The diagnosis and prevalence of subclinical endometritis in cows evaluated by different cytologic thresholds.

The aims of our study were to determine (1) how the prevalence of cytologically determined subclinical endometritis varies when using three different cytological threshold ratios to categorize cows as either with or without endometritis, (2) how the number of animals categorized as having endometritis changes from the fourth to the sixth wk postpartum when using each threshold, (3) how subclinical endometritis influences the number of days open, and (4) how the results of cytological and bacterial examinations correlate. To answer these questions, 222 clinically healthy cows in two herds were examined in the fourth (Exam 1) and the sixth wk (Exam 2) postpartum, when endometrial surface scrapings for bacteriologic and cytologic examination were collected by cytobrush from their uterine horns. After each examination, all cows were categorized using three different thresholds: (1) > 18% polymorphonuclear leucocytes in Exam 1 and > 10% in Exam 2, (2) > 8% in both exams, and (3) > 5% in both exams. It was found that: (1) The number of cows categorized as having endometritis increased as the threshold was lowered, and ranged from 18.9% to 75.4% according to herd, time of examination, and the threshold used; (2) with all three thresholds and in both herds, the number of cows categorized as having endometritis in Exam 1 was approximately double that in Exam 2; whereas depending on the herd and the threshold used, 6.1% to 17.0% of the cows that were negative in the first exam were positive in the second, and 7.4% to 33.3% were positive in both exams; (3) cows were open for a significantly greater number of days if categorized as having endometritis with the first threshold in Exam 1 (mean ± SEM 151.5 ± 9.5 vs. 115.9 ± 7.8; P < 0.01), or with either the first or the second threshold in Exam 2 (mean ± SEM 155.0 ± 15.0 vs. 125.1 ± 6.6; P < 0.05); and (4) the most common bacteria were Streptococcus acidominimus and Escherichia coli, and the correlation between cytologic and bacteriologic findings was low (Φ = 0.08 to 0.17 for different tested thresholds). Subclinical endometritis seems to be associated more with the postpartum recovery of the endometrium than with bacterial infection.