The entomological impact of passive metofluthrin emanators against indoor Aedes aegypti: A randomized field trial

Abstract

Background: In the absence of vaccines or drugs, insecticides are the mainstay of Aedes-borne disease control. Their utility is challenged by the slow deployment of resources, poor community compliance and inadequate household coverage. Novel application methods are required.

Methodology and principal findings: A 10% w/w metofluthrin "emanator" that passively disseminates insecticide from an impregnated net was evaluated in a randomized trial of 200 houses in Mexico. The devices were introduced at a rate of 1 per room and replaced at 3-week intervals. During each of 7 consecutive deployment cycles, indoor resting mosquitoes were sampled using aspirator collections. Assessments of mosquito landing behaviours were made in a subset of houses. Pre-treatment, there were no differences in Aedes aegypti indices between houses recruited to the control and treatment arms. Immediately after metofluthrin deployment, the entomological indices between the trial arms diverged. Averaged across the trial, there were significant reductions in Abundance Rate Ratios for total Ae. aegypti, female abundance and females that contained blood meals (2.5, 2.4 and 2.3-times fewer mosquitoes respectively; P<0.001). Average efficacy was 60.2% for total adults, 58.3% for females, and 57.2% for blood-fed females. The emanators also reduced mosquito landings by 90% from 12.5 to 1.2 per 10-minute sampling period (P<0.05). Homozygous forms of the pyrethroid resistant kdr alleles V410L, V1016L and F1534C were common in the target mosquito population; found in 39%, 24% and 95% of mosquitoes collected during the trial.

Conclusions/significance: This is the first randomized control trial to evaluate the entomological impact of any volatile pyrethroid on urban Ae. aegypti. It demonstrates that volatile pyrethroids can have a sustained impact on Ae. aegypti population densities and human-vector contact indoors. These effects occur despite the presence of pyrethroid-resistant alleles in the target population. Formulations like these may have considerable utility for public health vector control responses.

Fig 1

Map of the city of Ticul, Yucatán state (Mexico) showing the random distribution of metofluthrin-treated houses (red dots), untreated controls (blue dots) and the subset of houses selected to assess mosquito landing behaviour (circled dots).

Fig 2. Placement of the metofluthrin passive emanators in a room of treatment house.

The emanators consist of metofluthrin impregnated mesh contained in a plastic housing (A). Metofluthrin impregnated emanators were hung from ceilings, above head height, to keep them clear of routine household movement and activity. The emanator in this room is circled in red (B).

Fig 3

Influence of the metofluthrin emanators on indoor adult Ae. aegypti collections. Black triangles represent treated houses and grey circles represent controls for A) total Ae. aegypti, B) female Ae. aegypti and C) blood-fed Ae. aegypti. The dotted lines represent emanator deployment. The x-axis labels denote the pre-treatment baseline, and every subsequent deployment cycle (P1-P7). Data presented as means + 95% CI.

Fig 4

Efficacy of metofluthrin emanators against Ae. aegypti in comparison to baseline. Efficacy in relation to the untreated baseline for A) total Ae. aegypti, B) female Ae. aegypti and C) blood-fed Ae. aegypti. The dotted lines represent average efficacy for each measure. The x-axis labels denote successive deployment cycles (P1-P7). Data presented as means ± 95% CI.

Fig 5

Impact of emanators on mosquito landing behaviour. Ae aegypti landings in the control (black bars) and treatment houses (grey bars). Pooled data from across both assessments is presented as means per house and 95% CI. These measures were made immediately following the first deployment of the emanators in a subset of recruited households (n = 16). Data for all rooms and houses were pooled and averaged for the control and treatment houses.

Fig 6. Frequencies of tri-locus genotypes of mosquitoes collected during the trial.

The order of the genotypes on each row of the y axis is 410 / 1016 / 1534. The resistant mutations screened for were V410L, V1016I and F1534C. Resistant homozygous forms (RR) are denoted by LL, II and CC respectively. The triple susceptible genotype is at the bottom of the graph and the triple resistant genotype is at the top of the graph.