Title : Evaluating the contribution of Aedes albopictus on dengue prevalence: A systematic literature review and meta-analysis
Abstract:
Introduction: Although Aedes aegypti is routinely cited as the principal mosquito vector of dengue virus, there is limited evidence base around its contribution relative to A. albopictus.
Methods: We performed a systematic review and meta-analysis to compare the dengue prevalence in A. albopictus and A. aegypti among studies simultaneously evaluating both. We searched EMBASE, PubMed, Scielo and Global Index Medicus databases. We performed meta-analysis using the inverse variance-weighted random effect method and separated studies reporting individual and pooled mosquito data. We reported risk ratios ‘RR’ to display the relation between A. albopictus and aegypti positivity for dengue (RR>1, greater contribution among A. albopictus). Sub-group meta-analysis was employed using country’s income level, WHO region, urbanization level, and indoors/outdoors activity. We assessed risk of bias and graded robustness of evidence using the MASTER quality assessment and LFK index, respectively. We employed sensitivity analyses utilising the leave-one-out technique.
Results: Of 5,412 records, we included 48 studies in our final sample (n=12, n=32, and n=4 among individual, pooled, and artificially-infected mosquito data). Most studies were performed in India and Brazil (n=7, each). Meta-analysis showed a higher prevalence among A. aegypti in studies using individual mosquito data (RR=0.95, 95%CI=0.40, 2.27), whereas A. albopictus showed higher prevalence in articles using pooled mosquito data (RR=1.03, 95%CI=0.61, 1.73). Importantly, in neither case was there evidence for a significant difference in prevalence between species. Subgroup analyses showed higher prevalence in A. albopictus specifically among upper-middle income countries analysing individual mosquito data (RR=1.92, 95% CI=1.03-3.58). Overall, we found substantial heterogeneity (I2>65%) and a relative change in estimates between -14% and 16% compared to main meta-analysis results when applying the leave-one-out technique. The risk assessment confirmed the absence of accommodating confounding variables in the evaluated studies, limiting their generalisability.
Discussion: Our findings showed that, where both vectors were caught and tested, A. aegypti did not show clear dominance as a dengue vector. Prevalence of infection among vectors is only one aspect in determining the relative role that these species have on dengue transmission; and, our findings urge more discriminatory future studies. Disentangling their contributions will improve current and future risk assessments, and have important implications for control strategies.
