Abstract
OBJECTIVE: To evaluate PCR assay sensitivity and specificity compared to that of microscopic agglutination test (MAT) for diagnosis of canine leptospirosis. Electronic records search was performed to identify dogs with results for both PCR and MAT testing for leptospirosis. METHODS: All dogs were clinically ill. Diagnosis of leptospirosis was defined as an unvaccinated dog with a positive MAT titer of ≥1:800 or a vaccinated dog or dog with an unknown vaccination status with a positive MAT titer of ≥1:1,600. Diagnosis of leptospirosis was excluded based on MAT titer <1:800 on both the initial and convalescent samples or an initial MAT titer <1:800 and an alternative definitive diagnosis. RESULTS: Forty-nine samples (urine, n = 39; blood, n = 10) were evaluated. Leptospirosis was diagnosed in 17 dogs and excluded in 26 dogs. Urine PCR assay demonstrated sensitivity of 69.2%, specificity of 100%, positive predictive value of 100%, and negative predictive value of 86.6%. Blood PCR assay demonstrated sensitivity of 25%, specificity of 100%, positive predictive value of 100%, and negative predictive value of 25%. Overall PCR sensitivity was 52.4%, specificity was 100%, positive predictive value was 100%, and negative predictive value was 73.7%. CONCLUSIONS: PCR assay performed on urine or blood has high specificity and positive predictive value when compared to MAT for diagnosis of clinical canine leptospirosis. Sensitivity and negative predictive value are moderate to low, so PCR testing should be performed in conjunction with paired MAT testing for canine leptospirosis. Prior antibiotic therapy does not preclude the use of the PCR test.