Diagnostic accuracy of a Mycobacterium phlei-absorbed serum enzyme-linked immunosorbent assay for diagnosis of bovine paratuberculosis in dairy cows.
The purpose of this study was to describe the responses of sera from five groups of cattle to an enzyme-linked immunosorbent assay (ELISA) for paratuberculosis by using serum absorbed with Mycobacterium phlei at a single working dilution. The infection status of the cattle was determined by fecal culture. Cattle with different levels of exposure (high versus low prevalence and test negative) and disease manifestation (clinically suspect infection versus subclinical infection) were examined, as follows: (i) two paratuberculosis-negative herds; (ii) a fecal culture-confirmed, clinically suspect cases of paratuberculosis; (iii) cows from a paratuberculosis-infected herd with a high infection rate, as determined by fecal culture, but with no clinical cases at the time of sampling; (iv) cows from three paratuberculosis-infected herds known to have paratuberculosis diagnosed on the farm (low infection rate determined by fecal culture); and (v) one fecal culture-negative herd with known serologically positive cattle. Results generally showed a decreased ELISA response when absorbed rather than nonabsorbed serum from each animal was used. The results of the fecal culture confirmed clinically suspect cases, which were analyzed in relation to the amount of colonies isolated from the animals on fecal culture (0, +, ++,+++ , ++++, and above). There was a significant increase in the ELISA response for animals with heavy Mycobacterium paratuberculosis shedding ( ++++ or above), when both unabsorbed and absorbed sera were used, compared with the response in animals that were fecal culture negative or that shed M. paratuberculosis at lower levels (less than +) (P less than 0.05). The effects on sensitivity and specificity by using different cutoff points for the five groups of cattle with different levels of exposure is described, since sera were not discretely segregated into distinct groups of positive and negative samples. The specificity of the ELISA in the two fecal culture-negative herds was 100% at an ELISA cutoff of an optical density (OD) of 0.1 and above for absorbed serum. For unabsorbed serum the specificity was 62.9% at a similar cutoff value. Similarly, the specificity of the fecal culture-negative, serologically positive herd increased from 37.5 to 72.2 at an ELISA cutoff value of 0.1 to 0.2 (OD) by using absorbed versus unabsorbed serum from 75.0 to 94.4 at an ELISA cutoff value of 0.2 to 0.3 (OD).