Browse the corpus

Leishmaniasis is a vector-borne parasitic disease of humans caused by over 20 different species of Leishmania (Kinetoplastida: Trypanosomatidae) transmitted by hematophagous female Phlebotomine sand flies (Diptera: Psychodidae) []. The disease is endemic in over 90 countries worldwide, with an estimated annual incidence of 700,000 to 1 million new cases globally [–]. The polymorphic outcomes of symptomatic (clinical) infection predominantly include cutaneous leishmaniasis (CL), mucocutaneous leishmaniasis (MCL), visceral leishmaniasis (VL, or kala-azar), and post-kala-azar dermal leishmaniasis (PKDL) []. CL is the most common form, presenting as simple to severe cutaneous ulcers and disfigurement which can progress to destruction of the oral-pharyngeal mucosa (MCL) or diffuse CL [,] resulting in significant morbidity. VL is the most severe form of leishmaniasis causing fever, weight loss, spleen and liver enlargement, with a fatality rate of >95% without effective diagnosis and treatment [].

Some 30 species of sand fly are proven or suspected vectors of Leishmania [,]. Sand fly blood-seeking behaviour is typically crepuscular/ nocturnal, though they may bite during day-light hours when disturbed []. A wide variety of vertebrates are indicated as sink (dead-end) or source (reservoir) hosts, the latter including humans, domestic dogs, lagomorphs, and synanthropic and wild rodents, depending on the Leishmania species-specific transmission cycle []. Military personnel are often exposed to infectious sand fly bites when deployed to endemic settings, notably in the Middle East [–], Africa [], Europe [,] and the Americas [–], where the cumulative incidence of CL for example ranges from 0.2% to 25.2% [,–,–]. VL is less common amongst the military [,–], accounting for, e.g., 1.2% of all leishmaniasis cases reported between 2001 and 2006 among US soldiers deployed to Iraq, Afghanistan, Oman and Saudi Arabia []. In the British military, only one VL case has been recorded to date, and which followed deployment to Iraq [MoD restricted data].

Development of leishmaniasis has significant health and financial consequences to both afflicted individuals and to military institutions. Patients suffer short to long-term health conditions resulting in prolonged time out of service; chemotherapeutic treatments are costly and often toxic []; and financial compensation claims against the MoD may be extensive []. Currently, there are no chemoprophylaxis or human vaccines to protect humans against leishmaniasis [], thus prevention primarily relies on mitigating against sand fly bites, on the control of the sand fly vector(s), and/or the mammalian reservoir populations []. Deployed personnel are usually supplied with personal protective measures (PPMs) accompanied by instructions of their use to best reduce exposure to sand fly bites. However, supply and compliant use of PPMs are not consistent, factors often associated with an increased risk of contracting leishmaniasis [,,,,–]. Previous reviews of leishmaniasis epidemiology in civilian populations do not address the specific circumstances of operational deployments [–]. This systematic review and meta-analysis of data specific to military populations aims to help identify knowledge gaps towards mitigating exposure to sand fly vectors and Leishmania infection during deployment.

A comprehensive literature search was developed collaboratively with the review team through an iterative process to identify peer-reviewed studies reporting any form of leishmaniasis among military personnel deployed to any location. The search was performed in multiple databases, including MEDLINE, EMBASE, Web of Science, Scopus, and ProQuest. Hand searches were also conducted in preprint servers (bioRxiv and medRxiv), article reference lists, and Google Scholar Citations for additional relevant studies. The search strategy combined MeSH terms and keywords, including terms related to sand fly, leishmaniasis, and military. Truncation and wildcards were used to capture variations of words such as ‘leishman*’ (to capture leishmaniasis, leishmanial, Leishmania ). Specific subtopics or aspects of leishmaniasis in the military was considered such as cumulative incidence, incidence proportion, attack rate, sand fly and Leishmania species characterisation, sand fly saliva antigen exposure and risk factors. The search was limited to studies published in English (or with translations into English) between January 1990 and October 2023.

Inclusion criteria were studies of deployed military populations reporting any of the following: cumulative incidence (incidence proportion or attack rate) of leishmaniasis, characterisation of sand fly and Leishmania species, sand fly saliva exposure, the effectiveness of PPMs and associated risk of developing laboratory confirmed asymptomatic or symptomatic leishmaniasis. Studies reporting cases of leishmaniasis without laboratory confirmation were excluded. Studies of non-military populations were also excluded. The screening results are reported following the PRISMA statement [].

Two authors independently conducted screening of titles and abstracts, review of full texts, data extraction, and performed quality assessment (rating) using the National Institutes of Health Quality Assessment Tools [], and the Newcastle Ottawa scale [] (Tables B and C in ). To ensure accuracy, the screening results were manually cross-checked, and any disagreements in data extraction or quality assessment were resolved through consensus. Data extraction included characteristics of the study design, setting, numbers and characteristics of confirmed subclinical and clinical leishmaniasis, study population size, and period of deployment to the endemic setting. Information on human immune responses to sand fly saliva proteins exposure (as a measure of individual sand fly bite exposure), and characterisation of the sand flies, the infecting Leishmania species and methods of detection, were also extracted (Table A in ).

The geographical locations of military deployments were grouped by WHO-defined geographical region []. The extracted data on the numbers of cases and the number of troops deployed over the reported deployment period were used to calculate the cumulative incidence [] otherwise known as the incidence proportion, as the number of new asymptomatic or symptomatic confirmed cases divided by the total number of troops at the start of the deployment period. Deployment periods were short (median 1 months, maximum 6 months), and we assumed that the deployed troops had no previous history of leishmaniasis, were fully susceptible to infection, and there was no significant loss-to-follow-up. PPMs and associated risk of developing leishmaniasis were assessed for their suitability to be pooled in a meta-analysis. Effect sizes for single proportions (cumulative incidence), and precomputed effect sizes for the effectiveness of PPMs and associated risk of developing leishmaniasis, were calculated and declared. Potential impacts of data transformations on effect sizes (Freeman-Tukey, and Logit) were compared to the raw proportions, confirming no substantial differences in the overall estimates (Figs F–I in ). Thus, the Freeman-Tukey method for estimating cumulative incidence was adopted and reported here. For all meta-analyses, data were fitted to random effects models using restricted maximum likelihood [,]. For evaluation of the effectiveness of PPMs and associated risk of developing leishmaniasis, the precalculated effect sizes (odds ratios [OR] and confidence intervals) reported in the identified publications were used to calculate the natural log OR (ln[OR]) and associated standard error (SE), after which they were pooled for meta-analysis.

The degree of heterogeneity between study outcomes was assessed using I² (I-squared), τ² (Tau-squared), H² (H-squared), and Q (Cochran’s Q test) statistics. Low values of τ² (near 0), I² (close to 0%), H² (near 1), and Q with p-values p > 0.05, suggested low or no heterogeneity, while higher values of τ² , I² >75%, H² >1, and Q with a low p-value p< 0.05, indicated substantial heterogeneity [,]. The Z-test was used to assess whether the overall pooled effect size was significantly different from zero. Publication bias was assessed using the Egger’s test []. For meta-analysis including at least five studies, sensitivity analyses were performed, showing changes in the summary effect estimate and confidence intervals resulting from excluding one study at a time from the meta-analysis (Tables E–H in ). As meta-analysis was not possible to address all the objectives of this review due to quantitative data limitations, a narrative synthesis was conducted following guidelines of Popay et al. [].

The literature search yielded a total of 1,456 studies (). Following the removal of duplicates, 610 studies underwent screening based on their titles and abstracts. Of these, 468 records were excluded because they were not relevant to the subject of this review. The full text of 142 studies were assessed for eligibility, of which 36 met the inclusion criteria (). Of the nine excluded non-English publications without an English translation, only two were potentially relevant.

The 36 included studies reported on military personnel deployed to the American region (AMR; 16 studies), the Eastern Mediterranean region (EMR; 15 studies), the South-East Asian region (SEAR; 2 studies) and 1 study each deployed to the African region (AFR), the European region (EUR), and to countries spanning two regions (EUR and EMR) (Table A in ). Out of the 36 studies, 8 (22.2%) reported the cumulative incidence of symptomatic leishmaniasis, 3 (8.3%) the cumulative incidence of asymptomatic leishmaniasis, 24 (66.7%) characterised the Leishmania species aetiology, 7 (19.4%) identified the local sand fly species, 4 (11.1%) reported on soldier immune responses to sand fly salivary proteins, and 13 (36.1%) reported on the effectiveness of PPMs and associated risk of developing subclinical and clinical leishmaniasis. Of the 24 studies that reported confirmed clinical outcomes, 20 (83.3%) studies reported CL, three (12.5%) reported VL, and one study (4.2%) reported both CL and VL (deployment to Afghanistan) [].

The 36 study designs comprised six cohort studies [,,,,,], fifteen cross-sectional studies [,,,,,,–], one mixture of cohort and case-control study [], two mixtures of cross-sectional and cohort study [,], four case-control studies [–], three case reports [,,] and five case series [,,–]. The publication dates of the studies ranged from 1992 to 2023, with the majority ( n = 20, 57.1%) published between 2011 and 2023. The number of participants per study ranged from 1 to 5000 soldiers. Study characteristics (location, study design and sample size) and quality ratings, as defined in Methods, are shown in Tables A-C in . Eleven studies reported cumulative incidence, eight of symptomatic CL [,,,,,,,] and three of asymptomatic (subclinical) confirmed Leishmania infections [,,]. Thirteen cases of symptomatic VL were reported in three studies [,,] which were too few to perform a meta-analysis, hence are described under the narrative accounts below.

Of the eight studies that reported the cumulative incidence of symptomatic CL (eleven estimates), five studies were among soldiers deployed to AMR [,,,,], two to EMR [,] and one to AFR []. The duration of deployment ranged from <1 month [,], 1 – 4 months [,,] to >4 months (maximum 6 months) [,,]. Meta-analysis of the pooled Freeman-Tukey transformed proportions estimated an overall cumulative incidence of symptomatic CL of 10% (95% C.I.: 5, 16) with significant heterogeneity ( I 2 = 97.84%) between studies (), and no evidence of publication bias ( z = 1.85, p = 0.065) (Fig A in ). The seven estimates from the five studies within AMR showed similar heterogeneity ( I 2 = 97.90%) (Fig D in ). High cumulative incidence of symptomatic CL was in troops deployed to AMR [,], though not exclusively so ( and in ). Pooled estimates by region suggested highest values in the EMR (14% [95% CI: 12, 16]), though no statistical difference was detected between WHO regions (Bayesian ANOVA [Q b ] = 1.90; P = 0.39). The cumulative incidence associated with wet versus dry season of deployment was not dissimilar (10% [95% CI: 5, 16]) (). The duration of deployment lasting 1-4 months appeared to result in a higher cumulative incidence (15% [95% CI: 07, 25]) compared to <1 month and >4 months deployments (<9%) ( and in ).

Asymptomatic Leishmania infections were detected in troops post-deployment by testing for anti- Leishmania antibodies using enzyme-linked immunosorbent assay (ELISA) [], or by both ELISA and screening for Leishmania DNA by polymerase chain reaction (PCR) [,]. Of the three identified studies, one reported on troops deployed to EMR [], one to EUR [] and one deployed to both EMR and EUR []. Two molecular studies detected L. donovani/infantum complex DNA in blood samples. One study of Austrian troops deployed to Syria (EMR), Lebanon (EMR), and Bosnia and Herzegovina (BIH-EUR), identified by applying Leishmania Internal Transcribed Spacer Reverse (LITSR) and Leishmania 5.8S ribosomal RNA (L5.8S) primer pairs []. The other study of US troops deployed to Iraq (EMR) identified based on TaqMan specific primers []. The third study detected anti- Leishmania antibodies using RIDASCREEN Leishmania AB ELISA in Austrian troops deployed to Kosovo []. Meta-analysis of the Freeman-Tukey transformed proportions indicated an overall cumulative asymptomatic Leishmania infection incidence of 11% (95% CI: 6, 17) (). The highest cumulative incidence value was for Iraq (EMR) (20% [95% CI: 14, 25]), with significant inter-study heterogeneity ( I 2 = 78.47%) (), and evidence of publication bias ( z = -2.64, p <0.001) (Fig B in ).

Thirteen studies reported on the effectiveness of PPMs and associated risk of developing leishmaniasis, of which eight studies could be pooled for meta-analysis [,,,–,,]. The majority of the studies reported risk factors for developing symptomatic CL. Considering the combined influences of up to six reported PPMs, the odds of leishmaniasis was reduced by 35% (OR = 0.65; 95% CI = 0.42, 0.89; P <0.001), with moderate heterogeneity ( I 2 = 66.86%) between studies (). The regular use of military-issued Long-Lasting Insecticidal Nets (LLINs) indicates a 49% reduction in the odds of leishmaniasis, with a statistically significant protective effect and no heterogeneity (OR = 0.51; 95% CI = 0.24, 0.77; P <0.001) between the four studies. The regular use of long-sleeved clothing (OR = 0.72; 95% CI = 0.26, 1.17; P <0.001 I 2 = 74.41%]), insect repellent (OR = 0.40; 95% CI = -0.23, 1.03; P = 0.21 I 2 = 81.79%), and insecticide treated uniforms (ITUs) (OR = 0.22; 95% CI = -0.36, 0.80; P = 0.46 I 2 = 0.00%) may reduce the odds of leishmaniasis by 28%, 60% and 78%, respectively, however the latter two PPMs were not individually significantly different from controls (). Knowledge about leishmaniasis (OR = 1.07; 95% CI = 0.47, 1.66; P <0.001 I 2 = 0.00%) and sleeping in protected accommodation (OR = 1.34; 95% CI = 0.36, 2.33; P = 0.01 I 2 = 71.48%), were both significantly associated with reduced odds of leishmaniasis (). Funnel plots and Egger’s test showed no evidence of publication bias for these evaluated PPMs (Fig C and Table L in ).

Significant heterogeneity ( I 2 > 75%) was detected in some meta-analyses (–); however, as there were few studies, further analyses were not performed to identify differentiating factors. The Sensitivity analyses performed for the above analyses by excluding individual studies with replacement did not identify particular studies that significantly influenced the overall direction of association, effect size, statistical significance, or level of heterogeneity (Tables E–H in ). Egger’s test results indicated evidence of small-study effects and publication bias in the analyses of asymptomatic Leishmania infections ( z = -2.64, p <0.001), but not in analyses of symptomatic CL ( z = 1.85, p = 0.065), or the effectiveness of PPMs and associated risk of developing CL (Figs A-C and Table L in ).

A narrative synthesis of symptomatic VL is provided as quantitative data to conduct a meta-analysis were not reported. Descriptions of Leishmania and sand fly species identifications, and troop immune responses to sand fly salivary antigen exposure, are also provided below.

Thirteen cases of VL were described, all among US soldiers; eight cases deployed during Operation Desert Storm in Saudi Arabia [], and five cases deployed during Operation Iraqi Freedom and Operation Enduring Freedom in Iraq, Afghanistan and Oman [,]. The eight cases from Saudi Arabia exhibited clinical symptoms of viscerotropic leishmaniasis which is a mild form of VL and usually associated with L. tropica infections. Of these eight cases, seven exhibited symptoms of varying degrees including unexplained fever, chronic fatigue, malaise, cough, intermittent diarrhoea, abdominal pain, adenopathy, and mild transient hepato-splenomegaly, whereas none of the soldiers presented cutaneous manifestations []. Six of the eight cases were confirmed L. tropica infections. The five cases deployed to Iraq, Afghanistan and Oman all presented classical signs of VL including fever, night sweats, cough, abdominal pain, diarrhoea, anorexia, nausea, cytopenias, elevated liver-associated enzymes, splenomegaly, hepatomegaly and weight loss [,]. Leishmania amastigotes were visualised in bone marrow biopsies from two cases from Iraq and in liver biopsies from two cases from Afghanistan []. The single case from Oman was confirmed as infected with a member of the L. infantum/donovani complex[].

Eighteen studies reported the specific identification of Leishmania species using molecular techniques: eight in AMR [,,,,,,,]), seven in EMR [,,,,,,], one in EUR [], and two in SEAR [,]. The majority (15/18) of identifications were from symptomatic CL patients [,,,,,,,,,,,,,,], two from VL [,], and one from an asymptomatic infected patient []. PCR was the most used nucleic acid amplification method, including conventional PCR (cPCR), real-time PCR (qPCR), nested PCR, and kinetoplast DNA (kDNA) PCR; nucleic acid sequence-based amplification (NASBA) was used in addition. The PCR assays targeted specific genes such as 5.8S rRNA, heat shock protein (HSP70), cytochrome b (Cytb), small subunit rRNA (ssrRNA), internal transcribed spacer (ITS1 and ITS2), and mini-exon. The identified Leishmania species from infected soldier samples are summarised in .

Seven studies reported the identification of sand fly species captured in the deployment foci [,,,,,,]. Four reports were from AMR [,,,], two from EMR [,] and one from AFR []. Two studies confirmed sand fly infections with Leishmania and/or the specific Leishmania identities (). Among the four sand fly studies conducted in the AMR, three [,,] employed morphological methods to identify the sand flies, but did not screen for Leishmania . One study [] identified sand flies on morphological and molecular characteristics, and confirmed their infection with L. (V.) braziliensis by qPCR and restriction enzyme analysis. In the two studies conducted in EMR, one based sand fly identification on morphology [], the other on morphology and molecular techniques [], but neither study screened for Leishmania infection. The single study conducted in AFR [] identified the sand flies morphologically, and sand fly infection with L. major on the basis of isoenzyme profiles. The sand fly species identities and those confirmed with in situ flagellate infections are listed in .

Host antigenic immune responses to proteins in sand fly saliva deposited when blood-feeding, usually measured by an enzyme-linked immunosorbent assay (ELISA) or Western blot analysis, are indicative of host exposure to sand fly bites. Four studies, all involving soldiers deployed to Iraq, measured their IgG antibody responses to sand fly whole salivary gland homogenate (SGH) antigens, two studies using Phlebotomus (Ph). papatasi SGH [,] and two studies using Ph. alexandri SGH [,]). Antibody levels were significantly higher in troops post-deployment compared to pre-deployment, and significantly higher in those with parasitologically confirmed CL infections (cases) versus negative-CL diagnosed controls []. The salivary proteins in Ph. papatasi SGH frequently recognised were MW30 and MW64 [], MW38 and 14 kDa []. Others included MW12/14, 15, 18, 26, 28, 32, 36, 42, 44, 46, and 52. Higher levels of IFN-γ, IL-6, IL-13, IL-10 and IL-17 cytokines were detected in soldiers exposed to Ph. alexandri SGH than in controls; highest levels were observed in SGH-positive asymptomatic soldiers[]. Usually, these immune responses decline soon after repatriation to non-endemic regions, but with some exceptions [].

CL is the most common form of leishmaniasis reported among deployed military personnel, similar to in civilian populations [–]. The meta-analysis of 12 estimates from eight studies showed a mean cumulative incidence of 10% (), with a potential higher cumulative incidence in EMR countries (14%) compared to AFR (8%) or AMR (9%), though statistical differences between WHO regions were not detected. Notwithstanding, a high degree of heterogeneity (97.8%) was observed between estimates, as also detected between the seven estimates (four countries: Belize, French Guiana, Panama, and Peru) within the AMR region (97.9%) (Table D in ). The possibility to pool estimates by aetiological agent(s) rather than region did not substantially alter the mean cumulative incidence value for symptomatic CL: L. ( V .) braziliensis and L. (L) mexicana , being the most common, occur in AMR () whereas the limited data excluded further pooling by aetiology. As such, the calculated incidence estimates presented here should be treated as preliminary indicators intended to lay a foundation for military health planning and to stimulate future generation of context-specific comparable data.

Heterogeneities within and between regions were not unexpected not least due to their geographical diversity, differences in aetiology and epidemiology, vector species’ phenology, blood-feeding behaviour and vectorial competence, reservoirs, and timing of troop deployment relative to the climatic and environment conditions governing sand fly abundance [,,–]. In AMR countries, CL results from numerous Leishmania species, transmitted by multiple known and putative vector species via a range of forest- and peridomestic-dwelling mammals (and possibly domestic animals) as hosts [,–]. Current knowledge of the potential multiple transmission cycles in this region is poor. Human CL also occurs in the Old World due to L. major and L. tropica in EMR and AFR regions through better defined distinct zoonotic and anthroponotic cycles, respectively.

Sand fly abundance and biting intensities are generally climate-sensitive particularly to temperature and humidity [,–]. Greater sand fly vector abundance is variably reported in dry or wet seasons and related to clinical CL incidence [,,]. In the current study, the cumulative incidence of symptomatic CL was not statistically dissimilar between wet (10%) and dry (10%) seasons, though it should be acknowledged that temporal correlations between vector abundance and infection incidence are complex to interpret; the precise timing of inoculation is unknown, and infection prepatent periods are highly variable []. Indeed, “season” is a crude measure of the multiple climatic variables influencing sand fly population infection rates. Demography can also play a role: vector populations may be more infectious with increasing age, i.e., when the parous rate is highest- often at the end of the sand fly season []. For shorter durations of troop deployment, the timing of transmission may be more accurately attributed to a particular period or season. suggests that leishmaniasis outbreaks primarily occur during the wet season in the New World, but during the dry season in the Old World. These trends are generally consistent with data reported in civilian populations [], likely reflecting physiological and behavioural adaptations to local or regional environmental and climatic conditions [,–]. For example, sand fly vectors in European states, tend to be most active in temperatures >15°C, but optimal temperature and humidity requirements for sand fly metabolism, growth and survival, and Leishmania development within the vector, vary considerably between species [,–]. Heavy rainfall on the other hand, kills immature stages.

The analysis revealed differences in CL incidence with military deployment durations. The reason for the higher incidence in those deployed for 1–4 months compared to those lasting less than one month, or more than four months, is unclear. Speculatively, this may stem from heightened exposure during 1–4 months of deployment when personnel are less adapted to protective measures, in contrast to longer deployments giving more time to adhere to PPMs and make operational adjustments to reduce exposure. Future studies are needed to tease out such relationships.

The cumulative incidence of subclinical Leishmania infections was 11% across the studies, but again with significant inter-study heterogeneity (I2 = 78.47%). The identified infecting Leishmania species included the L. donovani/infantum complex and L. tropica among troops deployed to the EMR and EUR [,], with the highest incidence in Iraq (20%), compared to an incidence of <10% for L. donovani/infantum complex in Lebanon and BIH (). Whilst L. donovani and L. infantum both typically cause VL, they differ in transmission patterns and geographical distribution. Leishmania donovani , found predominantly in the Indian subcontinent and East Africa [,], is anthroponotic, and transmitted by clinical cases as demonstrated by xenodiagnosis studies [,]. In contrast, VL due to L. infantum , is a zoonosis with domestic dogs as key reservoir, being prevalent in the Americas, Mediterranean countries, the Baltics, and parts of Central and East Asia [,–]. Interestingly, some strains of L. infantum are associated with CL development in EUR, AMR and EMR [,,]. Troops deployed to Syria and Bosnia reported lower incidences of infections with the L. donovani/infantum species complex compared to deployments to Iraq [].