Aim: To determine the association between white-spot lesions in kidneys and serological and cultural prevalence of leptospirosis in sheep, and to evaluate the diagnostic value of these lesions in individual sheep and lines of sheep at slaughter as indicators of past or current episodes of leptospirosis.
Methods: Lines of lambs were randomly selected, and within lines individual lambs were randomly selected at slaughter. Blood samples and entire kidneys were collected. Serum was tested using the microscopic agglutination test (MAT) for antibody against Leptospira borgpetersenii serovar Hardjobovis or Leptospira interrogans serovar Pomona. Kidneys were cultured for the presence of Leptospira spp. The association between grossly visible white-spotted kidneys (WSK) and the serological status, and between WSK and culture status was evaluated at both line and individual levels. A fixed-effect multivariable logistic regression model was fitted to the line-level data, and included within-line prevalence of carcasses with WSK and line size. A random-effect multivariable logistic regression model was fitted to the individual-level data. This model included WSK lesion score and a random line effect.
Results: White-spot lesions in kidneys were significantly associated with the serological status for Leptospira spp. in individual sheep. A strong positive dose-response relationship between sero-status and the number of white spots on kidneys was observed. However, the sensitivity of WSK to detect seropositive carcasses was low (51 (95% CI=43-59)%), and specificity was moderately low (86 (95% CI=84-87)%). Due to a low observed seroprevalence of 5.2 (95% CI=3.9-7.1)% to serovar Hardjo or Pomona, the positive predictive value (PPV) of WSK for serology was only 18 (95% CI=14-22)%, and the negative predictive value (NPV) was 96 (95% CI=96-97)%. Carcasses with high WSK lesion scores (more than five white spots or white mottling on one or both kidneys) were 6.1 (95% CI=4.3-8.3) times more likely to be seropositive to either serovar than were carcasses with low scores (one to five white spots on one or both kidneys). However, the test sensitivity and PPV for these criteria were regarded unacceptably low (27 (95% CI=20-34)% and 27 (95% CI=21-35)%, respectively). Consideration of lesion status of lines rather than individual animals resulted in a high sensitivity of 98 (95% CI=87-100)%, but very low specificity of 15 (95% CI=8-27)% and a PPV of 48 (95% CI=37-59)%. Due to the low sensitivity of WSK and low prevalence of culture- positive carcasses, the PPV for WSK was as low as 4 (95% CI=2-12)%.
Conclusions: Whereas highly significant associations, including a strong dose-response effect, were observed between WSK and MAT serology, WSK was a poor predictor for the antibody and pathogen status of sheep carcasses with respect to leptospirosis.