Skip to main content
PMC home page
Bulletin of the World Health Organization logoLink to Bulletin of the World Health Organization
. 2023 Dec 8;102(3):204–215. doi: 10.2471/BLT.23.290424

The impact of mosquito sampling strategies on molecular xenomonitoring prevalence for filariasis: a systematic review

Impact des stratégies d’échantillonnage de moustiques sur la prévalence de la xénosurveillance moléculaire de la filariose: revue systématique

Impacto de las estrategias de muestreo de mosquitos en la prevalencia del xenomonitoreo molecular de la filariasis: una revisión sistemática.

تأثير استراتيجيات جمع عينات البعوض على انتشار الرصد الخارجي الجزيئي لداء الفيلاريات: مراجعة منهجية

蚊虫采样策略对分子异种监测的患病率的影响：系统评价

Влияние стратегий отбора образцов комаров на выявление распространенности филяриоза методом молекулярного ксеномониторинга: систематический обзор

Lisa J Reimer a,✉, Joseph D Pryce a
PMCID: PMC10898278  PMID: 38420575

Abstract

Objective

To explore the impact of mosquito collection methods, sampling intensity and target genus on molecular xenomonitoring detection of parasites causing lymphatic filariasis.

Methods

We systematically searched five databases for studies that used two or more collection strategies for sampling wild mosquitoes, and employed molecular methods to assess the molecular xenomonitoring prevalence of parasites responsible for lymphatic filariasis. We performed generic inverse variance meta-analyses and explored sources of heterogeneity using subgroup analyses. We assessed methodological quality and certainty of evidence.

Findings

We identified 25 eligible studies, with 172 083 mosquitoes analysed. We observed significantly higher molecular xenomonitoring prevalence with collection methods that target bloodfed mosquitoes compared to methods that target unfed mosquitoes (prevalence ratio: 3.53; 95% confidence interval, CI: 1.52–8.24), but no significant difference compared with gravid collection methods (prevalence ratio: 1.54; 95% CI: 0.46–5.16). Regarding genus, we observed significantly higher molecular xenomonitoring prevalence for anopheline mosquitoes compared to culicine mosquitoes in areas where Anopheles species are the primary vector (prevalence ratio: 6.91; 95% CI: 1.73–27.52). One study provided evidence that reducing the number of sampling sites did not significantly affect molecular xenomonitoring prevalence. Evidence of differences in molecular xenomonitoring prevalence between sampling strategies was considered to be of low certainty, due partly to inherent limitations of observational studies that were not explicitly designed for these comparisons.

Conclusion

The choice of sampling strategy can significantly affect molecular xenomonitoring results. Further research is needed to inform the optimum strategy in light of logistical constraints and epidemiological contexts.

Introduction

Lymphatic filariasis is a disabling and debilitating disease caused by the filarial worms Wuchereria bancrofti, Brugia malayi or B. timori that are transmitted by mosquitoes of the genera Culex, Anopheles, Aedes and Mansonia. The disease is targeted for elimination using mass drug administration and vector control; many countries have already achieved the goal of elimination as a public health problem. The World Health Organization (WHO) recommends that countries continue disease surveillance using cross-sectional surveys, routine surveillance of target populations, or molecular xenomonitoring to ensure infection levels remain below target thresholds or to confirm interruption of transmission.1

Molecular xenomonitoring is used as a surveillance strategy for vector-borne diseases such as lymphatic filariasis and onchocerciasis. The technique detects the presence of pathogen genetic material such as deoxyribonucleic acid (DNA) in disease vectors (e.g. mosquitoes). This method therefore gives a measure of the vector population’s exposure to pathogens picked up from infected humans, allowing it to serve as a proxy for the presence of human disease.

Molecular xenomonitoring overcomes many of the key challenges of lymphatic filariasis case surveillance: it does not rely on human blood sampling, it is relatively inexpensive and it allows integrated surveillance of multiple diseases.2 Innovations in mosquito trap design and field-friendly amplification and detection techniques are now bringing molecular xenomonitoring into the reach of control programmes, even those that lack specialist entomology training.3 A meta-analysis showed that molecular xenomonitoring had a high sensitivity at low microfilaria prevalence in communities, and demonstrated a strong correlation between molecular xenomonitoring and microfilaria prevalence when a consistent method is applied.4

A major limitation of molecular xenomonitoring is that there is no standardized protocol for sampling mosquitoes.3 Current WHO guidelines on molecular xenomonitoring5 indicate that any collection method can be used, and provide minimal instruction on frequency, scale, target species and sample sizes. However, each of these variables may influence the likelihood of a molecular xenomonitoring survey detecting mosquitoes positive for filarial DNA. Consequently, standardized guidelines, along with those tailored to specific settings, are needed. These guidelines will ensure that collection strategies effectively detect areas of potential disease transmission, and that results are comparable across timepoints and evaluation units. Developing such guidelines for sampling mosquitoes requires an understanding of how collection strategies influence the prevalence of filarial DNA in wild-caught mosquitoes.

Several mosquito collection methods can be used for lymphatic filariasis xenomonitoring.5 Each method exploits a specific stage of the mosquito’s gonotrophic cycle and therefore predominantly collects, though not exclusively, mosquitoes from that stage. In this review we have broadly categorized these methods into three groups: (i) fed collection methods, which target mosquitoes that have recently fed on blood. These methods use indoor resting catches (in which mosquitoes are collected either by direct aspiration or through the use of insecticides), or exit traps (traps fixed to windows to collect mosquitoes as they attempt to leave a building); (ii) gravid collection methods, which lure and trap gravid, ovipositing females using gravid traps; and (iii) unfed collection methods, which trap mosquitoes by exploiting their host-seeking behaviour. These methods use light traps (with or without carbon dioxide), odour-baited traps (such as the Biogents’ BG-Sentinel trap, Regensburg, Germany) or human landing catches (in which mosquitoes are caught as they alight on a human collector).

As the presence of parasite DNA in the mosquito is dependent on a previous bloodmeal from a lymphatic filariasis-infected host, bloodfed or gravid mosquitoes have a higher likelihood of containing filarial DNA than unfed mosquitoes.

A second factor that may affect the molecular xenomonitoring prevalence is the intensity of the sampling protocol. Lymphatic filariasis is a highly focal disease,6 and local transmission is influenced by a combination of environmental, climatic and socioeconomic conditions. Small hotspots of high transmission can persist in districts that have very low transmission overall.7,8 The likelihood of detecting a positive mosquito may increase with a higher number of sampling locations.

A third factor affecting molecular xenomonitoring prevalence is the predominant genus of mosquitoes collected. Molecular xenomonitoring prevalence is likely to be higher for mosquitoes that act as vectors for the disease than those that do not. In competent vectors, microfilariae that have been ingested develop into infective stage larvae, a process that takes a minimum of 10–12 days, so parasite DNA may be detectable for the remainder of the mosquito’s life.9,10 In non-competent vectors, parasite DNA is transient and often expelled within 48 hours of bloodmeal ingestion.11 Another consideration is bloodmeal size; Culex quinquefasciatus can ingest twice as many microfilariae as Aedes aegypti under experimental conditions.12 Anthropophilic (human-seeking) vectors, such as Anopheles gambiae, are also likely to have greater exposure to filarial DNA than those that feed from a variety of hosts.

In this systematic review we aimed to determine how sampling strategy affects molecular xenomonitoring prevalence and informs molecular xenomonitoring implementation. We compared molecular xenomonitoring prevalence when measured using two or more different methods from within the following three categories: (i) mosquito collection methods: fed, gravid and unfed; (ii) sampling intensity; and (iii) mosquito genera: Anopheles, Culex, Mansonia, Aedes and Armigeres.

Methods

We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines13 using pre-determined methods from a protocol registered with the PROSPERO international database of prospectively registered systematic reviews (registration number CRD42020200351).14

Eligibility criteria

Studies were included in our review if they met all of the following criteria: (i) they used two or more collection strategies; (ii) wild mosquito populations were sampled; and (iii) they used molecular methods (polymerase chain reaction (PCR), genetic sequencing or loop-mediated isothermal amplification) to both measure and report the molecular xenomonitoring prevalence for the causative agents of lymphatic filariasis (W. bancrofti, B.malayi, B. timori). We did not apply any language restrictions to the inclusion criteria.

Search

We searched five bibliographic databases (CINAHL Complete, eBook Collection, Global Health, Global Health Archive and MEDLINE Complete) for all records up to and including 8 February 2023 using the EBSCOhost research platform.15

We also checked the reference lists in included studies to identify further studies meeting the inclusion criteria. The search strategy is outlined in Box 1. Two reviewers assessed abstracts and selected papers for full-text screening. We resolved discrepancies by thorough review and discussion.

Box 1. Search terms used in the systematic review on the impact of mosquito sampling strategies on molecular xenomonitoring prevalence for filariasis.

((Xenosurveillance OR Xeno-surveillance) OR (Xenomonitor* OR Xeno-monitor*) OR

((“Molecular screen*” OR “Molecular diagnos*” OR PCR OR “Polymerase chain reaction” OR sequencing OR LAMP)) AND (Mosquito* OR Aedes OR Anopheles OR Culex OR Mansonia))) AND (Onchocerc* OR “River blindness” OR Filaria* OR Elephantiasis OR “Wucheria bancrofti” OR “W.°bancrofti” OR “Brugia malayi” OR “B.°malayi” OR “Brugia timori” OR “B.°timori” OR Loa OR Loiasis OR “African eye worm” OR Mansonel*

Note: This search strategy was used for another research project which identified molecular xenomonitoring studies for any disease caused by filarial worms.4,16 Specific inclusion criteria for this review restricted studies to those relating only to lymphatic filariasis.

Data extraction

We extracted study information at the smallest available level, e.g. individual villages within a district. Where molecular xenomonitoring prevalence and associated 95% confidence intervals (CIs) were not reported, we calculated these using data reported in the study. Where mosquitoes were screened in pools, we estimated the molecular xenomonitoring prevalence and 95% CI from the total number of mosquitoes screened, the size of the pools and the proportion of pools that were positive using the Poolscreen algorithm, version 2.0.17 If the sizes of individual pools were not reported, we used the mean pool size for Poolscreen calculations.

Methodological quality assessment

We used the QUADAS–2 tool to assess the methodological quality of the included studies.18 The tool (available in the online repository)19 was adapted for evaluation of community-level surveillance methods before the start of the screening process.

Statistical analysis and data synthesis

We calculated prevalence ratios with 95% CIs using RevMan 5 (Cochrane, London, United Kingdom of Great Britain and Northern Ireland) to compare molecular xenomonitoring prevalence between different sampling strategies. Because pooled mosquito data have a higher level of uncertainty than the same number of samples screened individually, we transformed the reported samples sizes to effective samples sizes (available in the online repository)19 to ensure RevMan calculated the correct CIs.

We conducted all meta-analyses using generic inverse variance models in RevMan5. We performed fixed effects meta-analyses if heterogeneity was absent or moderate (I2 < 60%), and random effects meta-analyses if heterogeneity was considerable or substantial (I2 ≥ 60%).20

We performed subgroup analyses by filarial parasite species, primary vector genus and trapping methods used to explore reasons for substantial heterogeneity. To assess significant differences between subgroups, we used χ2-squared tests with P-values less than 0.1 deemed statistically significant. We also performed sensitivity analyses to evaluate the effect of exclusion of trials that had a high risk of bias for any of the QUADAS-2 domains. Where 10 or more studies were included in a meta-analysis, we investigated the risk of publication bias using funnel plots.21

We assessed the certainty that the true differences between sampling methods lie close to those estimated by our meta-analyses using the Grading of Recommendations Assessment, Development and Evaluation approach.22 As all the included studies were observational studies, the evidence for each outcome started as low certainty; this grade could be further downgraded due to concerns about any of the following five domains: risk of bias; imprecision; inconsistency; indirectness; and reporting bias.23

Results

We identified 1142 records through electronic database searching, and three records through contact with study authors. After removal of duplicates, we screened 407 records. Of these, we identified 107 records for full-text assessment and 25 of these met the inclusion criteria.3,24–47 These were included in qualitative and quantitative synthesis (Fig. 1).

Fig. 1.

Fig. 1

Open in a new tab

Flowchart showing the selection of studies included in the systematic review on the impact of mosquito sampling strategies on molecular xenomonitoring prevalence for filariasis

Characteristics of included studies

Of the included studies, 12 studies were conducted in the WHO African Region, eight in the South-East Asia Region, two each in the Western Pacific Region and in the Region of the Americas, and one in the Eastern Mediterranean Region. The study characteristics are summarized in Table 1. A total of 172 083 mosquitoes were analysed across the included studies with a mean of 6373 per study (range: 208–57 357).

Table 1. Characteristics of included studies in the systematic review on the impact of mosquito sampling strategies on molecular xenomonitoring prevalence for filariasis.

Study Study location (country or territory) Setting Parasite of interest Primary vector No. of mosquitoes Comparison
Comparison 1: collection methods
Bockarie et al., 200024 Papua New Guinea Rural W. bancrofti An. punctulatus 621 Human landing catch versus light trap
Coulibaly et al., 202225 Mali Rural W. bancrofti An. gambiae 1364 Human landing catch versus Ifakara tent trap versus Biogents’ sentinel trap
Hoti et al., 200226 India Rural W. bancrofti Cx. quinquefasciatus 4940 Human landing catch versus indoor resting catch
Irish et al., 201527 United Republic of Tanzania Rural W. bancrofti Cx. quinquefasciatus 5737 Gravid trap versus light trap
Njenga et al., 202228 Kenya Rural W. bancrofti An. gambiae s.l. and Cx. quinquefasciatus 3652 Ifakara tent trap versus light trap
Opoku et al., 201829 Ghana Rural W. bancrofti An. gambiae s.l. and An. funestus 734 Gravid trap versus light trap versus Biogents’ sentinel trap versus exit trap versus indoor resting catch
Owusu et al., 201530 Ghana Rural W. bancrofti An. gambiae s.l. 4500 Gravid trap versus indoor resting catch
Pam et al., 201731 Nigeria Urban W. bancrofti An. gambiae s.l. 10 528 Gravid trap versus exit trap versus indoor resting catch
Pryce et al., 202232 Cameroon Rural W. bancrofti An. gambia s.l. 376 Indoor resting catch versus Biogents’ sentinel trap
Ramesh et al., 20183 Brazil Urban W. bancrofti Cx. quinquefasciatus 856 Indoor resting catch versus light trap
Comparison 2: sampling intensity
Rao et al., 201633 Sri Lanka Rural W. bancrofti Cx. quinquefasciatus 57 357 300 versus 150 versus 75 trap locations
Comparison 3: mosquito genera
de Souza et al., 201434 Sierra Leone and Liberia Urban W. bancrofti An. gambiae s.l. 16 073 Anopheles versus Culex
Dyab et al., 201635 Egypt Both W. bancrofti Cx. pipiens 1600 Aedes versus Anopheles versus Culex
Entonu et al., 202036 Nigeria Rural W. bancrofti An. gambiae 3000 Anopheles versus Culex versus Aedes
Fischer et al., 200237 Indonesia Rural B. timori An. barbirostris 1266 Anopheles versus Culex
Kouassi et al., 201538 Guinea Urban W. bancrofti An. gambiae s.l. 3747 Anopheles versus Culex
Lupenza et al., 202139 United Republic of Tanzania Rural W. bancrofti Cx. quinquefasciatus 7346 Culex versus Anopheles
McPherson et al., 202240 Samoa Rural W. bancrofti Ae. polynesiensis 8506 Aedes versus Culex
Mulyaningsih et al., 201941 Indonesia Rural B. malayi Unknown. Typically Mansonia spp. 1280 Armigeres versus Culex versus Mansonia
Nirwan et al., 202242 Indonesia Rural W. bancrofti, B. malayi, B. timori Unknown in this location 3907 Armigeres versus Culex versus Mansonia versus Aedes
Njenga et al., 202228 Kenya Rural W. bancrofti An. gambiae s.l. and Cx. quinquefasciatus 3652 Anopheles versus Culex versus Mansonia vs Aedes
Nurjana et al., 202043 Indonesia Rural B. malayi Unknown in this location 2989 Culex versus Mansonia versus Armigeres versus Aedes versus Anopheles
Pryce et al., 202232 Cameroon Rural W. bancrofti An. gambiae s.l. 350 Anopheles versus Culex
Ridha et al., 202044 Indonesia Rural W. bancrofti and B. malayi Unknown in this location 802 Anopheles versus Culex versus Mansonia
Schmaedick et al., 201445 American Samoa Rural W. bancrofti Ae. polynesiensis 17 448 Aedes versus Culex
Supriyono & Tan, 202046 Indonesia Rural B. malayi Unknown. Typically Mansonia spp. 208 Aedes versus Anopheles versus Culex versus Mansonia
Yokoly et al., 202047 Côte d'Ivoire Rural W. bancrofti An. gambiae s.l. 9244 Anopheles versus Culex
Open in a new tab

Ae.: Aedes; An.: Anopheles; B.: Brugia; Cx.: Culex; s.l.: sensu lato; spp.: species.

Note: Biogents AG is located in Regensburg, Germany.

Methodological quality

We provide a summary of the methodological quality in Table 2. The sampling methods used in two studies were judged to introduce a high risk of bias.26,39 One study26 screened mosquitoes caught by the indoor resting catch method in pools of 10, and those caught by human landing catch method in pools of 50; their study also reported reduced sensitivity of molecular detection in pool sizes of 50 compared to microscopic detection following dissection. This approach prevents an unbiased comparison of mosquitoes collected by the two catch methods. Another study39 used light traps to preferentially collect Anopheles mosquitoes and gravid traps to preferentially collect Culex mosquitoes, and they presented their data according to species. The comparisons we have made between species are likely confounded by the fact that gravid traps collect mosquitoes that have taken at least one bloodmeal.

Table 2. Methodological quality assessment of studies included in the systematic review on the impact of mosquito sampling strategies on molecular xenomonitoring prevalence for filariasis.

Study Sampling site selection risks Sampling site selection applicability Sampling methods risk Sampling methods applicability Flow and timing risks
  Comparison 1: collection methods
Bockarie et al., 200024 Low Low Unclear Low Low
Coulibaly et al., 202225 Low Low Low Low Low
Hoti et al., 200226 Unclear Low High Low Unclear
Irish et al., 201527 Low Low Low Low Low
Njenga et al., 202228 Low Low Low Low Low
Opoku et al., 201829 Unclear Low Low Low Low
Owusu et al., 201530 Low Low Unclear Low Unclear
Pam et al., 201731 Low Low Unclear Low Unclear
Pryce et al., 202232 Low Unclear Unclear Low Low
Ramesh et al., 20183 Low Low Low Low Low
  Comparison 2: sampling intensity
Rao et al., 201633 Low Low Low Low Low
  Comparison 3: mosquito genera
de Souza et al., 201434 Low Low Low Low Low
Dyab et al., 201635 Unclear Low Low Low Low
Entonu et al., 202036 Low Low Low Low Low
Fischer et al., 200237 Unclear Low Low Low Low
Kouassi et al., 201538 Low Low Unclear Low Low
Lupenza et al., 202139 Low Low High Low Low
McPherson et al., 202240 Low Low Low Low Low
Mulyaningsih et al., 201941 Unclear Low Low Low Low
Nirwan et al., 202242 Low Low Low Unclear Low
Njenga et al., 202228 Low Low Low Low Low
Nurjana et al., 202043 Low Low Low Low Low
Pryce et al., 202232 Low Unclear Unclear Low Low
Ridha et al., 202044 Unclear Low Low Unclear Low
Schmaedick et al., 201445 Low Low Low Low Low
Supriyono & Tan, 202046 Unclear Low Low Low Low
Yokoly et al., 202047 Unclear Low Unclear Low Low
Open in a new tab

Note: Some articles are included in more than one comparison and hence appear twice in the table.

Summary of findings

Molecular xenomonitoring prevalence was significantly higher when fed collection methods were used compared with unfed. Higher prevalence was also observed in anopheline mosquitoes than culicine mosquitoes. However, the certainty of the evidence scored very low (Table 3). A summary of the assessments and the justifications for each rating are also provided in Table 3. Below, we discuss in detail the data underpinning each of these assessments.

Table 3. The relative effect of different mosquito sampling strategies on molecular xenomonitoring prevalence and evaluations of the certainty of the evidence.

Comparison Relative effect, prevalence ratio (95% CI) No. of mosquitoes (studies) Quality of the evidence Comments
Comparison 1: collection methods
Fed versus gravid collection methods 1.54 (0.46–5.16) 12 711 (3) Very lowa,b,c Downgraded for imprecision, inconsistency and indirectness
Fed versus unfed collection methods 3.53 (1.52–8.24) 5 167 (2) Very lowd,e Downgraded for indirectness and risk of bias
Gravid versus unfed collection methods 0.20 (0.01–3.40) 5 927 (1) Very lowf,g Downgraded for serious imprecision and indirectness
Comparison 2: sampling intensity
300 versus 150 trapping locations 0.89 (0.54–1.46) 29 797 (1) Very lowg Downgraded for indirectness
300 versus 75 trapping locations 0.90 (0.54–1.51) 28 624 (1) Very lowg Downgraded for indirectness
150 versus 75 trapping locations 1.01 (0.60–1.68) 27 921 (1) Very lowg Downgraded for indirectness
Comparison 3: mosquito genera
Anopheles versus Culex mosquitoes:
  Anopheles areas 6.91 (1.73–27.52) 28 974 (6) Very lowb,h Downgraded for inconsistency and imprecision
  Culex areas 2.68 (0.08–94.93) 53 610 (2) Very lowa Downgraded for imprecision
Aedes versus Culex mosquitoes 1.07 (0.52–2.19) 64 705 (2) Very lowa,i Downgraded for imprecision and indirectness
Open in a new tab

CI: confidence interval.

a Downgraded for imprecision: the 95% CIs include both no difference and a large difference.

b Downgraded for inconsistency: there was substantial heterogeneity that could not be explained by subgroup analyses.

c Downgraded for indirectness: all studies were conducted in areas where the primary filariasis vectors belonged to the Anopheles gambiae species complex; the findings may not apply to other transmission areas.

d Downgraded for indirectness: over 92% of the weight of the meta-analysis came from a single study; the findings may not apply to other transmission areas.

e Downgraded for risk of bias: the sampling methods for the main study contributing to the analysis were judged to have a high risk bias as mosquitoes collected using fed traps were screened in smaller pool sizes than those collected by gravid traps; this may favour higher molecular xenomonitoring prevalence for mosquitoes from fed collection methods.

f Downgraded twice for serious imprecision: the 95% CIs include large differences in favour of each type of collection method.

g Downgraded for indirectness: a single study contributed to the meta-analysis; the findings may therefore not apply to other transmission areas.

h Downgraded for imprecision: although the 95% CIs are extremely wide and include both a small difference and a very large difference.

i Downgraded for indirectness: studies were conducted in American Samoa and Samoa; the findings may not apply to other transmission areas.

Note: We used the Grading of Recommendations Assessment, Development and Evaluation approach for grading the quality of the evidence.22

Comparison 1: collection methods

Fed versus gravid

For this subgroup analysis, we included three studies.29–31 Higher molecular xenomonitoring prevalence was observed for mosquitoes collected using fed traps than for mosquitoes collected using gravid traps, but this difference was not statistically significant (prevalence ratio: 1.54; 95% CI: 0.46–5.16; Fig. 2). There was substantial heterogeneity between studies (I2: 61%). A subgroup analysis by the specific type of trapping method used did not explain this heterogeneity.

Fig. 2.

Effect of mosquito collection method, fed versus gravid, on molecular xenomonitoring prevalence

CI: confidence interval.

Note: We used a random-effects meta-analysis.

Fig. 2

Open in a new tab

Fed versus unfed

Analysis from two studies showed that molecular xenomonitoring prevalence wasapproximately 3.5 times higher for mosquitoes collected using fed traps than for mosquitoes collected using unfed traps (prevalence ratio: 3.53; 95% CI: 1.52–8.24; Fig. 3).26,29 Note that 92.7% of the weight of the meta-analysis came from a single study26 that was considered at high risk of bias. A sensitivity analysis excluding this study suggested there was no difference between the two collection methods.

Fig. 3.

Effect of mosquito collection method, fed versus unfed, on molecular xenomonitoring prevalence

CI: confidence interval.

Note: We used a fixed-effects meta-analysis.

Fig. 3

Open in a new tab

Gravid versus unfed

Only one site provided evidence from sufficiently large effective sample sizes to contribute to the meta-analysis.27 This reported a higher molecular xenomonitoring prevalence for mosquitoes collected using unfed collection methods than for mosquitoes collected using gravid collection methods; however, this finding was not statistically significant (prevalence ratio: 0.20; 95% CI: 0.01–3.40).

Comparison 2: sampling intensity

None of the included studies provided a comparison of different longitudinal collection intensities (for example, nightly collections versus monthly collections). One study compared molecular xenomonitoring prevalence based on different densities of trapping locations (300 versus 150 versus 75 locations) on W. bancrofti detection rates.33 The study found that sampling mosquitoes from 300 locations did not lead to higher molecular xenomonitoring prevalence than when sampling the same number of mosquitoes from 75 sampling locations (prevalence ratio: 0.90; 95% CI: 0.54–1.51). We did not observe any difference in molecular xenomonitoring prevalence between any of the three sampling strategies.

Comparison 3: mosquito genera

Anopheles versus Culex

Eight studies, from 11 study sites, provided comparisons of molecular xenomonitoring prevalence for Anopheles and Culex mosquitoes.30,31,34,35,37–39,47

In almost all included studies, the numbers of Culex mosquitoes collected far outweighed the number of Anopheles mosquitoes. In areas where the primary vector is Anopheles, molecular xenomonitoring prevalence was approximately seven times higher for Anopheles mosquitoes than for Culex mosquitoes (prevalence ratio: 6.91; 95% CI: 1.73–27.52; Fig. 4). In areas of Culex-transmitted lymphatic filariasis, the molecular xenomonitoring prevalence was also higher for Anopheles mosquitoes, although the CI for this estimate was wide (prevalence ratio: 2.68; 95% CI: 0.08–94.93). In Anopheles areas, there was substantial heterogeneity between studies (I2: 73%) which was not explained by subgroup analyses. The high heterogeneity was due to several studies showing very large differences between mosquito genera. One study in Indonesia reported a much higher prevalence of B. timori DNA in Anopheles mosquitoes than Culex mosquitoes, resulting in a prevalence ratio of 172.7.37 Three additional studies also provided study areas with a prevalence ratio of 20 or greater.30,31,38

Fig. 4.

Effect of collected mosquito genera on molecular xenomonitoring prevalence

CI: confidence interval.

Note: We used a random-effects meta-analysis.

Fig. 4

Open in a new tab

Aedes versus Culex

Two studies from American Samoa and Samoa, comprising 11 study sites, provided comparisons for a meta-analysis of molecular xenomonitoring prevalence between Aedes and Culex mosquitoes.40,45 There was no difference between the two genera (prevalence ratio: 1.07; 95% CI: 0.52–2.19).

Other mosquito genera

Six studies provided comparisons between other mosquito genera including Armigeres and Mansonia species.28,41–44,46 The limited number of studies contributing to each comparison and the number of mosquitoes positive for filarial DNA in each study precluded a quantitative synthesis for this outcome.

Discussion

Our findings suggest that mosquito collection methods can have an important impact on molecular xenomonitoring prevalence, although precise estimates of the impact were difficult to obtain.

Multiple studies conducted in Anopheles-transmitted lymphatic filariasis areas reported substantially higher molecular xenomonitoring prevalence when targeting bloodfed mosquitoes than gravid mosquitoes. However, this effect was not consistent between studies and we were unable to determine the reasons for this heterogeneity. The lack of studies from areas where the primary vector is Culex is an important gap. Gravid traps are an efficient tool for collecting Culex mosquitoes, and have been used as the sole collection method for molecular xenomonitoring by elimination programmes in Bangladesh and Sri Lanka.2,33

Our meta-analysis showed a large difference in molecular xenomonitoring prevalence between fed and unfed collection methods. However, most of the weight of this analysis was contributed by a single study in an area of Cx. quinquefasciatus-transmitted lymphatic filariasis,26 and the applicability of this result to other transmission settings is limited. While it is logical that molecular xenomonitoring prevalence would be higher when targeting recently bloodfed mosquitoes, there remains uncertainty from the available data that this effect will always be observed.

Strong evidence was reported by one high quality study in Sri Lanka that a reduced number of sampling sites per evaluation unit, from 300 to 75, did not lead to reduced detection of W. bancrofti DNA in Cx. quinquefasciatus. This finding, observed in two post-transmission assessment survey communities, supports the feasibility of molecular xenomonitoring, although evidence from other areas will be required to determine whether this approach is applicable to other transmission zones.

Lymphatic filariasis programmes looking for the most sensitive approach for detecting Anopheles-transmitted lymphatic filariasis may wish to consider using a sampling strategy that preferentially targets Anopheles mosquitoes, since these methods are likely to collect a higher proportion of parasite-positive mosquitoes. However, this advantage will need to be balanced against the convenience of other collection methods such as gravid traps (which are an efficient method for collecting large numbers of Culex mosquitoes).9 Historically, the collection of bloodfed Anopheles mosquitoes has depended on indoor collections using aspirators or pyrethrum spray – an approach that is labour-intensive and typically results in modest collection numbers.2 Given the significantly higher molecular xenomonitoring prevalence for Anopheles mosquitoes, programmatic use of molecular xenomonitoring for the detection of ongoing cases of lymphatic filariasis in extremely low prevalence areas may depend upon the development of new tools that are efficient at collecting bloodfed or gravid Anopheles mosquitoes.

We assume that the primary explanation for variation in sensitivity between studies is the differences in the sampling strategy. However, primer and probe design, as well as the equipment used, can affect PCR results, and direct comparisons between molecular xenomonitoring methods have shown variation in sensitivity.48,49

There are narrative reviews on the status of molecular xenomonitoring. For example, one review50 highlighted the need for systematic methods and new WHO guidelines to be developed to supplement post-validation surveillance. Another review51 proposed that molecular xenomonitoring has enormous potential for the surveillance of vector-borne diseases, with the capacity for it to replace (rather than supplement) current human surveillance strategies. However, they identified several key barriers that must be overcome, including the development of protocols that account for heterogeneity in pathogen infection rates both within the mosquito and the human population.

Our review provides evidence supporting the development of standardized molecular xenomonitoring sampling protocols that specifically consider mosquito collection methods and genus. However, the certainty of evidence for every comparison is very low due to the inherent limitations of observational data, and specific concerns regarding the comparisons drawn from the available literature. Consequently, there is a need for further research in these areas to inform an optimum molecular xenomonitoring sampling strategy.

Competing interests:

None declared.

References

  • 1.Validation of elimination of lymphatic filariasis as a public health problem. Geneva: World Health Organization; 2017. Available from: https://www.who.int/publications/i/item/9789241511957 [cited 2023 Nov 22]. [Google Scholar]
  • 2.Pilotte N, Unnasch TR, Williams SA. The current status of molecular xenomonitoring for lymphatic filariasis and onchocerciasis. Trends Parasitol. 2017. Oct;33(10):788–98. 10.1016/j.pt.2017.06.008 [DOI] [PubMed] [Google Scholar]
  • 3.Ramesh A, Cameron M, Spence K, Hoek Spaans R, Melo-Santos MAV, Paiva MHS, et al. Development of an urban molecular xenomonitoring system for lymphatic filariasis in the Recife Metropolitan Region, Brazil. PLoS Negl Trop Dis. 2018. Oct 16;12(10):e0006816. 10.1371/journal.pntd.0006816 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Pryce J, Reimer LJ. Evaluating the diagnostic test accuracy of molecular xenomonitoring methods for characterizing community burden of lymphatic filariasis. Clin Infect Dis. 2021. Jun 14;72 Suppl 3:S203–9. 10.1093/cid/ciab197 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Lymphatic filariasis: a handbook of practical entomology for national lymphatic filariasis elimination programmes. Geneva: World Health Organization; 2013. [Google Scholar]
  • 6.Michael E, Singh BK, Mayala BK, Smith ME, Hampton S, Nabrzyski J. Continental-scale, data-driven predictive assessment of eliminating the vector-borne disease, lymphatic filariasis, in sub-Saharan Africa by 2020. BMC Med. 2017. Sep 27;15(1):176. 10.1186/s12916-017-0933-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Srividya A, Subramanian S, Jambulingam P, Vijayakumar B, Dinesh Raja J. Mapping and monitoring for a lymphatic filariasis elimination program: a systematic review. Res Rep Trop Med. 2019. May 27;10:43–90. 10.2147/RRTM.S134186 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Kwarteng EVS, Andam-Akorful SA, Kwarteng A, Asare DB, Quaye-Ballard JA, Osei FB, et al. Spatial variation in lymphatic filariasis risk factors of hotspot zones in Ghana. BMC Public Health. 2021. Jan 28;21(1):230. 10.1186/s12889-021-10234-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Okorie PN, de Souza DK. Prospects, drawbacks and future needs of xenomonitoring for the endpoint evaluation of lymphatic filariasis elimination programs in Africa. Trans R Soc Trop Med Hyg. 2016. Feb;110(2):90–7. 10.1093/trstmh/trv104 [DOI] [PubMed] [Google Scholar]
  • 10.Kwansa-Bentum B, Aboagye-Antwi F, Otchere J, Wilson MD, Boakye DA. Implications of low-density microfilariae carriers in Anopheles transmission areas: molecular forms of Anopheles gambiae and Anopheles funestus populations in perspective. Parasit Vectors. 2014. Apr 1;7(1):157. 10.1186/1756-3305-7-157 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Pilotte N, Cook DAN, Pryce J, Zulch MF, Minetti C, Reimer LJ, et al. Laboratory evaluation of molecular xenomonitoring using mosquito and tsetse fly excreta/feces to amplify Plasmodium, Brugia, and Trypanosoma DNA. Gates Open Res. 2020. Jun 3;3:1734. 10.12688/gatesopenres.13093.2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Albuquerque CM, Cavalcanti VM, Melo MA, Vercosa P, Regis LN, Hurd H. Bloodmeal microfilariae density and the uptake and establishment of Wuchereria bancrofti infections in Culex quinquefasciatus and Aedes aegypti. Mem Inst Oswaldo Cruz. 1999. Sep-Oct;94(5):591–6. 10.1590/S0074-02761999000500005 [DOI] [PubMed] [Google Scholar]
  • 13.Moher D, Liberati A, Tetzlaff J, Altman DG; PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med. 2009. Jul 21;6(7):e1000097. 10.1371/journal.pmed.1000097 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Welcome to PROSPERO. International prospective register of systematic reviews [internet]. York: University of York; 2023. Available from: https://www.crd.york.ac.uk/prospero/ [cited 2023 Nov 10].
  • 15.EBSCOhost research platform [internet]. Ipswich: EBSCO Information Services; 2023. Available from: https://www.ebsco.com/products/ebscohost-research-platform [cited 2023 Nov 10].
  • 16.Pryce J, Unnasch TR, Reimer LJ. Evaluating the diagnostic test accuracy of molecular xenomonitoring methods for characterising the community burden of Onchocerciasis. PLoS Negl Trop Dis. 2021. Oct 12;15(10):e0009812. 10.1371/journal.pntd.0009812 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Katholi CR, Unnasch TR. Important experimental parameters for determining infection rates in arthropod vectors using pool screening approaches. Am J Trop Med Hyg. 2006. May;74(5):779–85. 10.4269/ajtmh.2006.74.779 [DOI] [PubMed] [Google Scholar]
  • 18.Whiting PF, Rutjes AW, Westwood ME, Mallett S, Deeks JJ, Reitsma JB, et al. QUADAS-2 Group. QUADAS-2: a revised tool for the quality assessment of diagnostic accuracy studies. Ann Intern Med. 2011. Oct 18;155(8):529–36. 10.7326/0003-4819-155-8-201110180-00009 [DOI] [PubMed] [Google Scholar]
  • 19.Reimer LJ. Xenomonitoring methods. [online repository]. Charlottesville: Center for Open Science; 2023. 10.17605/OSF.IO/NRH4V [DOI] [Google Scholar]
  • 20.Deeks JJ, Higgins JPT, Altman DG. Analysing data and undertaking meta–analyses. Cochrane handbook for systematic reviews of interventions. London: The Cochrane Collaboration; 2019. pp. 241–84. 10.1002/9781119536604.ch10 [DOI] [Google Scholar]
  • 21.Sterne JAC, Egger M, Moher D, Boutron I. Chapter 10: Addressing reporting biases. In: Higgins JPT, Churchill R, Chandler J, Crumpston MS, editors. Cochrane handbook for systematic reviews of interventions, version 5.2.0. London: The Cochrane Collaboration; 2017. Available from: http:www.training.cochrane.org/handbook [cited 2022 Feb 7].
  • 22.Guyatt GH, Oxman AD, Schünemann HJ, Tugwell P, Knottnerus A. GRADE guidelines: a new series of articles in the Journal of Clinical Epidemiology. J Clin Epidemiol. 2011. Apr;64(4):380–2. 10.1016/j.jclinepi.2010.09.011 [DOI] [PubMed] [Google Scholar]
  • 23.Balshem H, Helfand M, Schünemann HJ, Oxman AD, Kunz R, Brozek J, et al. GRADE guidelines: 3. Rating the quality of evidence. J Clin Epidemiol. 2011. Apr;64(4):401–6. 10.1016/j.jclinepi.2010.07.015 [DOI] [PubMed] [Google Scholar]
  • 24.Bockarie MJ, Fischer P, Williams SA, Zimmerman PA, Griffin L, Alpers MP, et al. Application of a polymerase chain reaction-ELISA to detect Wuchereria bancrofti in pools of wild-caught Anopheles punctulatus in a filariasis control area in Papua New Guinea. Am J Trop Med Hyg. 2000. Mar;62(3):363–7. 10.4269/ajtmh.2000.62.363 [DOI] [PubMed] [Google Scholar]
  • 25.Coulibaly YI, Sangare M, Dolo H, Doumbia SS, Coulibaly SY, Dicko I, et al. Comparison of different sampling methods to catch lymphatic filariasis vectors in a Sudan savannah area of Mali. Am J Trop Med Hyg. 2022. Feb 28;106(4):1247–53. 10.4269/ajtmh.21-0667 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Hoti S, Patra KP, Vasuki V, Lizotte MW, Hariths VR, Sushma N, et al. Evaluation of Ssp I polymerase chain reaction assay in the detection of Wuchereria bancrofti infection in field–collected Culex quinquefasciatus and its application in the transmission studies of lymphatic filariasis. J Appl Entomol. 2002;126(7-8):417–21. 10.1046/j.1439-0418.2002.00635.x [DOI] [Google Scholar]
  • 27.Irish SR, Stevens WM, Derua YA, Walker T, Cameron MM. Comparison of methods for xenomonitoring in vectors of lymphatic filariasis in northeastern Tanzania. Am J Trop Med Hyg. 2015. Nov;93(5):983–9. 10.4269/ajtmh.15-0234 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Njenga SM, Kanyi HM, Mwatele CM, Mukoko DA, Bockarie MJ, Kelly-Hope LA. Integrated survey of helminthic neglected tropical diseases and comparison of two mosquito sampling methods for lymphatic filariasis molecular xenomonitoring in the River Galana area, Kilifi County, coastal Kenya. PLoS One. 2022. Dec 9;17(12):e0278655. 10.1371/journal.pone.0278655 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Opoku M, Minetti C, Kartey-Attipoe WD, Otoo S, Otchere J, Gomes B, et al. An assessment of mosquito collection techniques for xenomonitoring of anopheline-transmitted Lymphatic Filariasis in Ghana. Parasitology. 2018. Nov;145(13):1783–91. 10.1017/S0031182018000938 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Owusu IO, de Souza DK, Anto F, Wilson MD, Boakye DA, Bockarie MJ, et al. Evaluation of human and mosquito-based diagnostic tools for defining endpoints for elimination of Anopheles-transmitted lymphatic filariasis in Ghana. Trans R Soc Trop Med Hyg. 2015. Oct;109(10):628–35. 10.1093/trstmh/trv070 [DOI] [PubMed] [Google Scholar]
  • 31.Pam DD, de Souza DK, D’Souza S, Opoku M, Sanda S, Nazaradden I, et al. Is mass drug administration against lymphatic filariasis required in urban settings? The experience in Kano, Nigeria. PLoS Negl Trop Dis. 2017. Oct 11;11(10):e0006004. 10.1371/journal.pntd.0006004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Pryce J, Pilotte N, Menze B, Sirois AR, Zulch M, Agbor JP, et al. Integrated xenosurveillance of Loa, Wuchereria bancrofti, Mansonella perstans and Plasmodium falciparum using mosquito carcasses and faeces: a pilot study in Cameroon. PLoS Negl Trop Dis. 2022. Nov 2;16(11):e0010868. 10.1371/journal.pntd.0010868 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Rao RU, Samarasekera SD, Nagodavithana KC, Punchihewa MW, Dassanayaka TDM, P K D G, et al. Programmatic use of molecular xenomonitoring at the level of evaluation units to assess persistence of lymphatic filariasis in Sri Lanka. PLoS Negl Trop Dis. 2016. May 19;10(5):e0004722. 10.1371/journal.pntd.0004722 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.de Souza DK, Sesay S, Moore MG, Ansumana R, Narh CA, Kollie K, et al. No evidence for lymphatic filariasis transmission in big cities affected by conflict-related rural-urban migration in Sierra Leone and Liberia. PLoS Negl Trop Dis. 2014. Feb 6;8(2):e2700. 10.1371/journal.pntd.0002700 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Dyab AK, Galal LA, Mahmoud AE, Mokhtar Y. Finding Wolbachia in filarial larvae and Culicidae mosquitoes in Upper Egypt governorate. Korean J Parasitol. 2016. Jun;54(3):265–72. 10.3347/kjp.2016.54.3.265 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Entonu ME, Muhammad A, Ndams IS. Evaluation of Actin-1 expression in wild caught Wuchereria bancrofti-infected mosquito vectors. J Pathog. 2020. May 4;2020:7912042. 10.1179/000349802125002239 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Fischer P, Wibowo H, Pischke S, Rückert P, Liebau E, Ismid IS, et al. PCR-based detection and identification of the filarial parasite Brugia timori from Alor Island, Indonesia. Ann Trop Med Parasitol. 2002. Dec;96(8):809–21. 10.1179/000349802125002239 [DOI] [PubMed] [Google Scholar]
  • 38.Kouassi BL, de Souza DK, Goepogui A, Narh CA, King SA, Mamadou BS, et al. Assessing the presence of Wuchereria bancrofti in vector and human populations from urban communities in Conakry, Guinea. Parasit Vectors. 2015. Sep 26;8(1):492. 10.1186/s13071-015-1077-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Lupenza E, Gasarasi DB, Minzi OM. Lymphatic filariasis, infection status in Culex quinquefasciatus and Anopheles species after six rounds of mass drug administration in Masasi District, Tanzania. Infect Dis Poverty. 2021. Mar 1;10(1):20. 10.1186/s40249-021-00808-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.McPherson B, Mayfield HJ, McLure A, Gass K, Naseri T, Thomsen R, et al. Evaluating molecular xenomonitoring as a tool for lymphatic filariasis surveillance in Samoa, 2018–2019. Trop Med Infect Dis. 2022. Aug 22;7(8):203. 10.3390/tropicalmed7080203 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Mulyaningsih B, Umniyati SR, Hadisusanto S, Edyansyah E. Study on vector mosquito of zoonotic Brugia malayi in Musi Rawas, south Sumatera, Indonesia. Vet World. 2019. Nov;12(11):1729–34. 10.14202/vetworld.2019.1729-1734 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Nirwan M, Hadi UK, Soviana S, Satrija F, Setiyaningsih S. Diversity, domination and behavior of mosquitoes in filariasis endemic area of Bogor District, west Java, Indonesia. Biodiversitas. 2022;23(4):2093–100. 10.13057/biodiv/d230444 [DOI] [Google Scholar]
  • 43.Nurjana MA, Anastasia H, Widjaja J, Srikandi Y, Nur Widayati A, Murni M, et al. Program Pengendalian Filariasis di Kabupaten Donggala, Sulawesi Tengah. J Dis Vector. 2020;14(2):103–12. Indonesian. 10.22435/vektorp.v14i2.2512 [DOI] [Google Scholar]
  • 44.Ridha MR, Rahayu N, Hairani B, Perwitasari D, Kusumaningtyas H. Biodiversity of mosquitoes and Mansonia uniformis as a potential vector of Wuchereria bancrofti in Hulu Sungai Utara District, south Kalimantan, Indonesia. Vet World. 2020. Dec;13(12):2815–21. 10.14202/vetworld.2020.2815-2821 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Schmaedick MA, Koppel AL, Pilotte N, Torres M, Williams SA, Dobson SL, et al. Molecular xenomonitoring using mosquitoes to map lymphatic filariasis after mass drug administration in American Samoa. PLoS Negl Trop Dis. 2014. Aug 14;8(8):e3087. 10.1371/journal.pntd.0003087 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Supriyono S, Tan S. DNA of Brugia malayi detected in several mosquito species collected from Balangan District, South Borneo Province, Indonesia. Vet World. 2020. May;13(5):996–1000. 10.14202/vetworld.2020.996-1000 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Yokoly FN, Zahouli JBZ, Méite A, Opoku M, Kouassi BL, de Souza DK, et al. Low transmission of Wuchereria bancrofti in cross-border districts of Côte d’Ivoire: a great step towards lymphatic filariasis elimination in West Africa. PLoS One. 2020. Apr 13;15(4):e0231541. 10.1371/journal.pone.0231541 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Zulch MF, Pilotte N, Grant JR, Minetti C, Reimer LJ, Williams SA. Selection and exploitation of prevalent, tandemly repeated genomic targets for improved real-time PCR-based detection of Wuchereria bancrofti and Plasmodium falciparum in mosquitoes. PLoS One. 2020. May 1;15(5):e0232325. 10.1371/journal.pone.0232325 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Rodríguez A, Rodríguez M, Córdoba JJ, Andrade MJ. Design of primers and probes for quantitative real-time PCR methods. Methods Mol Biol. 2015;1275:31–56. 10.1007/978-1-4939-2365-6_3 [DOI] [PubMed] [Google Scholar]
  • 50.Riches N, Badia-Rius X, Mzilahowa T, Kelly-Hope LA. A systematic review of alternative surveillance approaches for lymphatic filariasis in low prevalence settings: implications for post-validation settings. PLoS Negl Trop Dis. 2020. May 12;14(5):e0008289. 10.1371/journal.pntd.0008289 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Cameron MM, Ramesh A. The use of molecular xenomonitoring for surveillance of mosquito-borne diseases. Philos Trans R Soc Lond B Biol Sci. 2021. Feb 15;376(1818):20190816. 10.1098/rstb.2019.0816 [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Bulletin of the World Health Organization are provided here courtesy of World Health Organization

RESOURCES