Insecticide-treated screening of windows for household protection against insecticide-resistant Anopheles gambiae sensu lato in Côte d'Ivoire: a semi-field trial.

backgroundDespite the significant impact of long-lasting insecticidal nets (LLINs) and indoor residual spraying (IRS) on malaria, 597,000 deaths from malaria were still recorded in 2023. Additional measures are clearly needed to complement current tools. We assessed the efficacy of insecticide-treated window screens (a piece of PermaNet 3.0 containing deltamethrin + piperonyl butoxide) versus untreated window screening and eave tube inserts treated with β-cyfluthrin as household interventions against Anopheles gambiae sensu lato.

methodBetween August and October 2022, five experimental houses in Kolongonouan village, central Côte d'Ivoire, were used under semi-field conditions to evaluate the following interventions: (i) standard control house with small openings in the eaves and no window screens (SCH); (ii) insecticide-treated screening (ITS) of windows with small openings in the eaves (ITS op); (iii) ITS of windows with eaves blocked to prevent mosquito entry via eaves (ITS blq); (iv) insecticide treatment applied to the eaves using In2Care eave tubes, without window screening (ET); and (v) untreated window screening together with insecticide-treated eave tubes (SET). The efficacy of these treatments on the indoor and outdoor density of naturally recruiting Anopheles gambiae mosquitoes was assessed by human landing catches from 6:00 pm to 8:00 am. The impact of these treatments on mosquito mortality and blood-feeding rates was further assessed using release-recapture experiments, in which large enclosures were built around individual houses and approximately 100 non-blood-fed female An. gambiae were released each night over 15 nights. In addition, for the ITS blq treatment, the importance of physical integrity was assessed by evaluating the effect of damaging the screening with two or four 4 cm × 4 cm holes in each screened window.

resultsInsecticide-treated window screening plus eave blocking (ITS blq) significantly reduced by 31.8% the number of An. gambiae entering houses overnight relative to standard control houses (SCH) (P < 0.001). The use of untreated screening + eave tubes (SET) induced a similar reduction (27.0%, P < 0.001). The house entry reduction magnitude for eave tubes alone (ET) was lower (23.5%), though it remained significant relative to the control (P < 0.001). The impact of the ITS op (insecticide-treated window screening with open eaves) was lower and nonsignificant (18.1%; P = 0.4019). There were no significant differences in house entry rates between ET, ITS blq, and ITS op (P > 0.05). The peak in mosquito abundance within houses occurred between 11:00 pm and 3:00 am. There were no significant differences between treatments in the numbers of mosquitoes collected outdoors. In the release-recapture experiments, all treatments significantly reduced the blood-feeding rate and increased the mortality of An. gambiae, with again a similar impact for ITS blq and SET. Damaging the ITS with two holes still led to a significant reduction in blood-feeding rate, but protection was lost with four holes. Mortality rate also declined with increasing levels of damage, although it remained significantly higher than that with the untreated window screening.

conclusionsThis study demonstrated that the combination (insecticide-treated window screening and blocking of eave access) reduced mosquito entry, increased mortality, and lowered blood-feeding rates of An. gambiae to levels comparable to insecticide-treated eave tubes combined with untreated window screening (SET). A previous cluster randomized controlled trial of SET in this area demonstrated a significant reduction in malaria incidence and prevalence. Insecticide-treated window screens could therefore yield a similar epidemiological impact while potentially being simpler and less expensive to implement. Larger-scale epidemiological trials involving communities are needed to test this assumption and further optimize the approach.