early nutritional intervention. It is also noted that “healthy” Portuguese children show high levels of saturated fatty acids, as well as an omega-6/omega-3 ratio that is much higher than recommended. Even though we live in a country where, ideally, a so-called Mediterranean diet is practiced, the authors question whether the younger members of our society are actually consuming this type of diet. Abbreviations BMI:Body mass index NF-kB:Nuclear factor-kB INF-γ:Interferon-γ EPA:Eicosapentaenoic acid DHA:Docosahexaenoic acid BHT:Butylated hydroxytoluene LA/ALA:Linoleic acid/alpha-linolenic acid EPA/AA:Eicosapentaenoic acid/arachidonic acid. Conflicts of Interest The authors declare that they have no conflicts of interest. Table 1 Characteristics of participants. Type 1 diabetes New onset diabetes Controls Number 23 6 12 M/F ratio 12/11 3/3 7/5 Age (years) 7.15 (4.0–9.7) 7.36 (3.9–9.7) 6.6 (3.6–10.3) BMI (kg/m2) 17.03 16.94 16.97 Duration of disease (years) 3.29 (0.5–7.2) HbA1c (%) 8.17 (6.3–10.8) 9.9 (8.2–11.9) Table 2 Plasma fatty acids. Plasma FA NOD Diabetes (D) Controls (C) NPT n Mean SD N Mean SD n Mean SD NOD versus D NOD versus C C versus D P P P

Type 1 diabetes New onset diabetes Controls Number 23 6 12 M/F ratio 12/11 3/3 7/5 Age (years) 7.15 (4.0–9.7) 7.36 (3.9–9.7) 6.6 (3.6–10.3) BMI (kg/m2) 17.03 16.94 16.97 Duration of disease (years) 3.29 (0.5–7.2) HbA1c (%) 8.17 (6.3–10.8) 9.9 (8.2–11.9) Table 2 Plasma fatty acids. Plasma FA NOD Diabetes (D) Controls (C) NPT n Mean SD N Mean SD n Mean SD NOD versus D NOD versus C C versus D P P P Sat. FA 8 35,12 3,33 34 31,77 2,87 10 34,45 1,39 0,009 0,897 0,004 Omega-3 8 4,37 1,05 34 4,21 0,98 10 3,76 0,52 0,44 0,043 0,085 Omega-6 8 34,82 4,18 34 39,11 3,85 10 34,97 2,87 0,012 0,829 0,001 LA/ALA 7 220,45 86,35 28 184,82 101,94 9 498,84 285,05 0,199 0,042 0,002 NOD: new onset diabetes; Sat. FA: saturated fatty acids; SD: standard deviation; NPT: nonparametric test. Table 3 Plasma fatty acid profile. Plasma fatty acid profile New onset diabetes Diabetes Controls NPT n Mean SD N Mean SD n Mean SD Nod versus D Nod versus C C versus D P P P Butyric acid 8 1,55 2,31 34 0,78 1,18 10 1,19 0,74 0,199 0,515 0,058 Caproic acid 6 0.03 0.02 29 0.07 0.04 9 0.03 0.01 0.014 0.955 0.004 Lauric acid 7 0.14 0.11 11 0.06 0.02 9 0.10 0.07 0.056 0.408 0.056 Myristic acid 8 1.05 0.55 34 0.67 0.30 10 0.83 0.25 0.037 0.408 0.096 Pentadecanoic acid 8 0.28 0.12 32 0.29 0.26 10 0.29 0.14 0.415 0.762 0.358 Palmitic acid 8 23.82 2.36 34 21.01 1.71 10 22.99 1.48 0.001 0.515 0.002

Butyric acid 8 1,55 2,31 34 0,78 1,18 10 1,19 0,74 0,199 0,515 0,058 Caproic acid 6 0.03 0.02 29 0.07 0.04 9 0.03 0.01 0.014 0.955 0.004 Lauric acid 7 0.14 0.11 11 0.06 0.02 9 0.10 0.07 0.056 0.408 0.056 Myristic acid 8 1.05 0.55 34 0.67 0.30 10 0.83 0.25 0.037 0.408 0.096 Pentadecanoic acid 8 0.28 0.12 32 0.29 0.26 10 0.29 0.14 0.415 0.762 0.358 Palmitic acid 8 23.82 2.36 34 21.01 1.71 10 22.99 1.48 0.001 0.515 0.002 Palmitoleic acid 8 1.71 0.80 34 1.03 0.38 10 1.75 0.44 0.022 0.515 0.0004 Heptadecanoic acid 8 0.29 0.07 34 0.25 0.07 10 0.24 0.05 0.114 0.068 0.710 Stearic acid 8 7.86 0.98 34 8.47 1.56 10 8.62 0.71 0.210 0.083 0.945 Oleic acid 8 22.92 2.52 34 22.35 4.87 10 24.04 1.67 0.320 0.315 0.031 Linoleic acid 8 26.82 3.55 34 30.72 3.67 10 27.87 3.00 0.012 0.573 0.041 γ-Linolenic acid 8 0.54 0.20 34 0.50 0.12 10 0.64 0.32 0.937 0.515 0.066 ALA 7 0.14 0.08 28 0.20 0.09 9 0.08 0.06 0.104 0.042 0.0002 Arachidonic acid 8 7.23 1.45 34 7.70 1.69 10 6.22 0.85 0.304 0.101 0.001 DHA 8 2.12 0.75 34 1.95 0.74 10 1.35 0.33 0.582 0.016 0.002 ALA: α-linolenic acid; NPT: nonparametric test. Table 4 Plasma lipid profile. Plasma lipid profile New onset diabetes Diabetes Nonparametric test N Min. Max. Mean SD n Min. Max. Mean SD NOD versus D Total cholesterol (mg/dl) 5 175.0 215.0 187.0 17.13 31 123.0 224.0 162.26 22.34 0.012 Total triglycerides (mg/dl) 5 50.0 81.0 62.8 12.28 30 31.0 118.0 59.43 20.92 0.477 HDL cholesterol (mg/dl) 5 80.0 150.0 107.8 27.22 30 52.0 135.0 91.7 19.16 0.321 LDL cholesterol (mg/dl) 5 49.0 87.0 66.6 13.59 31 21.0 75.0 58.1 11.09 0.282

1. Background Globally, pneumonia is recognized as a leading cause of under-five mortality. Pneumonia alone is responsible for more deaths than combined deaths from malaria, tuberculosis, and Human Immunodeficiency Virus (HIV) in this category of children. South Asia and Sub-Saharan Africa carry the heaviest global burden of pneumonia estimated at 18% and 90% deaths among children under-five years, respectively [1–3]. To address this problem, the World Health Organization (WHO) formulated guideline of integrated management of child illnesses (IMCI) to direct health workers in assessing and managing pneumonia step by step in children presenting with cough and difficulty in breathing (DIB) at health facilities [4]. Pneumonia is a leading cause of under-five mortality in South Sudan, accounting for 19% of all childhood deaths [5]. Consequently, the IMCI guideline was introduced by Government of South Sudan in 2007 to address the high mortality rates in selected health facilities due to shortage of Human Resources for Health [6]. In Aweil East County (AEC), Northern Bar El Ghazal State (NBEGS), South Sudan, reports indicate pneumonia is incorrectly diagnosed and treated in spite of the presence of the IMCI guideline [7]. Without addressing this practice, it is obvious that most, if not all, children with pneumonia will be incorrectly managed resulting in antibiotic resistance and high mortality rates [3]. In this study, we assessed the level and factors associated with health worker adherence to the IMCI guideline when treating children with cough or DIB in AEC, NBEGS, South Sudan.

that most, if not all, children with pneumonia will be incorrectly managed resulting in antibiotic resistance and high mortality rates [3]. In this study, we assessed the level and factors associated with health worker adherence to the IMCI guideline when treating children with cough or DIB in AEC, NBEGS, South Sudan. 2. Methods This study was conducted at 36 health facilities in AEC, NBEGS, South Sudan. Health services delivery in South Sudan is provided at four different levels (central, state, county, and community levels) each with a different diagnostic capacity and staffing requirements. Health services are further categorized as community, primary health, secondary, and specialized care interlinked with a referral system [8]. Community health workers, maternal and child health workers, and Home Health Promoters provide community healthcare at Primary Healthcare Units (PHCUs) and Primary Healthcare Centers (PHCCs) as entry points. PHCUs provide the first level of interaction between the community and the formal health system in providing basic preventive, promotive, and curative care to about 15,000 people. PHCCs provide diagnostic/laboratory services and maternity care to an estimated 50,000 people alongside services provided at PHCUs. County and State Hospital levels provide secondary, comprehensive-obstetric, inpatient, and surgical care to an estimated 300,000 and 500,000 people, respectively [8].

about 15,000 people. PHCCs provide diagnostic/laboratory services and maternity care to an estimated 50,000 people alongside services provided at PHCUs. County and State Hospital levels provide secondary, comprehensive-obstetric, inpatient, and surgical care to an estimated 300,000 and 500,000 people, respectively [8]. This study used an analytical cross-sectional design to assess and determine health workers' adherence to the IMCI guideline when treating children under 5 years with cough or DIB. The 36 health facilities (one State Hospital, one County Hospital, six PHCCs, and 28 PHCUs) were purposively selected but respondents were consecutively selected health workers that treated children aged 2–59 months with cough or DIB at the outpatient and pediatric units. Within 95% confidence limits, 5% precision, 50% estimated adherence to the IMCI guideline, and 360 health workers previously trained in the IMCI guideline in AEC, 232 health workers based on Krejcie and Morgan (1970) formula [9, 10] were required in this study. Data was collected by trained and supervised Research Assistants between November 2016 and January 2017 using a standardized pretested questionnaire and observation checklist adopted from the WHO IMCI flowchart for managing children with cough or DIB. Nine questions on assessing, documenting presenting complaints, performing physical inspection, and prescribing appropriate treatment to measure health worker adherence to the IMCI guideline were developed from the WHO IMCI flowchart:Asking about the child's ability to drink or breast-feed

g children with cough or DIB. Nine questions on assessing, documenting presenting complaints, performing physical inspection, and prescribing appropriate treatment to measure health worker adherence to the IMCI guideline were developed from the WHO IMCI flowchart:Asking about the child's ability to drink or breast-feed Asking about the child's current or past vomiting history Asking for history of convulsions Checking the child for lethargy or unconsciousness Asking about cough and its duration Assessing the child's breathing patterns by taking off the child's clothes and looking at the chest movements Assessing the child's respiratory rate by counting the number of breaths/minute Correctly classifying the cough or DIB Prescribed antibiotics as per the IMCI guideline. The Research Assistants were health workers with at least 5 years of experience in the IMCI clinical practice.

Assessing the child's breathing patterns by taking off the child's clothes and looking at the chest movements Assessing the child's respiratory rate by counting the number of breaths/minute Correctly classifying the cough or DIB Prescribed antibiotics as per the IMCI guideline. The Research Assistants were health workers with at least 5 years of experience in the IMCI clinical practice. The outcome variable was adherence to the IMCI guideline when treating children aged 2–59 months with cough or DIB measured on a dichotomous scale (“no” or yes). To be adherent, a health worker had to correctly follow all the nine steps in the IMCI flowchart. Failure to adhere to any one of the nine steps constituted nonadherence. The independent variables were health facility staffing levels, availability of the IMCI guideline, presence of essential IMCI drugs and supplies, staff workload, frequency of IMCI support supervisions, importance attached to the IMCI guideline by a health workers, perceived work load, previous training in the IMCI guideline, attitudes of health workers towards the IMCI guideline, perceived benefits of the IMCI guideline on patient waiting time, cadre and qualifications of health workers, years of service, and perceived motivation in using the IMCI guideline.

by a health workers, perceived work load, previous training in the IMCI guideline, attitudes of health workers towards the IMCI guideline, perceived benefits of the IMCI guideline on patient waiting time, cadre and qualifications of health workers, years of service, and perceived motivation in using the IMCI guideline. Data was double entered in EpiData version 3.1 (EpiData Association, Odense, Denmark) [11] and exported to STATA version 12 (StataCorp, College Station, TX, USA) for univariate, bivariate, and multivariate analysis. Univariate analysis involved the computation of descriptive statistics of means and standard deviations (SD) for numerical variables and medians with interquartile ranges (IQR) for categorical variables. In bivariate analysis, the association between categorical independent variables and adherence to IMCI guideline was examined with the Chi-squared test for larger cell counts or Fisher's exact for smaller cell counts, and Student's t-test was used when the independent variable was numerical. Test of collinearity between variables was performed and variables with variance inflation factor greater than ten (VIF > 10) were excluded from the final model. However, none of the variables were collinear. In the final stepwise backward multivariate logistic regression model, significant variables at bivariate analysis and IMCI technical support supervision as a clinically relevant variable were included and the output was stated in odds ratios, 95% Confidence Intervals, and P values. Level of statistical significance was set at 5% all through.

ckward multivariate logistic regression model, significant variables at bivariate analysis and IMCI technical support supervision as a clinically relevant variable were included and the output was stated in odds ratios, 95% Confidence Intervals, and P values. Level of statistical significance was set at 5% all through. This study was approved by the Institutional Review Board of International Health Sciences University and AEC Health Department. In addition, approval was obtained from respective Health Facility Heads. Study participants likewise gave informed written consent after explaining the purpose, benefits, and risks involved. They were free to withdraw their participation in the study at any time. 3. Results 3.1. Sociodemographic Characteristics of Participants The mean age of the respondents was 32.41 ± 7.0 years (median, 32 years; IQR: 28–37), 154 (66.4%) were males, 71 (30.6%) were Anglicans, 104 (44.8%) ended their education at secondary level, 190 (82.0%) had certificates in medical training, and 124 (53.5%) were community health workers (CHWs) (Table 1).

of Participants The mean age of the respondents was 32.41 ± 7.0 years (median, 32 years; IQR: 28–37), 154 (66.4%) were males, 71 (30.6%) were Anglicans, 104 (44.8%) ended their education at secondary level, 190 (82.0%) had certificates in medical training, and 124 (53.5%) were community health workers (CHWs) (Table 1). 3.2. Adherence to the IMCI Guideline Of 232 participants, 193 (83.2%) had asked the mother about the child's ability to drink or breast-feed, 137 (59.1%) had asked about vomiting/vomited everything, 139 (59.9%) had asked about convulsions, 122 (52.6%) had asked about cough and its duration, 153 (66.0%) assessed or looked for lethargy or level of consciousness, 144 (62.1%) took off the child's clothes and observed for abnormal breathing patterns, 28 (18.7%) counted and recorded the child's respiratory rate, 151 (65.1%) classified the presenting complaint correctly, and 186 (80.2%) gave the correct antibiotics. Overall, 23 (9.9%: 95% CI: 6.4–14.5) respondents adhered to all the steps of the IMCI guideline when treating children with cough or DIB (Table 2). 3.3. Bivariate Analysis of Factors Associated with Adherence to IMCI Guideline Table 3 shows 18 (14.0%) respondents aged 32 years or less compared to five (4.8%) above 32 years that were adherent (P = 0.02). Eight (4.2%) respondents that had certificates, 13 (35.1%) that had diploma, and two (40.0%) that had bachelor's degree were adherent (P < 0.001).

3.3. Bivariate Analysis of Factors Associated with Adherence to IMCI Guideline Table 3 shows 18 (14.0%) respondents aged 32 years or less compared to five (4.8%) above 32 years that were adherent (P = 0.02). Eight (4.2%) respondents that had certificates, 13 (35.1%) that had diploma, and two (40.0%) that had bachelor's degree were adherent (P < 0.001). Two (2.0%) respondents that stated it took a long time to follow the steps of the IMCI guideline in contrast to 21 (15.7%) that reported it took a short time were adherent (P = 0.009). 14 (7.8%) respondents that had never received technical support supervision in the IMCI guideline compared to 9 (17.0%) that had ever received were adherent (P = 0.05). Four (3.8%) respondents that reported the number of staffs in the health facility was inadequate to treat children compared to 19 (15.1%) that stated it was adequate were adherent (P = 0.004). 11 (6.5%) respondents that reported recent shortage of IMCI drugs in the preceding month compared to 12 (19.0%) that reported no shortage were adherent (P = 0.004). Two (2.0%) respondents that had difficulty in using the IMCI guideline and 21 (16.1%) that had no difficulty in using the guideline were adherent (P < 0.001).

(6.5%) respondents that reported recent shortage of IMCI drugs in the preceding month compared to 12 (19.0%) that reported no shortage were adherent (P = 0.004). Two (2.0%) respondents that had difficulty in using the IMCI guideline and 21 (16.1%) that had no difficulty in using the guideline were adherent (P < 0.001). 3.4. Univariable Analysis of Factors Associated with Adherence to IMCI Guideline In unadjusted analysis (Table 3), respondents aged more than 32 years compared to less than or equal to 32 years had reduced adherence (unadjusted odds ratio (UOR) = 0.31; 95% CI: 0.11–0.88, P = 0.027). Conversely, adherence increased with holding a diploma (UOR = 12.32; 95% CI: 4.63–32.77, P < 0.001) or degree (UOR = 15.17; 95% CI: 2.21–103.89, P = 0.006), short time taken to follow steps of the guideline (UOR = 8.92; 95% CI: 2.04–39.02; P = 0.004), ever receiving IMCI technical support supervision (UOR = 2.41; 95% CI: 0.98–5.94, P = 0.056), adequate staffing at health facility (UOR = 4.53; 95% CI: 1.49–13.76; P = 0.008), no recent shortage of IMCI drugs (UOR = 3.38; 95% CI: 1.41–8.12; P = 0.006), and no difficulties in using the guideline (UOR = 9.63; 95% CI: 2.20–42.13; P = 0.003).

technical support supervision (UOR = 2.41; 95% CI: 0.98–5.94, P = 0.056), adequate staffing at health facility (UOR = 4.53; 95% CI: 1.49–13.76; P = 0.008), no recent shortage of IMCI drugs (UOR = 3.38; 95% CI: 1.41–8.12; P = 0.006), and no difficulties in using the guideline (UOR = 9.63; 95% CI: 2.20–42.13; P = 0.003). 3.5. Multivariate Analysis of Factors Associated with Adherence to IMCI Guideline After adjusting for health worker academic qualifications, age, IMCI support supervision, staffing, time taken to follow the steps of the guideline, difficulties in using the guideline and availability of IMCI drugs (Table 3), holding a diploma (AOR = 6.97; 95% CI: 1.82–26.67; P = 0.005) compared to certificate, a short time taken to follow the steps of the IMCI guideline (AOR = 12.0; 95% CI: 2.73–61.66; P < 0.001), no recent shortage of IMCI drugs (AOR = 7.11; 95% CI: 2.37–21.38; P < 0.001), and absence of difficulties in using the guideline (AOR = 27.7; 95% CI: 5.40–142.25; P < 0.001) were associated with increased adherence. However, maternal age above 32 years (AOR = 0.63; 95% CI: 0.19–2.15; P = 0.463), holding a degree (AOR = 1.02; 95% CI: 0.10–10.43; P = 0.986), ever receiving the IMCI technical support supervision (AOR = 0.94; 95% CI: 0.27–3.24; P = 0.921), and adequate staffing at the health facility (AOR = 2.96; 95% CI: 0.85–10.32; P = 0.089) were not associated with adherence.

.63; 95% CI: 0.19–2.15; P = 0.463), holding a degree (AOR = 1.02; 95% CI: 0.10–10.43; P = 0.986), ever receiving the IMCI technical support supervision (AOR = 0.94; 95% CI: 0.27–3.24; P = 0.921), and adequate staffing at the health facility (AOR = 2.96; 95% CI: 0.85–10.32; P = 0.089) were not associated with adherence. 4. Discussion This study assessed the level and factors associated with health worker adherence to the IMCI guideline when treating children with cough or DIB in AEC, NBEGS, South Sudan. The level of adherence was low at 9.9% suggesting most children that presented at the health facilities with cough or DIB were inappropriately managed. In Tanzania [12] and elsewhere [13], low adherence to the IMCI guideline was previously reported [12, 13]. According to the WHO, at least two in every five health workers must adhere to the IMCI guideline for effective management of childhood pneumonia [14]. With low health worker adherence to the IMCI guideline, the potential impact of the IMCI strategy on under-five morbidity and mortality is thus lost [15]. Our results compare unfavorably with those in Benin and Senegal where 64% and 16% of health workers, respectively, were adherent [16].

t of childhood pneumonia [14]. With low health worker adherence to the IMCI guideline, the potential impact of the IMCI strategy on under-five morbidity and mortality is thus lost [15]. Our results compare unfavorably with those in Benin and Senegal where 64% and 16% of health workers, respectively, were adherent [16]. In this study, the major steps in the WHO IMCI flowchart commonly missed by health workers include asking for vomiting, convulsions, cough or DIB, and the duration, assessing for lethargy or level of consciousness and counting the respiratory rate. Similar gaps in following the steps of the IMCI guideline were observed in Benin, Senegal, Tanzania, and Botswana [12, 16, 17]. However, our findings reflect a much lower level of adherence compared to the studies in Benin, Senegal, and Botswana. These gaps should be addressed through onsite mentorships of health workers in using the IMCI guideline followed by technical support supervision. Earlier in Tanzania, incorrect antibiotics were prescribed to children diagnosed with severe pneumonia [12]. Similarly in Botswana, children were prescribed incorrect antibiotics after diagnosis of pneumonia [18]. Our result is therefore not surprising because it generally appears that there is lack of adherence to the IMCI guideline among health workers in Sub-Saharan Africa.

In this study, the major steps in the WHO IMCI flowchart commonly missed by health workers include asking for vomiting, convulsions, cough or DIB, and the duration, assessing for lethargy or level of consciousness and counting the respiratory rate. Similar gaps in following the steps of the IMCI guideline were observed in Benin, Senegal, Tanzania, and Botswana [12, 16, 17]. However, our findings reflect a much lower level of adherence compared to the studies in Benin, Senegal, and Botswana. These gaps should be addressed through onsite mentorships of health workers in using the IMCI guideline followed by technical support supervision. Earlier in Tanzania, incorrect antibiotics were prescribed to children diagnosed with severe pneumonia [12]. Similarly in Botswana, children were prescribed incorrect antibiotics after diagnosis of pneumonia [18]. Our result is therefore not surprising because it generally appears that there is lack of adherence to the IMCI guideline among health workers in Sub-Saharan Africa. This study found diploma and degree compared to certificate holders were more adherent to the IMCI guideline. In South Sudan, the health worker force is dominated by CHWs usually trained for 9 months. Their medical skills and knowledge are therefore unmatched with diploma and degree holders usually trained for minimally 3 years. Our results contradict with previous study in Tanzania that found low adherence to IMCI guidelines among clinicians with better knowledge of IMCI [13]. Additionally, it controverts earlier evidence that well-trained health workers in rural and hard to reach areas hardly stick to guidelines set for treating pneumonia [19]. Regardless of previous evidences, level of training and academic qualifications might determine adherence to national and international guidelines for disease management. A preservice training in the IMCI guideline might thus be critical in the training of CHWs so as to increase adherence [20] particularly in AEC, South Sudan.

ss of previous evidences, level of training and academic qualifications might determine adherence to national and international guidelines for disease management. A preservice training in the IMCI guideline might thus be critical in the training of CHWs so as to increase adherence [20] particularly in AEC, South Sudan. We found adherence to the IMCI guideline increased with no reported recent shortage of IMCI drugs. This was not surprising because availability of IMCI drugs ensures correct prescription as per defined standards. Our result agrees with a study in Kenya [21] that found high adherence to the IMCI guideline at health facilities without shortage of IMCI drugs.

ne increased with no reported recent shortage of IMCI drugs. This was not surprising because availability of IMCI drugs ensures correct prescription as per defined standards. Our result agrees with a study in Kenya [21] that found high adherence to the IMCI guideline at health facilities without shortage of IMCI drugs. In this study, respondents that indicated it takes a short time to follow the steps of the IMCI guideline and that it is not difficult to use the guideline had increased adherence. This result implies that the IMCI guideline is less adhered to by health workers that face difficulty in following its steps resulting in lengthy time requirement. The combination of these two findings suggests that guideline complexity or simplicity is a crucial determinant of adherence. Our result is in conformity with an earlier study that evaluated the performance of CHWs in managing multiple childhood illness at outpatient and inpatient units in Siaya district of Kenya, where mistakes occurred in symptom assessment, disease classification, and prescription of correct doses of medications due to guideline complexity and inadequate supervision [22]. In addition, it is consistent with results of a past study in Senegal that found low adherence to the IMCI guideline on regular basis among Medical Doctors, Nurses, and Midwives after IMCI training due to difficulties encountered in following the required steps [16].

uideline complexity and inadequate supervision [22]. In addition, it is consistent with results of a past study in Senegal that found low adherence to the IMCI guideline on regular basis among Medical Doctors, Nurses, and Midwives after IMCI training due to difficulties encountered in following the required steps [16]. Another supporting evidence of our findings comes from previous study that assessed the safety and effectiveness of modifying the IMCI guideline that initially resulted in most children with severe pneumonia being treated locally at first-level facilities. In the study, the modification of the IMCI guideline increases the number of children that were correctly managed and reduced unnecessary referrals [23]. These research evidences [16, 22–24] confirm that difficulties in using the IMCI guideline are a common phenomenon that probably accounts for low adherence. In enhancing adherence to the IMCI guideline, capacity building of health workers through onsite mentorships, in-service training, and Continuous Professional Education (CPE) is important. In Kenya, CPE sessions and clinical mentorships increased health worker adherence to the IMCI guideline [21]. However in South Sudan, CPE remains a critical challenge of HRH [25], a factor that would have improved HRH capacity in adhering to national and international guidelines.

Another supporting evidence of our findings comes from previous study that assessed the safety and effectiveness of modifying the IMCI guideline that initially resulted in most children with severe pneumonia being treated locally at first-level facilities. In the study, the modification of the IMCI guideline increases the number of children that were correctly managed and reduced unnecessary referrals [23]. These research evidences [16, 22–24] confirm that difficulties in using the IMCI guideline are a common phenomenon that probably accounts for low adherence. In enhancing adherence to the IMCI guideline, capacity building of health workers through onsite mentorships, in-service training, and Continuous Professional Education (CPE) is important. In Kenya, CPE sessions and clinical mentorships increased health worker adherence to the IMCI guideline [21]. However in South Sudan, CPE remains a critical challenge of HRH [25], a factor that would have improved HRH capacity in adhering to national and international guidelines. 5. Study Limitations This study established low adherence to the IMCI guideline in treating children below 5 years with cough or DIB in AEC, NBEGS, South Sudan. Nevertheless, it has a number of drawbacks. First, there is possibility of interobserver and intraobserver variation when observing health workers in following the steps of the IMCI guideline. However, we minimized this pitfall by using health workers that had experience in the IMCI as Research Assistants. Secondly, self-reporting bias is likely especially in providing information on ever receiving technical support supervision, reporting shortage of IMCI drugs, and difficulties in following the steps of the IMCI guideline. The absence of data on the actual number of health workers in the units so as to establish the staffing norm is another limitation that must be considered in the interpretation of our results.

technical support supervision, reporting shortage of IMCI drugs, and difficulties in following the steps of the IMCI guideline. The absence of data on the actual number of health workers in the units so as to establish the staffing norm is another limitation that must be considered in the interpretation of our results. 6. Conclusion and Recommendations Our study is the first in AEC, NBEGS, South Sudan, to indicate low adherence of health workers to the IMCI guideline when treating children under five years of age with cough or DIB. Adherence was high among well-trained health workers that held diploma or degrees, among health workers that took a short time to follow the guideline steps, in health facilities where no shortage of IMCI drugs was reported, and when health workers had no difficulty in using the guideline. In improving health worker adherence to the IMCI guideline in South Sudan, the Government should address shortages of HRH (Human Resources for Health) by recruiting and retaining well-trained high-level professionals, building the capacity of existing health workforce in the IMCI guideline and ensuring a sustainable supply of IMCI drugs. Abbreviations AEC:Aweil East County AIDS:Acquired immunodeficiency syndrome AOR:Adjusted odds ratio CHWs:Community Health Workers CPE:Continuous Professional Education DIB:Difficulty in breathing HIV:Human Immunodeficiency Virus HRH:Human Resources for Health IMCI:Integrated Management of Childhood Illnesses IQR:Interquartile range NBEGS:Northern Bar El Ghazal State PHCC:Primary health care center PHCU:Primary health care unit UOR:Unadjusted odds ratio

CHWs:Community Health Workers CPE:Continuous Professional Education DIB:Difficulty in breathing HIV:Human Immunodeficiency Virus HRH:Human Resources for Health IMCI:Integrated Management of Childhood Illnesses IQR:Interquartile range NBEGS:Northern Bar El Ghazal State PHCC:Primary health care center PHCU:Primary health care unit UOR:Unadjusted odds ratio VIF:Variance inflation factor WHO:World Health Organization. Additional Points Availability of Data and Materials. The data set used in this study shall be provided on request in order to protect respondent anonymity. Consent Informed written consent was obtained from all participants prior to participation in the study. Privacy, confidentiality, and free will to withdraw from the study at any point in time were maintained. Conflicts of Interest The authors declare that there are no conflicts of interest regarding the publication of this paper. Authors' Contributions Jonathan Izudi participated in the study design and statistical analysis and wrote the first and final manuscripts. Stanley Anyigu participated in the study design, ethical approval, and data collection and entry and proof-read all manuscripts. David Ndungutse guided the entire conduct of the research and reviewed study protocol, draft, and final manuscripts. Jonathan Izudi is the first author of this paper although all authors contributed substantially in the writing and revisions. Table 1 Characteristics of respondents. Variable Frequency (number = 232) Percentage (% = 100) Age (years) 19–32 129 55.6 33–55 103 44.4 Age in years (mean ± SD) 232 32.41 ± 7.0 Sex Female 78 48.3 Male 154 66.4 Religion

Authors' Contributions Jonathan Izudi participated in the study design and statistical analysis and wrote the first and final manuscripts. Stanley Anyigu participated in the study design, ethical approval, and data collection and entry and proof-read all manuscripts. David Ndungutse guided the entire conduct of the research and reviewed study protocol, draft, and final manuscripts. Jonathan Izudi is the first author of this paper although all authors contributed substantially in the writing and revisions. Table 1 Characteristics of respondents. Variable Frequency (number = 232) Percentage (% = 100) Age (years) 19–32 129 55.6 33–55 103 44.4 Age in years (mean ± SD) 232 32.41 ± 7.0 Sex Female 78 48.3 Male 154 66.4 Religion Catholic 112 48.3 Moslem 43 18.5 Anglican 71 30.6 Others 6 2.6 Highest level of education Primary 58 25.0 Secondary 104 44.8 Tertiary 70 30.2 Academic qualifications Certificate 190 81.9 Diploma 37 16.0 Undergraduate 5 2.2 Designation Community health worker 124 53.5 Nursing officer 69 29.7 Clinical officer 34 14.7 Medical officer 5 2.2 Table 2 Adherence to IMCI guideline in the treatment of children with cough or difficulty in breathing in Aweil East County, South Sudan.

Academic qualifications Certificate 190 81.9 Diploma 37 16.0 Undergraduate 5 2.2 Designation Community health worker 124 53.5 Nursing officer 69 29.7 Clinical officer 34 14.7 Medical officer 5 2.2 Table 2 Adherence to IMCI guideline in the treatment of children with cough or difficulty in breathing in Aweil East County, South Sudan. Variable Number = 232 Percentage (% = 100) Health worker asked the mother whether the child was able to drink or breast-feed Yes 193 83.2 No 39 16.8 Health worker asked the mother whether the child was vomiting/vomited everything Yes 137 59.1 No 95 40.9 Health worker asked the mother whether the child had had convulsions Yes 139 59.9 No 93 40.1 Health worker asked the mother if the child had had cough and for how long Yes 122 52.6 No 110 47.4 Health worker assessed or looked at the child for lethargy or level of consciousness Yes 153 66.0 No 79 44.0 Health worker took off the cloth of the child and looked at the chest Yes 144 62.1 No 88 37.9 Health worker counted the respiratory rate of the child Yes 28 18.7 No 122 81.3 Health worker classified the pneumonia correctly Yes 151 65.1 No 81 34.9 Health worker gave the correct antibiotics as per the IMCI guideline Yes 186 80.2 No 46 19.8 Adherence to the IMCI guideline No 23 9.9 Yes 209 90.1 Table 3 Unadjusted and adjusted analysis of factors associated with adherence to the IMCI guideline when treating children with cough or difficulty in breathing in Aweil East County, Northern Bahr El Ghazal State, South Sudan.

IMCI guideline Yes 186 80.2 No 46 19.8 Adherence to the IMCI guideline No 23 9.9 Yes 209 90.1 Table 3 Unadjusted and adjusted analysis of factors associated with adherence to the IMCI guideline when treating children with cough or difficulty in breathing in Aweil East County, Northern Bahr El Ghazal State, South Sudan. Variable Adhered to IMCI guide? Univariable analysis Multivariate analysis No (n = 209) Yes (n = 23) UOR (95% CI) P value AOR (95% CI) P value Number (%) Number (%) Age group (years) Less than or equal to 32 111 (86.1) 18 (14.0) 1   1 More than 32 98 (95.2) 5 (4.8) 0.31 (0.11–0.88) 0.027 0.63 (0.19–2.15) 0.463 Academic qualifications Certificate 182 (95.8) 8 (4.2) 1   1 Diploma 24 (64.9) 13 (35.1) 12.32 (4.63–32.77) <0.001 6.97 (1.82–26.67) 0.005 Degree 3 (60.0) 2 (40.0) 15.17 (2.21–103.89) 0.006 1.02 (0.10–10.43) 0.986 Time taken to follow the IMCI steps Long 96 (98.0) 2 (2.0) 1   1 Short 113 (84.3) 21 (15.7) 8.92 (2.04–39.02) 0.004 12 (2.73–61.66) <0.001 IMCI support supervision Never received 165 (92.2) 14 (7.8) 1   1 Ever received 44 (83.0) 9 (17.0) 2.41 (0.98–5.94) 0.056 0.94 (0.27–3.24) 0.921 Inadequate staffing Yes 102 (96.2) 4 (3.8) 1   1 No 107 (84.9) 19 (15.1) 4.53 (1.49–13.76) 0.008 2.96 (0.85–10.32) 0.089 Recent shortage of IMCI drugs at health facility Yes 158 (93.5) 11 (6.5) 1   1 No 51 (81.0) 12 (19.0) 3.38 (1.41–8.12) 0.006 7.11 (2.37–21.38) <0.001 The IMCI guideline is difficult to use Yes 100 (98.0) 2 (2.0) 1   1 No 109 (83.9) 21 (16.1) 9.63 (2.20–42.13) 0.003 27.7 (5.40–142.25) <0.001

IMCI support supervision Never received 165 (92.2) 14 (7.8) 1   1 Ever received 44 (83.0) 9 (17.0) 2.41 (0.98–5.94) 0.056 0.94 (0.27–3.24) 0.921 Inadequate staffing Yes 102 (96.2) 4 (3.8) 1   1 No 107 (84.9) 19 (15.1) 4.53 (1.49–13.76) 0.008 2.96 (0.85–10.32) 0.089 Recent shortage of IMCI drugs at health facility Yes 158 (93.5) 11 (6.5) 1   1 No 51 (81.0) 12 (19.0) 3.38 (1.41–8.12) 0.006 7.11 (2.37–21.38) <0.001 The IMCI guideline is difficult to use Yes 100 (98.0) 2 (2.0) 1   1 No 109 (83.9) 21 (16.1) 9.63 (2.20–42.13) 0.003 27.7 (5.40–142.25) <0.001 Note. Percentages were calculated as n/N for each row; level of statistical significance was set less than 5%; AOR: adjusted odds ratio; UOR: unadjusted odds ratio.

1. Introduction Lower respiratory tract infections such as bronchiolitis and viral pneumonia place tremendous strain on the health of young children and the healthcare system. The propensity of these viral infections to affect the most vulnerable of pediatric populations along with their highly variable clinical course leads to frequent hospital admissions that often occur regardless of severity [1]. The respiratory syncytial virus (RSV) alone affects roughly 800,000 children in the United States leading to approximately 20% of the annual birth cohort requiring medical attention yearly [2]. This results in $500 million of direct hospital costs in the United States alone [3].

often occur regardless of severity [1]. The respiratory syncytial virus (RSV) alone affects roughly 800,000 children in the United States leading to approximately 20% of the annual birth cohort requiring medical attention yearly [2]. This results in $500 million of direct hospital costs in the United States alone [3]. Hypoxemia requiring supplemental oxygen (O2) is a key determinant in the decision to hospitalize infants with bronchiolitis and contributes to increased length of stay (LOS) [4]. Updated clinical practice guidelines released by the American Academy of Pediatrics (AAP) in 2014 identify O2 supplementation and hydration as the mainstay of treatment for bronchiolitis [5]. They establish a blood O2 saturation (spO2) of <90%, measured by pulse oximetry as a threshold for initiating O2 therapy and encourage discontinuing O2 supplementation and spO2 monitoring after improvement [6]. However, the integration of those guidelines can be quite challenging as they require ongoing “deimplementation,” a term describing the practice of discouraging care not supported by evidence-based research [7, 8]. These practices are difficult to discourage and often need a concerted effort to eliminate, even in the face of well-executed collaboration [9]. While there are numerous studies evaluating the impact of deimplementing interventions such as chest X-rays, antibiotics, and corticosteroids [10–16], few studies specifically evaluate the impact of a protocol based O2 supplementation practice [17–19]. Our institution developed and implemented an inpatient pediatric O2 supplementation and pulse oximetry protocol based on AAP guidelines [5] that aligns with the realities of our clinical practice. We hypothesize that the implementation of this protocol and deimplementation of certain practices lead to a significant decrease in hospital LOS for children admitted with bronchiolitis.

c O2 supplementation and pulse oximetry protocol based on AAP guidelines [5] that aligns with the realities of our clinical practice. We hypothesize that the implementation of this protocol and deimplementation of certain practices lead to a significant decrease in hospital LOS for children admitted with bronchiolitis. 2. Methods 2.1. Intervention A multidisciplinary team of clinicians at Helen DeVos Children's Hospital, a tertiary care pediatric hospital, designed and implemented a standardized O2 supplementation and continuous pulse oximetry protocol (Figure 1) in February 2013. Significant changes to prior medical practice included lowering the threshold for O2 supplementation from 94% to 90% and clearly defining both the steps and duration over which O2 supplementation would be titrated and discontinued. This protocol created a formal algorithm that was in clear contrast to a previously highly variable system that was driven by individual care providers. We performed education in the form of online modules, didactics, and formalized multidisciplinary rounds with respiratory therapists, nurses, and providers at the hospital. We also used education to stress and deemphasize some of the practices that were not supported by clinical evidence (Table 1).

dual care providers. We performed education in the form of online modules, didactics, and formalized multidisciplinary rounds with respiratory therapists, nurses, and providers at the hospital. We also used education to stress and deemphasize some of the practices that were not supported by clinical evidence (Table 1). 2.2. Study Sample In order to assess this quality initiative, we performed a retrospective review of the medical charts of infants and children (≤24 months of age) hospitalized with bronchiolitis during the preintervention study period of November 1, 2011, through April 30, 2012 (control group), and during the postintervention study period of November 1, 2013, through April 30, 2014 (O2 protocol group). Exclusion criteria included hospital admission directly to the pediatric intensive care unit (PICU), home O2 used immediately before or after hospitalization, presence of a tracheostomy tube, congenital heart disease, sickle cell disease, severe anemia, hypotonia, cystic fibrosis, and age > 24 months. 2.3. IRB Statement Spectrum Health Institutional Review Board reviewed the study as a quality improvement (QI) project and thus was exempted from full review.

2.2. Study Sample In order to assess this quality initiative, we performed a retrospective review of the medical charts of infants and children (≤24 months of age) hospitalized with bronchiolitis during the preintervention study period of November 1, 2011, through April 30, 2012 (control group), and during the postintervention study period of November 1, 2013, through April 30, 2014 (O2 protocol group). Exclusion criteria included hospital admission directly to the pediatric intensive care unit (PICU), home O2 used immediately before or after hospitalization, presence of a tracheostomy tube, congenital heart disease, sickle cell disease, severe anemia, hypotonia, cystic fibrosis, and age > 24 months. 2.3. IRB Statement Spectrum Health Institutional Review Board reviewed the study as a quality improvement (QI) project and thus was exempted from full review. 2.4. Data Sources The medical records of patients who met the entry criteria with discharge diagnoses of “acute bronchiolitis” were reviewed. Patients were identified using electronic medical record search queries for International Classification of Diseases, Ninth Revision (ICD-9), primary or secondary diagnosis codes 466.1 (acute bronchiolitis), 466.11 (acute bronchiolitis due to respiratory syncytial virus), and 466.19 (acute bronchiolitis due to other infectious organisms).

fied using electronic medical record search queries for International Classification of Diseases, Ninth Revision (ICD-9), primary or secondary diagnosis codes 466.1 (acute bronchiolitis), 466.11 (acute bronchiolitis due to respiratory syncytial virus), and 466.19 (acute bronchiolitis due to other infectious organisms). 2.5. Study Variables Patient age at admission, gender, respiratory distress score (RDS), duration of supplemental O2, LOS, seven-day readmission rate, rate of PICU transfer, and coexisting medical problems were collected. LOS, as noted in hours, was determined as the period from the time of admission to the inpatient unit until the time of discharge. We evaluated bronchiolitis illness severity at the time of the first respiratory therapist evaluation by using a modified RDS tool, an evaluation tool based on respiratory rate, accessory muscle use, wheezing, O2 requirement, and inspiratory to expiratory ratio (Table 2) [20]. RDS values ≥3 were considered to be indicative of moderate to severe bronchiolitis.

ty at the time of the first respiratory therapist evaluation by using a modified RDS tool, an evaluation tool based on respiratory rate, accessory muscle use, wheezing, O2 requirement, and inspiratory to expiratory ratio (Table 2) [20]. RDS values ≥3 were considered to be indicative of moderate to severe bronchiolitis. 2.6. Statistical Methods The data was analyzed using IBM Statistics SPSS v. 21 (Armonk, New York). Quantitative data was compared using a t-test and was reported as the mean ± SD. Nominal data was compared using the χ2 test and Fisher's exact test (when appropriate) and was reported as percentages. Due to the nonnormal distribution of LOS and duration of O2 supplementation, both of these variables were transformed prior to analysis, while the summary statistics shown are for the untransformed data. The LOS was transformed using the natural log, while the duration of O2 supplementation was transformed using the inverse hyperbolic sine. In addition, a multiple regression analysis was performed, using the log transformed LOS as the dependent variable, with patient age, O2 protocol group versus control group, PICU transfer, and RDS score as the independent variables. Significance was assessed at p < 0.05.

s transformed using the inverse hyperbolic sine. In addition, a multiple regression analysis was performed, using the log transformed LOS as the dependent variable, with patient age, O2 protocol group versus control group, PICU transfer, and RDS score as the independent variables. Significance was assessed at p < 0.05. 3. Results From the collective study periods, a total of 263 children met the study criteria: 141 children in the control group and 122 children in the O2 protocol group. Table 3 shows the patient's demographics and characteristics. There were no significant differences based on age, gender, and PICU transfer rates between control and O2 protocol groups (Table 3). Only three children were readmitted within seven days, one in the control group and two in the O2 protocol group. Contrary to our hypothesis, there was not a statistically significant difference in LOS between the control and O2 protocol groups. However, the O2 protocol group had a significantly higher severity of illness at admission based on their initial RDS independent of assessment rates.

3. Results From the collective study periods, a total of 263 children met the study criteria: 141 children in the control group and 122 children in the O2 protocol group. Table 3 shows the patient's demographics and characteristics. There were no significant differences based on age, gender, and PICU transfer rates between control and O2 protocol groups (Table 3). Only three children were readmitted within seven days, one in the control group and two in the O2 protocol group. Contrary to our hypothesis, there was not a statistically significant difference in LOS between the control and O2 protocol groups. However, the O2 protocol group had a significantly higher severity of illness at admission based on their initial RDS independent of assessment rates. Next, we assessed the relationship between LOS and bronchiolitis severity, as defined by the RDS. Patients in the O2 protocol group with mild bronchiolitis (RDS < 3) had a statistically significant shorter LOS compared to control (RDS ≥ 3) (p = 0.005). Interestingly, O2 protocol group subjects with moderate to severe bronchiolitis (RDS ≥ 3) had a 29% increase in LOS, although this was not statistically significant (p = 0.535). Furthermore, O2 protocol group subjects with mild disease had a significantly shorter LOS compared with the O2 protocol patients with moderate to severe bronchiolitis (Table 4). This data suggests that LOS may be dependent on disease severity. O2 protocol group subjects with mild bronchiolitis disproportionately improve compared with moderate to severe disease. Furthermore, the O2 protocol may have a negative affect when bronchiolitis is more severe. However, no direct conclusive relationship between LOS and the protocol could be inferred for moderate to severe disease due to the disproportionately higher RDS patients in the O2 protocol.

pared with moderate to severe disease. Furthermore, the O2 protocol may have a negative affect when bronchiolitis is more severe. However, no direct conclusive relationship between LOS and the protocol could be inferred for moderate to severe disease due to the disproportionately higher RDS patients in the O2 protocol. We performed a multiple regression analysis to independently assess each subject group and variables affecting the LOS (Table 5). We found age, RDS, and PICU transfer all had a significant correlation with LOS. A one-month increase in age significantly decreased LOS. RDS significantly impacted LOS, whereas a one-unit increase in RDS significantly increased LOS by 11.3% and a two-unit increase in RDS increased LOS by 23.9%. Transfer to the PICU significantly increased LOS by 2.9 fold. The O2 protocol had a significant inverse association with LOS compared to the control group; patients in the O2 protocol group had a 19.7%, decrease in LOS (Table 4).

ignificantly increased LOS by 11.3% and a two-unit increase in RDS increased LOS by 23.9%. Transfer to the PICU significantly increased LOS by 2.9 fold. The O2 protocol had a significant inverse association with LOS compared to the control group; patients in the O2 protocol group had a 19.7%, decrease in LOS (Table 4). 4. Discussion High variability in clinical care often contributes to higher healthcare costs and poor adherence to evidence-based practices [21]. For this reason, health care professionals have developed protocols to drive therapies and reduce the lack of concordance. Studies have shown clinical outcomes from nonphysician directed protocols compare favorably with physician driven interventions in multiple settings [22–24]. Such protocols used in the PICU have the potential to save money and reduce resource allocation when used in the non-ICU setting. Our study suggests that this is potentially true. We reduced LOS for patients with mild bronchiolitis (initial RDS < 3) after the implementation of the O2 protocol and deimplementation of unnecessary practices. However, LOS appeared to increase for patients with higher RDS following protocol implementation. The actual impact of RDS on LOS is somewhat confounded by the fact that RDS was higher in the postimplementation years even though the rates of assessment were similar in both groups. When we controlled for the RDS effect by using multivariate regression analysis there was a demonstrable aggregate benefit that was more than compensated for the increased LOS in sicker patients.

he fact that RDS was higher in the postimplementation years even though the rates of assessment were similar in both groups. When we controlled for the RDS effect by using multivariate regression analysis there was a demonstrable aggregate benefit that was more than compensated for the increased LOS in sicker patients. Deimplementation and deinnovation are quality improvement (QI) terms that emphasize the abandonment of unnecessary care that is not supported by evidence-based research [7, 8]. These terms focus on the ideal of discouraging use rather than discouraging underuse and have been used in the context of eliminating nonevidenced-based practices in bronchiolitis [9]. We paired the initiation of O2 use protocol along with deimplementation of unnecessary practices and hypothesized that a collective approach would have specific value in reducing unnecessary care. This provided a mechanism to overcome the inertia of so-called “established” clinical practice and increase the provider's sense of efficacy.

iation of O2 use protocol along with deimplementation of unnecessary practices and hypothesized that a collective approach would have specific value in reducing unnecessary care. This provided a mechanism to overcome the inertia of so-called “established” clinical practice and increase the provider's sense of efficacy. The AAP prioritizes the prevention of unnecessary care [5, 6], and a recent study showed benefit to eliminating practices such as X-rays and alpha-agonist therapy in community hospital settings [9]. The deimplementation of such practices requires constant education. It was our experience that the respiratory therapists were the strongest advocates for avoiding such unnecessary therapy. Applications of nebulized β-agonist therapy and/or hypertonic saline were rarely tried, and when attempted, they were discontinued once lack of efficacy was established. Such an approach has benefit in saving money, eliminating unnecessary interventions, and focusing care on the sickest during the time of the year that hospitals are busiest [25].

agonist therapy and/or hypertonic saline were rarely tried, and when attempted, they were discontinued once lack of efficacy was established. Such an approach has benefit in saving money, eliminating unnecessary interventions, and focusing care on the sickest during the time of the year that hospitals are busiest [25]. There is wide variation in the clinical course of bronchiolitis and thus it is difficult to distinguish which patients will require only titration of O2 therapy from those who will require a more involved escalation of care [26, 27]. For example, there have been efforts in the past to define clinical criteria that could predict inpatient LOS for children with bronchiolitis that have shown that initial spO2 values do not predict LOS in children with bronchiolitis [26]. Our protocol attempts to surmount some of that dilemma by predicating appropriate care and letting the clinical course dictate the level of care the patient receives.

ld predict inpatient LOS for children with bronchiolitis that have shown that initial spO2 values do not predict LOS in children with bronchiolitis [26]. Our protocol attempts to surmount some of that dilemma by predicating appropriate care and letting the clinical course dictate the level of care the patient receives. This study does have limitations in that the single center retrospective design with historical controls limits our ability to conclusively state that there is benefit to the O2 supplementation protocol. This study design by nature is an observational study. It is possible that natural cycle of viral virulence, unknown changes to childhood immunity, and other unknown variations may affect this study. In addition, other interventions independent of this oxygen protocol may have played a role in these outcomes, in particular LOS. One criticism for our study might be that the RDS score we used is more of an amalgam of other scores available in the literature rather than a validated score. However, the score was uniformly applied to all patients, and clearly an increase in the score was associated with an increase in respiratory distress. Also, the RDS scores were notably higher on average in the O2 protocol group. It is possible this is due to a true difference in illness severity between the seasons evaluated, or the increased frequency of respiratory therapist monitoring mandated by the O2 protocol resulted in increased provider confidence in keeping patients with higher illness severity on general inpatient floors. Regardless, we presented data that the weaning protocol might have benefit in a selected group of milder bronchiolitis cases.

frequency of respiratory therapist monitoring mandated by the O2 protocol resulted in increased provider confidence in keeping patients with higher illness severity on general inpatient floors. Regardless, we presented data that the weaning protocol might have benefit in a selected group of milder bronchiolitis cases. 5. Conclusion This study is unique in that it highlights both the benefit and the unforeseen effect of applying protocols to patient care. Our application of the AAP guidelines in a collaborative manner saw a decrease in LOS of children with milder bronchiolitis while LOS for the sicker patients increased. However, the overall effect was one of the benefits with potential to exert an impact on appropriate hospital triage and cost. This translates into a benefit for both patients and their families. Acknowledgments The authors thank Matthew Pridgeon, MD, who helped to conceive and implement the oxygen supplementation protocol, and Stephanie Raymundo, MD, who aided in data acquisition and in conceptualizing this study. Abbreviations RSV:Respiratory syncytial virus O2:Oxygen LOS:Length of stay AAP:American Academy of Pediatrics spO2:Blood oxygen saturation PICU:Pediatric intensive care unit QI:Quality improvement ICD-9:International Classification of Diseases, Ninth Revision RDS:Respiratory distress score. Conflicts of Interest The authors declare that they have no conflicts of interest. Figure 1 Standardized O2 supplementation and continuous pulse oximetry protocol. Table 1 Clinical practices discouraged on initiation of protocol. Discouraged Encouraged

ICD-9:International Classification of Diseases, Ninth Revision RDS:Respiratory distress score. Conflicts of Interest The authors declare that they have no conflicts of interest. Figure 1 Standardized O2 supplementation and continuous pulse oximetry protocol. Table 1 Clinical practices discouraged on initiation of protocol. Discouraged Encouraged Chest radiography Weaning O2 flow and FIO2 Viral panel testing Accepting saturation of >90% if child does not appear distressed Blood draws for laboratory testing Discontinuing supplemental gas flow when patient is on room air β-Agonist therapy Calling the physician for any perception of deterioration Steroids Spot checks in pulse oximetry Antibiotic therapy Table 2 Respiratory distress score (RDS) calculation. Respiratory distress score Respiratory rate □ 0 Normal (respiratory rate (rr) up to 40 breathes/minute) □ 1 Elevated (RR 40–60 breaths/minute) □ 2 Tachypnea (RR greater than 60 breaths/minute) Accessory muscle use □ 0 Normal □ 1 Retractions (substernal, subcostal, intercostal) □ 2 Neck or abdominal muscle use Wheezing □ 0 None or scattered end expiratory wheezes □ 1 Wheezes throughout expiration □ 2 Entire inspiration and expiration wheezes Oxygen requirement □ 0 Maintains SpO2 above 90% on room air □ 1 Maintains SpO2 above 90% on less than 1 Lpm oxygen □ 2 Requires 1 Lpm oxygen or more to maintain SpO2 above 90% Inspiratory to expiratory ratio □ 0 I : E ratio less than 1 : 2 □ 1 I : E ratio 1 : 2 to 1 : 3 □ 2 I : E ratio greater than 1 : 3 Table 3 Demographic and clinical variables.

ins SpO2 above 90% on room air □ 1 Maintains SpO2 above 90% on less than 1 Lpm oxygen □ 2 Requires 1 Lpm oxygen or more to maintain SpO2 above 90% Inspiratory to expiratory ratio □ 0 I : E ratio less than 1 : 2 □ 1 I : E ratio 1 : 2 to 1 : 3 □ 2 I : E ratio greater than 1 : 3 Table 3 Demographic and clinical variables. Variable Control group (n = 141) O2 protocol group (n = 122) p value Age (months) 6.2 ± 5.5 7.0 ± 6.3 0.31 Gender: male/female 77 (54.6%)/64 (45.4%) 70 (57.4%)/52 (42.6%) 0.65 RDS assessed 125/141 (88.7%) 110/122 (90.2%) 0.69 RDS 2.0 ± 1.5 2.7 ± 1.6 <0.001 LOS (h)a 69.6 ± 67.5 72.5 ± 77.4 0.374 LOS: RDS < 3a,b 70.6 ± 60.3 51.6 ± 42.6 0.005 LOS: RDS ≥ 3a,c 74.0 ± 53.4 95.2 ± 95.7 0.535 Duration of O2 supplementation (h)a 38.3 ± 58.4 40.9 ± 62.9 0.638 Number requiring supplemental O2 134/141 (95.0%) 113/122 (92.6%) 0.414 PICU transfer 10/141 (6.7%) 13/122 (9.7%) 0.310 7-day readmission 1/141 (0.7%) 2/122 (1.6%) 0.598 Data are presented as the mean ± SD or as percentages; O2, oxygen; LOS, length of stay; RDS, respiratory distress score; PICU, pediatric intensive care unit. aData were analyzed using log transformed data, values shown are untransformed data. bControl group n = 87; oxygen protocol group n = 49. cControl group n = 38; oxygen protocol group n = 61. Table 4 Study group LOSa and PICU transfer compared by RDS.

Variable Control group (n = 141) O2 protocol group (n = 122) p value Age (months) 6.2 ± 5.5 7.0 ± 6.3 0.31 Gender: male/female 77 (54.6%)/64 (45.4%) 70 (57.4%)/52 (42.6%) 0.65 RDS assessed 125/141 (88.7%) 110/122 (90.2%) 0.69 RDS 2.0 ± 1.5 2.7 ± 1.6 <0.001 LOS (h)a 69.6 ± 67.5 72.5 ± 77.4 0.374 LOS: RDS < 3a,b 70.6 ± 60.3 51.6 ± 42.6 0.005 LOS: RDS ≥ 3a,c 74.0 ± 53.4 95.2 ± 95.7 0.535 Duration of O2 supplementation (h)a 38.3 ± 58.4 40.9 ± 62.9 0.638 Number requiring supplemental O2 134/141 (95.0%) 113/122 (92.6%) 0.414 PICU transfer 10/141 (6.7%) 13/122 (9.7%) 0.310 7-day readmission 1/141 (0.7%) 2/122 (1.6%) 0.598 Data are presented as the mean ± SD or as percentages; O2, oxygen; LOS, length of stay; RDS, respiratory distress score; PICU, pediatric intensive care unit. aData were analyzed using log transformed data, values shown are untransformed data. bControl group n = 87; oxygen protocol group n = 49. cControl group n = 38; oxygen protocol group n = 61. Table 4 Study group LOSa and PICU transfer compared by RDS. Outcome measure RDS < 3 (n = 136) RDS ≥ 3 (n = 99) p value Control group LOS (h)b 70.6 ± 60.3 74.0 ± 53.4 0.697 O2 protocol group LOS (h)c 51.3 ± 41.4 92.4 ± 94.3 0.005 PICU transfer 9/136 (6.6%) 12/99 (12.1%) 0.144 Data are presented as the mean ± SD or as percentages; O2, oxygen; LOS, length of stay; RDS, respiratory distress score; PICU, pediatric intensive care unit. aData analyzed using log transformed data, values shown are untransformed data. bRDS < 3 group n = 87; RDS ≥ 3 group n = 38; cRDS < 3 group n = 49; RDS ≥ 3 group n = 61.

e presented as the mean ± SD or as percentages; O2, oxygen; LOS, length of stay; RDS, respiratory distress score; PICU, pediatric intensive care unit. aData analyzed using log transformed data, values shown are untransformed data. bRDS < 3 group n = 87; RDS ≥ 3 group n = 38; cRDS < 3 group n = 49; RDS ≥ 3 group n = 61. Table 5 Multiple regression analysis, with log transformed length of stay (LOS) as the dependent variable. Variable β-Coefficient 95% CI p value O2 protocol group∗ −0.22 −0.42–−0.02 0.030 Age −0.02 −0.04–−0.01 0.008 RDS 0.11 0.04–0.17 0.001 PICU Transfer 1.06 0.72–1.41 <0.001 CI, confidence interval; O2, oxygen; RDS, respiratory distress score; PICU, pediatric intensive care unit. ∗Control group (reference group) versus the O2 protocol group.

1. Introduction Pneumonia is the single greatest cause of death in children worldwide, with an estimated 1.3 million deaths in 2011 and more than 90% occurring in developing countries [1–3]. It is responsible for 4% of deaths in newborns and 14% of deaths in pediatric patients [4]. The incidence of CAP is lower in developed countries: in the US it is about 35–40/1000/person-years in children < 5 years old, 20/1000 person-years in children 5–10 years old, and 10/1000 person-years in children > 10 years old. Despite this, approximately 50% of children with CAP < 5 years old, 20% between 5–10 years old, and 10% of children > 10 years old need to be hospitalized [5]. These numbers demonstrate the burden that CAP represents for society and for economic healthcare resources. 2. Materials and Methods In the first part of the study, we compared the latest national and international guidelines on pediatric CAP, including all those who were published since 2005 to 2016, focusing on their recommendations for first-line therapies. Then we performed a search on PubMed and Scopus databases, looking for studies published from 2010 to 2016 about CAP antimicrobial therapy in children, trying to get data from as many different countries as possible. We also performed hand-search of references of relevant articles. Our search included both retrospective and prospective studies, mainly cross-sectional and hospital based, including both inpatients and outpatients. All of them except for one [6] included pediatric patients only.

s many different countries as possible. We also performed hand-search of references of relevant articles. Our search included both retrospective and prospective studies, mainly cross-sectional and hospital based, including both inpatients and outpatients. All of them except for one [6] included pediatric patients only. To get a more extensive review of CAP prescribing behavior, for those countries where specific studies on antimicrobial prescriptions for CAP were not available, a search for articles on antimicrobial prescriptions in pediatric age groups was performed. All articles including CAP as reason for treatment were included. 3. Results and Discussion 3.1. Different Countries, Same Pathogens Organisms responsible for CAP vary stratifying children by age because of the developing immune system and age-related exposures: viruses or mixed infections are more common amongst younger patients (children under 5 years of age), while exclusive bacterial origin and atypical etiology (mainly Mycoplasma pneumoniae and Chlamydophila pneumoniae) are more often identified in older children [7, 8]. S. pneumoniae and Haemophilus influenzae are the commonest bacterial pathogens isolated in children under five years with CAP accounting for 30%–50% and 10%–30%, respectively [9]. Around 50% of deaths due to pneumonia are attributable to these organisms [10].

umoniae) are more often identified in older children [7, 8]. S. pneumoniae and Haemophilus influenzae are the commonest bacterial pathogens isolated in children under five years with CAP accounting for 30%–50% and 10%–30%, respectively [9]. Around 50% of deaths due to pneumonia are attributable to these organisms [10]. Viral etiology has been documented in up to 80% of CAP cases in children younger than 2 years and much less in older children (10–16 years). The most frequently identified viral pathogen in younger children is Respiratory Syncytial Virus (RSV), rarely detected in older children. Less frequent are Adenoviruses, Bocavirus, Human Metapneumovirus, Influenza A and B Viruses, Parainfluenza Viruses, Coronaviruses, and Rhinovirus. Up to 33% of hospitalized children are simultaneously infected by 2 or more viruses. Mixed infections (both of viral and bacterial etiology) have been documented in 2–50% of children with CAP, more frequently in inpatients, which are more seriously ill than outpatients [3, 11]. Atypical pneumonia caused by different pathogens is characterized by a different clinical course: slowly progressing, with malaise, sore throat, low-grade fever, and cough developing over 3–5 days. The main organisms responsible for atypical pneumonia are M. pneumoniae in older children and C. pneumoniae in infants. Legionella species are rarely identified in children [8, 12, 13].

ed by a different clinical course: slowly progressing, with malaise, sore throat, low-grade fever, and cough developing over 3–5 days. The main organisms responsible for atypical pneumonia are M. pneumoniae in older children and C. pneumoniae in infants. Legionella species are rarely identified in children [8, 12, 13]. The etiologic definition is difficult for many reasons, such as low yield of blood cultures, difficulty in obtaining adequate sputum specimens from younger children, frequent specimen contaminations by upper airways bacterial flora and invasiveness of pulmonary biopsy, lung aspiration, and bronchoalveolar lavage which are rarely performed [13]. However, over the last 10 years, there have been improvements in PCR techniques for viral identification on nasopharyngeal aspirates or secretion, and molecular assays are now commonly used in Europe and in the US.

iveness of pulmonary biopsy, lung aspiration, and bronchoalveolar lavage which are rarely performed [13]. However, over the last 10 years, there have been improvements in PCR techniques for viral identification on nasopharyngeal aspirates or secretion, and molecular assays are now commonly used in Europe and in the US. Vaccines are the most effective strategy for prevention of pediatric CAP. Haemophilus influenzae type B (HiB) conjugate vaccine and 7-valent pneumococcal conjugate vaccines (PCV7) dramatically decreased the incidence of bacterial CAP after introduction of universal vaccination campaigns [14, 15]. PCVs have been included for some years in the immunization schedules of children in their first year of life in many countries and they have completely modified the burden of pneumococcal diseases among these children and their unvaccinated contacts of any age [16]. Currently, the polyvalent pneumococcal vaccine (PCV13) confers immunity to approximately 85% of serotypes responsible for most invasive pneumococcal diseases [17]. 3.2. Same Pathogens, Same Treatment: International CAP Recommendations Since its introduction during the 20th century, antibiotic therapy, along with vaccines, has decreased CAP mortality of 97% in developed countries [14]. Most of the time the choice of an antimicrobial agent is empirical and based on the most common etiologies for each age group, on the local prevalence of causative organisms, and on the presence of risk factors for atypical or resistant bacteria [18].

nes, has decreased CAP mortality of 97% in developed countries [14]. Most of the time the choice of an antimicrobial agent is empirical and based on the most common etiologies for each age group, on the local prevalence of causative organisms, and on the presence of risk factors for atypical or resistant bacteria [18]. During the last 10 years, many guidelines have defined the best antimicrobial regimen for CAP in children considering spectrum of activity, antimicrobial susceptibility, tolerability, bioavailability, safety, and cost [19, 20]. As already highlighted by other authors, these guidelines present some differences in treatment strategies, but almost all agree on the first-line therapy to administer in case of CAP (Figure 1) [19]. For infants < 2 months of age, the association with ampicillin and aminoglycosides is the most suggested therapy, ensuring coverage for Group B streptococci and Gram-negatives. In case of atypical pneumonia, in this period of life, because of the possibility of Chlamydia trachomatis infection, macrolides are recommended [3, 19, 21–23]. For all children > 3 months of age, the narrowest regimen with S. pneumoniae activity is suggested worldwide. Penicillin is the ideal first-line therapy, being a narrow-spectrum agent achieving therapeutic concentrations for S. pneumoniae in the lung up to MIC of 4 mg/ml [24]. However, due to its limited bioavailability, oral amoxicillin is reported as an equivalent and more feasible option [24, 25].

ty is suggested worldwide. Penicillin is the ideal first-line therapy, being a narrow-spectrum agent achieving therapeutic concentrations for S. pneumoniae in the lung up to MIC of 4 mg/ml [24]. However, due to its limited bioavailability, oral amoxicillin is reported as an equivalent and more feasible option [24, 25]. Despite general agreement on the agent, differences in dose and posology have been reported, varying according to pneumococcal resistance [19]. Indeed, beta-lactam effectiveness is time dependent and S. pneumoniae does not develop resistance through β-lactamase enzyme production, but through the alteration of the cell wall's antimicrobial targets (penicillin-binding proteins) [26]. Thus, in the setting of resistant S. pneumoniae serotype, higher concentration at the infection site is needed in order to saturate penicillin-binding proteins and to overcome resistance [27].

mase enzyme production, but through the alteration of the cell wall's antimicrobial targets (penicillin-binding proteins) [26]. Thus, in the setting of resistant S. pneumoniae serotype, higher concentration at the infection site is needed in order to saturate penicillin-binding proteins and to overcome resistance [27]. A study of children with pulmonary pneumococcal infection [28] provided data to develop a model for describing amoxicillin pharmacokinetics administered with different patterns: 50 mg/kg/day in two or three administrations daily. The resulting curve, integrated with S. pneumoniae MIC for amoxicillin, showed that, for intermediate resistant S. pneumoniae (MIC 4 mg/ml) CAP, the amoxicillin plasma concentration remained above the pneumococcal MIC level for about 4 hours. Therefore, amoxicillin administered every 8 hours maintains blood and lung concentrations that are above S. pneumoniae MIC for enough time to allow S. pneumoniae eradication. A longer interval between administrations (every 12 hours), in case of intermediate resistant serotypes, would not permit having a sufficient antimicrobial plasma concentration [28]. Similarly, penicillin G needs more frequent administrations than other beta-lactams, because of its shorter half-life [13].

eradication. A longer interval between administrations (every 12 hours), in case of intermediate resistant serotypes, would not permit having a sufficient antimicrobial plasma concentration [28]. Similarly, penicillin G needs more frequent administrations than other beta-lactams, because of its shorter half-life [13]. Beta-lactam dose is the other key factor for pathogen eradication. Through the different guidelines, amoxicillin daily dose varies from 40–50 mg/kg to 90–100 mg/kg, with higher dosage recommended in areas with higher risk for antibiotic-resistant serotype, as in the US [13, 19]. In the same way, for inpatient parenteral therapy, higher doses of penicillin G or ampicillin are recommended [13].

the different guidelines, amoxicillin daily dose varies from 40–50 mg/kg to 90–100 mg/kg, with higher dosage recommended in areas with higher risk for antibiotic-resistant serotype, as in the US [13, 19]. In the same way, for inpatient parenteral therapy, higher doses of penicillin G or ampicillin are recommended [13]. The only two guidelines which suggest an aminopenicillin plus beta-lactamase inhibitor as first line are the Taiwan Pediatric Working Group and Asociacion Espanola de Pediatria de Atencion Primaria [29, 30]. Unlike the first one, in which aminopenicillin plus beta-lactamase inhibitor (e.g., amoxicillin-clavulanate) is suggested as first-line therapy for all children treated as outpatient, the Spanish guidelines recommend coamoxiclav only for children who are not fully immunized with conjugate vaccines for type B H. influenzae and for S. pneumoniae. Indeed, this population is at increased risk to develop a CAP by aggressive S. pneumoniae serotypes and other less common organisms, as H. influenza. Unlike Pneumococcus, type B and nontypeable H. influenzae became resistant to penicillin through the production of β-lactamase. Therefore, treatment with the association of amoxicillin with a β-lactamase inhibitor ensures a broader coverage [30]. It should be noted that the addition of a β-lactamase inhibitor does not change the amoxicillin kinetic curve; as a consequence, in order to treat a pneumococcal infection with the association of amoxicillin with clavulanate, the therapy should be administered every 8 hours [26].

ibitor ensures a broader coverage [30]. It should be noted that the addition of a β-lactamase inhibitor does not change the amoxicillin kinetic curve; as a consequence, in order to treat a pneumococcal infection with the association of amoxicillin with clavulanate, the therapy should be administered every 8 hours [26]. The WHO guidelines are the only one suggesting cotrimoxazole as alternative to amoxicillin in outpatient treatment. This recommendation derived from evidence of no difference in treatment failure rates between amoxicillin and cotrimoxazole [31–33]. Despite concerns about the increase of S. pneumoniae and H. influenzae resistant to cotrimoxazole, as demonstrated by some authors [34], the reason for this indication is mainly attributable to economic factors. Indeed, for children <10 kg, the cost of a five-day treatment with amoxicillin is higher than the same duration on cotrimoxazole [35–37]. No guidelines recommend oral cephalosporins as first-line therapy. Indeed, pharmacokinetic and pharmacodynamic studies showed that none of the available oral cephalosporins is able to exceed the pneumococcal MIC for more than 50% of the time between two administrations [26]. Moreover, recent US data on S. pneumoniae susceptibility to cefdinir and cefuroxime indicated only 70% to 80% efficacy, compared with 84% to 92% amoxicillin efficacy [38, 39].

none of the available oral cephalosporins is able to exceed the pneumococcal MIC for more than 50% of the time between two administrations [26]. Moreover, recent US data on S. pneumoniae susceptibility to cefdinir and cefuroxime indicated only 70% to 80% efficacy, compared with 84% to 92% amoxicillin efficacy [38, 39]. The only cephalosporin that has been demonstrated superior to penicillin in S. pneumoniae eradication, even if resistant, is ceftriaxone [40]. No microbiologic failures have been reported for S. pneumoniae with ceftriaxone MIC of 4.0 mg/mL [13, 41]. Thus, ceftriaxone or cefotaxime in standard doses is suggested by all guidelines as alternatives in case of first-line treatment failure, severe clinical conditions, or not fully immunized children [3, 7, 13, 21–23, 29, 30, 41]. Due to high prevalence of macrolide resistance circulating strains of S. pneumoniae, macrolides are not recommended as empiric therapy for CAP. Their use is suggested only when atypical etiology is suspected or in case of persistence of symptoms despite beta-lactams administration [7, 13, 42]. This strict indication for macrolides use derives from the evidence that Mycoplasma lower respiratory tract infection (LRTI) has a high rate of spontaneous clinical remission and the use of azithromycin has been associated with the selection of resistant organisms because of its prolonged serum elimination half-life [13]. Moreover, no significant benefits of antibiotic treatment in M. pneumonia infection have been documented [37].

ion (LRTI) has a high rate of spontaneous clinical remission and the use of azithromycin has been associated with the selection of resistant organisms because of its prolonged serum elimination half-life [13]. Moreover, no significant benefits of antibiotic treatment in M. pneumonia infection have been documented [37]. For complicated pneumonia (i.e., moderate parapneumonic effusion and necrotizing pneumonia), antimicrobial therapy must be broadened to cover less common but highly aggressive pathogens as Streptococcus pyogenes and S. aureus. As for S. pneumoniae, macrolides cannot be considered an effective empiric therapy because of the high level of resistance [13]. Despite the fact that no penicillin or cephalosporin resistance has been reported for S. pyogenes, some authors suggest that, in case of concomitant symptoms attributable to toxic shock syndrome, combination therapy with clindamycin decreases the severity of symptoms [43]. In fact, since clindamycin inhibits protein synthesis (by binding the 50S subunit of the bacterial ribosome), it inhibits the production of S. aureus toxins, resulting in a lower inflammatory reaction. Clindamycin may be bacteriostatic or bactericidal depending on the organism and drug concentration and is indicated by US guidelines as a good option for both methicillin susceptible S. aureus (MSSA) and community-acquired methicillin-resistant S. aureus (CA-MRSA) strains [13].

sulting in a lower inflammatory reaction. Clindamycin may be bacteriostatic or bactericidal depending on the organism and drug concentration and is indicated by US guidelines as a good option for both methicillin susceptible S. aureus (MSSA) and community-acquired methicillin-resistant S. aureus (CA-MRSA) strains [13]. Nowadays almost all MSSA have penicillin resistance which can be overcome with the addition of a β-lactamase inhibitor or through penicillinase-resistant beta-lactams, such as oxacillin or first-generation cephalosporins. MRSA strains have mecA gene that encodes penicillin-binding protein 2a, an enzyme that has low affinity for beta-lactams, leading to resistance to all antibiotics active against MSSA. During the last decade, both community-associated and hospital-acquired infections with MRSA have increased. MRSA, accounting for 20%–40% of all hospital-acquired pneumonia (HAP) and ventilator-associated pneumonia (VAP), have demonstrated a rapid increase as cause of pneumonia even in patients without exposure to the healthcare system [44]. This CA-MRSA has become an important cause of CAP complicated by empyema and necrosis [45].

accounting for 20%–40% of all hospital-acquired pneumonia (HAP) and ventilator-associated pneumonia (VAP), have demonstrated a rapid increase as cause of pneumonia even in patients without exposure to the healthcare system [44]. This CA-MRSA has become an important cause of CAP complicated by empyema and necrosis [45]. Since erythromycin resistance predicts inducible clindamycin resistance in many isolates, a D-test to assess clindamycin susceptibility should always be performed. In case of D-test positivity, the use of clindamycin should be avoided, since it is highly possible that the organism will become resistant during the infectious process, especially in high-inoculum infections such as empyema [45]. On the other hand, all CA-MRSA strains are susceptible to vancomycin, which is considered by all guidelines as the drug of choice if MRSA is suspected [7, 13]. Although linezolid has been recently demonstrated as efficient as vancomycin for the treatment of MRSA pneumonia, its use should be considered as a second-line treatment for cost consideration (linezolid costs >10 times more than vancomycin) and because linezolid-resistant MRSA has already been described [46, 47].

13]. Although linezolid has been recently demonstrated as efficient as vancomycin for the treatment of MRSA pneumonia, its use should be considered as a second-line treatment for cost consideration (linezolid costs >10 times more than vancomycin) and because linezolid-resistant MRSA has already been described [46, 47]. 3.3. Different Countries, Same Treatment? A worldwide review about CAP antimicrobial therapy in children includes 37 studies about antibiotics prescriptions in 50 countries published since 2010. The results are shown in Table 1 and Figure 2. Even if the studies were different in design and study population, their results give a good picture of the antibiotic prescription patterns in different environments, and they show the global heterogeneity in the application of the guidelines for the treatment of childhood pneumonia.

are shown in Table 1 and Figure 2. Even if the studies were different in design and study population, their results give a good picture of the antibiotic prescription patterns in different environments, and they show the global heterogeneity in the application of the guidelines for the treatment of childhood pneumonia. In fact, the first important result of our review is that the correct implementation of the guidelines is not confined to specific areas but may be variable even inside the same country. For example, Iroh Tam et al., through a 2-year retrospective study on hospitalized children with CAP in six US centres, showed that the most used antibiotics were third-generation cephalosporins (73%), and only 1% of the patients received amoxicillin. These findings during the first 2 years after US guidelines publication led the authors to recommend more strategies for educating healthcare providers [71]. On the other hand, Thomson et al. in another retrospective study set in an US hospital, with the same population (hospitalized children between 3 months old and 18 years old) in a 15-month period (May 2011–July 2012), had an opposite result, reporting that 63,6% of the pediatric CAP were treated with aminopenicillins and only 16.8% with third-generation cephalosporins [80]. We found a similar situation comparing studies from France [63, 78] and India [57, 65].

In fact, the first important result of our review is that the correct implementation of the guidelines is not confined to specific areas but may be variable even inside the same country. For example, Iroh Tam et al., through a 2-year retrospective study on hospitalized children with CAP in six US centres, showed that the most used antibiotics were third-generation cephalosporins (73%), and only 1% of the patients received amoxicillin. These findings during the first 2 years after US guidelines publication led the authors to recommend more strategies for educating healthcare providers [71]. On the other hand, Thomson et al. in another retrospective study set in an US hospital, with the same population (hospitalized children between 3 months old and 18 years old) in a 15-month period (May 2011–July 2012), had an opposite result, reporting that 63,6% of the pediatric CAP were treated with aminopenicillins and only 16.8% with third-generation cephalosporins [80]. We found a similar situation comparing studies from France [63, 78] and India [57, 65]. Interestingly, in France our data about CAP prescriptions derive from two different settings. Launay and colleagues investigated antimicrobial prescriptions and recommendations adherence in a French Emergency Pediatrics Department through a prospective two-period study, including all children aged one month to 15 years. The results were encouraging, with an increase of recommendation compliance from 18.8% to 48% between 2009 and 2012, and a consequent increase of amoxicillin monotherapy prescription from 54.2% to 71% [78]. Dubos et al., on the other hand, give us a picture of CAP antimicrobial prescriptions through general practitioners (GPs), private pediatricians, and pediatric fellows. The results of the standardized questionnaire submitted to every participant showed that CAP guidelines were insufficiently followed, with high rate of amoxicillin/clavulanate prescriptions (amoxicillin in monotherapy was prescribed in only 29% of cases, for 54% of cases associated with clavulanic acid) [63].

fellows. The results of the standardized questionnaire submitted to every participant showed that CAP guidelines were insufficiently followed, with high rate of amoxicillin/clavulanate prescriptions (amoxicillin in monotherapy was prescribed in only 29% of cases, for 54% of cases associated with clavulanic acid) [63]. In India, in addition, we found some of the lowest rates of prescription on aminopenicillins as single therapy. Choudry and Bezbaruah, in a prospective observational study based in a university hospital in Assam, including inpatients up to 12 years, reported 0% use of penicillin as single therapy in cases of pediatric pneumonia. The therapy mostly used (54% of cases) was the combination of amoxicillin/clavulanate [58]. Another prospective study by Moinuddin et al. was conducted over 9 months in 2012, in two hospitals in Bangalore. The most widely used therapy was amoxicillin + clavulanate (43,8%), with third-generation cephalosporins as the most prescribed class (ceftriaxone 36.2%, cefotaxime 21%). Penicillin in single therapy accounted only for 1% of prescriptions [57]. Cephalosporins were often reported to be the class with higher rates of prescription for CAP treatment, as reported by many centres in different countries, like Ethiopia [60], Saudi Arabia [62], Nepal [73], Serbia [83], Sudan [67], US [54, 71, 74], Italy [76], and other European countries [48, 82].

In India, in addition, we found some of the lowest rates of prescription on aminopenicillins as single therapy. Choudry and Bezbaruah, in a prospective observational study based in a university hospital in Assam, including inpatients up to 12 years, reported 0% use of penicillin as single therapy in cases of pediatric pneumonia. The therapy mostly used (54% of cases) was the combination of amoxicillin/clavulanate [58]. Another prospective study by Moinuddin et al. was conducted over 9 months in 2012, in two hospitals in Bangalore. The most widely used therapy was amoxicillin + clavulanate (43,8%), with third-generation cephalosporins as the most prescribed class (ceftriaxone 36.2%, cefotaxime 21%). Penicillin in single therapy accounted only for 1% of prescriptions [57]. Cephalosporins were often reported to be the class with higher rates of prescription for CAP treatment, as reported by many centres in different countries, like Ethiopia [60], Saudi Arabia [62], Nepal [73], Serbia [83], Sudan [67], US [54, 71, 74], Italy [76], and other European countries [48, 82]. Feleke and colleagues conducted their 5-month prospective study in a large government hospital in Ethiopia. The study includes all children admitted in that period and CAP accounted for 56.3% of all drug prescriptions. Ceftriaxone was the most prescribed drug (43.5%) followed by gentamicin (25.6%), and penicillin and ampicillin ranked the third and fourth place [62]. In a retrospective study by Zec et al., during a 6-month period in 2014, first- and third-generation cephalosporins were given to children with CAP in 40.4% and 31.7% of cases, respectively. Penicillin was used in 25% of cases [83]. In an Italian 1-day point-prevalence survey on antimicrobial use in hospitalized neonates and children in 2012, the main indication for treatment in children was LRTI (34%), with higher prevalence of third-generation cephalosporins (43.3%) followed by macrolides accounting for 26.8%. No ampicillin/amoxicillin prescription was reported [76].

int-prevalence survey on antimicrobial use in hospitalized neonates and children in 2012, the main indication for treatment in children was LRTI (34%), with higher prevalence of third-generation cephalosporins (43.3%) followed by macrolides accounting for 26.8%. No ampicillin/amoxicillin prescription was reported [76]. Association of aminopenicillins was found to be often prescribed: amoxicillin + clavulanate was reported to be the most used therapy by studies conducted in Saudi Arabia [51], France [63], and India [58], and a study conducted in Iraq, by Younis, reported that ampicillin + cloxacillin, alone and in combination, accounted for 50% of the antibiotic prescriptions for the children with respiratory tract infections [50]. One study, in particular, reported a high rate of prescriptions of macrolides. It was conducted in Norway, by Fossum and colleagues, and included the prescriptions of general practitioners in case of respiratory tract infections in patients < 6 years. They found that macrolides were prescribed in 44% of the cases of pneumonia, more than penicillin V, which was used in 31%, and that extended spectrum penicillin accounted for 24% of the prescriptions [55].

ncluded the prescriptions of general practitioners in case of respiratory tract infections in patients < 6 years. They found that macrolides were prescribed in 44% of the cases of pneumonia, more than penicillin V, which was used in 31%, and that extended spectrum penicillin accounted for 24% of the prescriptions [55]. Studies on the appropriateness of prescriptions or prescriber behavior were also found. In addition to the aforementioned French study, Maltezou et al. showed how Greek private-practice pediatricians guidelines compliance is only around 30.6% [64]. Moreover, Ceyhan et al., in a multicenter point-prevalence survey with respiratory infection as main diagnosis, showed how cephalosporins and penicillin (most of the time combined with b-lactamase inhibitors) were improperly prescribed in 36.1% and 43.7% of cases, respectively. These analyses highlighted how, even now, adherence to guidelines is still low. On the other hand, Usonis and colleagues through a questionnaire developed and distributed by the CAP Pediatric Research Initiative (CAP-PRI) working group and distributed across Europe showed high adherence to CAP guidelines, with a high prescription rate of narrow-spectrum penicillin for inpatients (amoxicillin (32%) and ampicillin (37%)) and outpatients (amoxicillin (84%)) [81].