Comparison of serological tests for equine trypanosomosis in naturally infected horses from Kazakhstan.

In this study, we compared the complement fixation test (CFT), the horse complement fixation test (HCFT) and a card agglutination test for trypanosomosis (CATT/T. evansi) for the diagnosis of equine trypanosomosis in the Republic of Kazakhstan. Cohen's kappa test was used to evaluate the concordance between the three tests. Kappa scores for CFT versus HCFT and CATT are both 0.6165 (95% Confidence Interval CI 0.414--0.819) indicating a "substantial" agreement between CFT and HCFT or CATT, respectively. Kappa for HCFT versus CATT is 0.395 (CI 0.142--0.648) indicating a "fair" agreement between the two tests. In the absence of a golden standard, seroprevalence and sensitivity and specificity of the three tests were estimated using maximum likelihood estimation. CFT has a sensitivity of 57.2% (CI 31.5--79.5%) and a specificity of 95.8% (CI 89.2--98.5%), HCFT has a sensitivity of 80.6% (CI 44.1--95.6%) and a specificity of 99.5% (CI 90.7--100%), CATT has a sensitivity of 80.2% (CI 44.5--95.2%) and a specificity of 98.5% (CI 79.5--99.9%). The seroprevalence of equine trypanosomosis in Kazakhstan was estimated at 16.4% (CI 9.4--27.0%). The data suggest that for epidemiological studies and the control of equine trypanosomosis serological tests prove useful since they have a high specificity and a satisfactory sensitivity. Field applicable tests, such as CATT/T. evansi may be used to replace laboratory-based tests, such as CFT and HCFT.