Aims: To develop and validate criteria for identification of abnormal lymph nodes (LN) at commercial slaughter, for the purpose of national surveillance for Mycobacterium avium subspecies paratuberculosis (Map) in New Zealand farmed deer. This included estimation of the predictive value of abnormal LN for Map infection; a standard circumference cut-point for measurement of abnormal LN; and identification of risk factors associated with increasing LN circumference.
Methods: In Study 1, official assessors sampled 129 LN with macroscopically visible abnormalities (abnormal LN) from 76 deer herds between May and November 2007. LN samples were cultured for Map, with culture-negative LN further examined for typical histopathological changes. The predictive value of abnormal LN for Map infection was calculated and significance of herd location (North or South Island) assessed. In Study 2, the circumferences of 412 mesenteric LN (MLN) from 79 herds were measured between October 2007 and January 2009, with samples cultured for Map and examined for eight histopathological features. The minimum circumference of an abnormal MLN was defined, based on an arbitrary >95% specificity of a culture-positive Map diagnosis. Associations between the predictor variables Map culture status, carcase weight, animal age and gender, and histopathological features, and increasing MLN circumference were assessed using a linear mixed-effects model.
Results: Based solely on culture, the predictive value of abnormal LN for Map infection was 92.2 (95% CI: 86.2-96.2)% with no difference between the North and South Islands (p = 0.09). Inclusion of three culture-negative LN with histopathological changes typical of Map infection increased the predictive value estimate to 94.6 (95% CI: 89.2-97.3)%. The minimum circumference of an abnormal MLN was defined as 55 mm, with a sensitivity of Map detection at this cut-point of approximately 12%. Increasing MLN circumference was positively associated with the presence of moderate follicular hyperplasia (p < 0.01), focal granulomas (p < 0.01) and a synergistic interaction between focal granulomas and Map status (p = 0.03).
Conclusions: Deer MLN with macroscopically visible abnormalities and/or a circumference of >55 mm have >95% likelihood of Map infection. However, sensitivity of Map diagnosis in MLN with circumference of >55 mm was 12%, indicating use of abnormal LN as a sole criterion in national surveillance for Map in slaughtered deer will underestimate animal-level prevalence.