Evaluation of milk cathelicidin for detection of bovine mastitis

Mastitis due to intramammary infection is one of the most economically relevant diseases in dairy cows, causing reductions in milk quality and quantity. Currently, mastitis monitoring is based on somatic cell count (SCC) and bacteriologic culture (BC) of milk. Nevertheless, inflammation-specific protein markers might provide more sensitive and reliable assays, enabling immunoassay-based screening strategies. Cathelicidin is an inflammatory protein released in milk that has recently demonstrated fair reliability and diagnostic potential for ewe mastitis. To assess its performance in cows, 531 quarter milk samples from 2 herds were tested using cathelicidin ELISA, SCC, and BC. We found that 29.0% of samples were positive for cathelicidin, 18.8% had SCC >200,000 cells/mL, and 13.7% were BC-positive. Cathelicidin showed a strong positive correlation with SCC as demonstrated by receiver operating characteristics curve analysis and by the clustering of cathelicidin-negative and cathelicidin-positive samples in association with low and high SCC values, respectively. For evaluating the diagnostic performance of a novel test, BC cannot be considered a reliable gold standard for true disease status because of its known limitations. Therefore, we assessed the sensitivity (Se) and specificity (Sp) of the milk cathelicidin ELISA using a latent class analysis approach together with BC and SCC by considering different diagnostic thresholds to identify the preferred Se/Sp combination. We modeled conditional dependence of cathelicidin and SCC to account for their close association. The cathelicidin ELISA showed higher Se than SCC and BC for almost all threshold combinations. In fact, at the best-performing threshold combination, the Se of cathelicidin was 80.6%, 6.2 percentage points higher than that of SCC >200,000 cells/mL (74.4%) and similar to that of SCC >100,000 cells/mL (80.2%). Most importantly, this Se was obtained with a loss in Sp of only 1.4 percentage points compared with SCC >200,000 cells/mL (94.9% Sp for cathelicidin vs. 96.3% for SCC >200,000). The limited Se of BC (38.8%) was also confirmed in this study, and BC showed a slightly lower Sp than both cathelicidin and SCC for most of threshold combinations. This study confirmed that cathelicidin is released in the milk of cows with mastitis and that its presence is highly correlated with SCC. The measurement of cathelicidin by ELISA may hold significant potential for improving the sensitivity of mastitis detection in dairy cows while maintaining high specificity.