A reliable method using the cytobrush for diagnosis of subclinical endometritis in dairy cattle during the late lactational period

This experiment was performed to assess reliability of the cytobrush-cytology method (CCM) in diagnosis of subclinical endometritis (SCE) using the biopsy-histopathology method (BHM) as a reference in late lactating dairy cows. Reproductive organs were collected from 115 slaughtered multiparous crossbred cows culled due to infertility 398 +/- 135 days subsequent to parturition. Samples were collected from the dorsal part of the corpus uteri for analyses. Inflammation status was graded histopathologically based on the cell percentages [(neutrophils, eosinophils, lymphocytes (LYM), macrophages (MAC), and plasma cells)]. Data were subjected to Friedman's test for group comparisons (method and diagnosis), concordance correlation and chi-square tests for consistency of results among methods, and the receiver operating characteristics curve analysis for reliability of the CCM. Percentages of LYM (2.67x) and MAC (3.00x) were greater when evaluated using BHM than with CCM (P < 0.05 for both). The agreement (Cohen's kappa value) of results among methods was 0.79 +/- 0.06. The sensitivity (Se) and specificity (Sp) of the CCM for defining endometrial inflammation were 79.3% and 100%, respectively. Among inflammatory cells, proportions of LYM and MAC in the CCM had merit for evaluation of uterine inflammation, with an Se of 74.1 and 84.5 and an Sp of 93.0 and 75.4 at the cut-off > 4 and > 0, respectively. The results indicate the CCM may be used in the diagnosis of SCE when the LYM and MAC percentages are considered in chronically infertile cows in the later stages of the lactational period.