Exploring sources of inaccuracy and irreproducibility in the CDC bottle bioassay through direct insecticide quantification

Abstract Background The Centers for Disease Control and Prevention (CDC) bottle bioassay is a commonly used susceptibility test for measuring insect response to insecticide exposure. However, inconsistencies and high variability in insect response when conducting CDC bottle bioassays have been reported in previous publications. We hypothesized that the CDC bottle bioassay results may be compromised when expected and actual insecticide concentrations in the bottles are not equivalent and that inadequate bottle cleaning and/or loss during insecticide introduction and bottle storage steps could be responsible. We explored this hypothesis by quantifying insecticides using gas chromatography tandem mass spectrometry (GC-MS/MS) in bottles that had been cleaned, prepared, and stored according to the CDC guidelines. Methods We investigated the bottle cleaning, preparation, and storage methods outlined in the CDC bottle bioassay procedure to identify sources of irreproducibility. We also investigated the effectiveness of cleaning bottles by autoclaving because this method is commonly used in insecticide assessment laboratories. The two insecticides used in this study were chlorpyrifos and lambda-cyhalothrin (λ-cyhalothrin). Insecticides were removed from glass bioassay bottles by rinsing with ethyl-acetate and n -hexane and then quantified using GC-MS/MS. Results The CDC bottle bioassay cleaning methods did not sufficiently remove both insecticides from the glass bottles. The cleaning methods removed chlorpyrifos, which has higher water solubility, more effectively than λ-cyhalothrin. Chlorpyrifos experienced significant loss during the bottle-coating process whereas λ-cyhalothrin did not. As for bottle storage, no significant decreases in insecticide concentrations were observed for 6 h following the initial drying period for either insecticide. Conclusions The CDC bottle bioassay protocol is susceptible to producing inaccurate results since its recommended bottle cleaning method is not sufficient and semi-volatile insecticides can volatilize from the bottle during the coating process. This can lead to the CDC bottle bioassay producing erroneous LC 50 values. High levels of random variation were also observed in our experiments, as others have previously reported. We have outlined several steps that CDC bottle bioassay users could consider that would lead to improved accuracy and reproducibility when acquiring toxicity data. Graphical Abstract