Comparability between insecticide resistance bioassays for mosquito vectors: time to review current methodology?

BACKGROUND: Insecticides play an integral role in the control of mosquito-borne diseases. With resistance to insecticides on the rise, surveillance of the target population for optimal choice of insecticides is a necessity. The Centers for Disease Control and Prevention (CDC) bottle assay and the World Health Organization (WHO) susceptibility test are the most frequently used methods in insecticide resistance monitoring. However, the two bioassays differ in terms of insecticide delivery and how insecticide susceptibility is measured. To evaluate how equivalent data from the two assays are, we compared the two methods side-by-side. METHODS: We did a literature search from 1998 to December 2014 to identify publications that performed both assays on the same mosquito population and compared the results. We then tested the WHO and CDC bioassays on laboratory strains of Aedes aegypti, Anopheles stephensi, An. gambiae and An. arabiensis with different insecticide resistance levels against permethrin, λ-cyhalothrin, DDT, bendiocarb and malathion. In addition, we also measured the relationship between time-to-knockdown and 24 h mortality. RESULTS: Both published data and results from the present laboratory experiments showed heterogeneity in the comparability of the two bioassays. Following their standard procedures, the two assays showed poor agreement in detecting resistance at the WHO cut-off mark of 90 % (Cohen’s κ = 0.06). There was better agreement when 24 h mortality was recorded in the CDC bottle assay and compared with that of the WHO susceptibility test (Cohen’s κ = 0.5148). Time-to-knockdown was shown to be an unreliable predictor of 24 h mortality. CONCLUSION: Even though the two assays can detect insecticide resistance, they may not be used interchangeably. While the diagnostic dose in the WHO susceptibility test does not allow for detecting shifts at low or extreme resistance levels, time-to-knockdown measured in the CDC bottle assay is a poor predictor of 24 h mortality. Therefore, dose–response assays could provide the most flexibility. New standardized bioassays are needed that produce consistent dose–response measurements with a minimal number of mosquitoes.