Plasticity of blood feeding behavior of Anopheles mosquitoes in Ethiopia: a systematic review

Background

The efficacy of vector control tools depends on the behavior of the vector species. Many studies have sought to determine the feeding behavior of Anopheles mosquitoes in different settings of Ethiopia. We have performed a systematic review aimed to generate pooled evidence on the overall and species-specific blood meal sources of Anopheles mosquitoes in Ethiopia.

Methods

A search for relevant articles was performed in two electronic databases (PubMed and Science Direct) and three search engines (Google Scholar, Research Gate and Google) between 11 March and 2 April 2024. Following the initial identification of articles, we used EndNote X8 software and removed duplicate articles and screened the remaining articles by careful reading of their titles and abstracts. The full text of articles that passed this screening phase was retrieved, read and evaluated against predetermined selection criteria. The final decision for inclusion in the systematic review was made after a methodological quality check using the JBI critical appraisal checklist. All relevant data were extracted from tables, figures and texts of the included articles using a premade template in Excel, and the data were analyzed using Stata version 14 software.

Results

Of the 2431 studies identified, 27 met the inclusion criteria; all were published between 1997 and 2024. At 215 data points (frequency of tests of each Anopheles species by location and method of mosquito collections), 18,771 Anopheles mosquitoes belonging to 23 species or species complexes were tested for blood meal sources. The commonest sources of blood meals for Anopheles mosquitoes were bovine (36.0%, n = 6758) and human (29.4%, n = 5520). Among the tested anophelines, Anopheles (An.) arabiensis accounted for 67.9% (n = 12,741), followed by An. pharoensis, An. demeilloni and An. stephensi at 10.0%, 5.6% and 4.4%, respectively. Overall, there was no difference in the mean proportion of An. arabiensis detected with domestic animal blood (33.4%, 95% confidence interval [CI] 32.4–34.4%) and those detected with human blood (31.8%, 95% CI 30.9–32.8%). However, a greater proportion of the outdoor collected An. arabiensis were found to feed on bovines (47.9%, 95% CI 35.3–60.6) compared to humans (12.9%, 95% CI 0.8–24.9, P < 0.01). The foraging ratio (FR), which accounts for host availability, was greater for bovines (FR = 0.7) than for humans (FR = 0.2) for An. arabiensis, indicating preferential feeding on bovine hosts. This host preference was supported by the host preference index (human:bovine = 0.4). Anopheles pharoensis was detected with a slightly higher human blood index (53.5%, n = 1005) compared to bovine blood index (45.2%, n = 849). In contrast, An. demeilloni, An. coustani and An. marshalli were detected with a higher bovine blood index. Recently invaded urban malaria vector, An. stephensi was found with a higher ovine blood index.

Conclusions

Bovine and human hosts are common sources of a blood meal for Anopheles mosquitoes. In terms of host availability, An. arabiensis showed preferential feeding on bovines/cattle. Targeting domestic animals, bovines and ovines with endectocides could supplement current vector control interventions.