The epidemiology of corresponding tick-borne diseases has changed as a result of the recent introduction of Rhipicephalus (Boophilus) microplus to West Africa. The current study aimed to assess the diagnostic performance of an indirect ELISA for the detection of Babesia bovis infection in cattle. In a cross-section study, using a Bayesian Latent Class Model and iELISA diagnostic test for cattle babesiosis due to Babesia bovis, accuracy has been assessed with RT-PCR as an imperfect reference test. A total of 766 cattle were tested. The optimal diagnostic performances were obtained with 5% percentage of positivity. Sensitivity and specificity were, respectively, 0.94 [Cr. I.: 0.85−0.99] and 0.89 [Cr. I.: 0.87−0.92]. Additional diagnostic characteristics revealed that the Positive Predictive Value (PPV) and Negative Predictive Value (NPV) were 96.6% [Cr. I.: 92.7−100%] and 82.2% [Cr. I.: 72−93%]. Overall, this test well discriminates an infected status from an uninfected status considering the area under the ROC curve (AUC) which was 0.78 [Cr. I: 0.72−0.85] and a Diagnostic Odds Ratio (DOR) of 127.8 [Cr. I.: 10.43−1562.27]. The AUC was significantly higher than 0.5 (p < 10−5). In consequence, this serologic assay could be suitable in moderate to high prevalence assessments.
Keywords: Babesia bovis; Bayesian; RT-PCR; cattle; iELISA; index test.