Evaluation of laboratory and field methods for measuring mosquito repellent efficacy

Mosquito repellents are on the market since more than 60 years and represent a good solution for travellers to endemic areas where mosquitoes may transmit several diseases. But how long does a repellent protect against mosquito bites? Mosquito repellents are usually tested in the laboratory with the arm-in-cage test method and the results are the basis for the label claim. The general assumption is that protection times measured under laboratory conditions are a good proxy for the repellent’s efficacy under end user conditions. However, it remains unclear how informative the arm-in-cage test is.
As a first approach to answer this question, a systematic literature search was conducted to review the efficacy of the four most used active ingredients DEET, PMD, Icaridin and EBAAP both under laboratory and field conditions in order to assess the predictive value of the arm-in-cage test. The available data from the literature were, however, insufficient to draw a clear conclusion.
Therefore a comparative study with field and laboratory experiments was conducted, measuring the efficacy of repellents. In two nature reserves, Langholz (Canton of Aargau) and Thurauen (Canton of Zurich) as well as under laboratory conditions the protection of 15% DEET and 15% PMD was assessed and compared between field and laboratory using the same 18 study participants. In the field, both DEET and PMD provided full protection up to at least 6 hours, while in the laboratory DEET 15% protected for a maximum of 30 minutes against Ae. aegypti, 2 hours against An. stephensi (95% CI: 1 – 3 hours) and 2 hours (95% CI: 1.5 – 3.5 hours) against Cx. quinquefasciatus. For PMD 15% median CPTs in the arm-in-cage test were slightly lower for An. stephensi and Cx. quinquefasciatus with times of 0.5 hour (95% CI: 0.5 – 1 hour) and 1 hour (95% CI: 0.5 – 1.0 hour), while average CPT for Ae. aegypti was again a maximum of 0.5 hour.
During these and previous experiments it was observed that mosquitoes being repeatedly exposed to repellents in the arm-in-cage test change their behaviour during the course of an experiment. The hypothesis was then that mosquitoes become adapted to odours trapped inside the cage; and hence are less responsive to neither the repellent nor the human odour cues. To test whether air trapped inside the cage may influence mosquito behaviour a subsequent series of experiments were carried out to investigate if air ventilation has an influence on the biting behaviour of the mosquitoes. Tests were performed with 10 study participants and the repellents DEET 30% and PMD 30%.
The complete protection times measured against Ae. aegypti with air ventilation were between 0.5 – 1.5 hours for DEET 30% and between 0.5 – 1 hour for PMD 30%. Without air ventilation the protection times for DEET 30% were between 0.5 and 2 hours and for PMD 30% 0.5 and 1 hour. DEET 30% repelled An. stephensi between 0.5 – 5 hours and PMD
30% protected the study participants between 0.5 – 1.5 hours with the air ventilation. Without air ventilation DEET 30% protected for 1 – 5 hours and PMD 30% for 0.5 – 2 hours. However, ventilation had no significant effect on the complete protection times, yet mosquitoes in air-ventilated cages seemed to remain more active over time in the negative controls.
Methods for repellent testing in the laboratory (arm-in-cage test) and in the field (HLC) are well explained in the guidelines of the WHO and the US EPA and I would highly recommend following these guidelines to get comparable results. This was one of the biggest problems in the review study where I was looking for protection times of repellents in field and laboratory experiments. Additionally, I would recommend measuring complete and relative protection as well as landing rates of mosquitoes in field and laboratory experiments to get a better overview of the biting pressure.
As long as the experimental designs vary in such an enormous way despite existing guidelines an interpretation of the studies and the measured protection times is extremely difficult. Our experiments were based on the guidelines and it was difficult anyway to compare the results of both methods. But it is possible figure out a tendency. In general it can be observed that repellents with good protection times in the laboratory were also effective in the field as shown in the review and my own field and laboratory studies.