A screening sampling plan to detect Mycobacterium avium subspecies paratuberculosis-positive dairy herds.

Mycobacterium avium ssp. paratuberculosis (MAP) is the etiological agent of paratuberculosis, a chronic contagious bacterial disease primarily affecting dairy cattle. Paratuberculosis represents a dual problem for the milk production chain: in addition to economic losses to affected herds, MAP may have zoonotic potential. Infected herds must be identified in order to implement programs designed to reduce the incidence of disease within and between herds and to prevent MAP from entering the food chain. The objective of this study was to evaluate the sensitivity and specificity of a screening sampling plan (SSP) to detect MAP-positive dairy herds by repetitive analysis of bulk tank milk (BTM) samples by ELISA and in-line milk filter (ILMF) samples by PCR. Samples from BTM and ILMF were collected twice from 569 dairy herds in southern Italy. Additionally, 12,016 individual milk samples were collected: 9,509 from 102 SSP-positive herds (SSP MAP-positive) and 2,507 from 21 randomly selected SSP-negative herds (SSP MAP-negative). There was a total of 126 SSP MAP-positive herds (i.e., 21.3% SSP MAP-positive herds; 95% confidence interval=18.0-24.9); the within-herd apparent prevalence (AP) ranged between 0.00 and 22.73% (mean 6.07%). A significant difference in within-herd AP was shown between SSP MAP-positive herds and SSP MAP-negative herds. A highly significant association was shown between the median AP herd status (>5%) and positivity to at least one ILMF or BTM sample. The SSP detected a minimum of 56.25% of low AP herds (AP ≤ 2.0%) up to a maximum of 100% of herds with a within-herd AP ≥ 8.0%. Overall, the SSP detected 85.57% of herds in which at least one individual milk sample was positive by ELISA. The proposed SSP was an inexpensive and useful tool to detect MAP-positive herds with a higher risk of infection diffusion and milk contamination. Although the SSP cannot be used for MAP-free certification of herds, it could be useful to prioritize appropriate control measures aimed at reducing the prevalence of infection in dairy herds and milk contamination.