Epidemiology, detection and milk production effects of Mycobacterium avium subspecies paratuberculosis on California dairies

Johne's disease is a granulomatous enteritis of cattle caused by Mycobacterium avium subsp. paratuberculosis (MAP). Interest in Johne's is promoted by reports of decreased milk production, however, most publications were cross-sectional. The longitudinal studies available lack assessment of confounding and model the standard lactation curve rather than that expressed by each cow. For the first study, mixed models based on causal diagrams were used to estimate the association between MAP and milk production, using cow-specific splines. Compared to test-negative cows, MAP seropositive and shedding cows produced 2.5 and 2.2 Kg less of milk per day, respectively.;Environmental sampling and pooled fecal culture were used to identify super-shedders, cows that shed MAP in tens of thousands per gram of feces. The objective was to estimate the cost-effectiveness of diagnostic strategies to identify super-shedders such as ranking pens on a dairy by MAP bioburden using environmental samples. Subsequently, cows in high MAP bioburden pens would be tested. Results of the second study found that the reliability of environmental sampling across collectors and over 5 days was excellent (81%) and good (67%), respectively.;Due to routine moving of cows between pens for management purposes, pen rankings must be available within 2-3 days. The availability of quantitative real-time PCR (qrt-PCR) made it possible to quantify MAP in samples within days compared to weeks using culture. Results of the third study found good correlation between assays for pooled and environmental samples (-0.66 and -0.76, respectively) and a non-linear xiiassociation between results of both assays that differed in magnitude between fresh and thawed samples.;A whole herd test was conducted including combinations of serum or milk ELISA, qrt-PCR or culture of environmental, individual or pooled fecal samples. Results of all the tests were used to simulate 15 diagnostic strategies to detect super-shedders on a large dairy and estimate their cost-effectiveness. A whole herd qrt-PCR survey sample was the most sensitive (80%) of all 15 strategies. The most cost-effective strategy was to rank milking pens by MAP bioburden, ELISA test milk of cows in high MAP bioburden pens, followed by qrt-PCR testing of ELISA positive cows.