Comparison of cytological and histological examinations in different locations of the equine uterus-an in vitro study.

Endometritis constitutes a major problem in managing broodmares. The histological occurrence of polymorphonuclear neutrophils (PMNs) in the stratum compactum of the endometrium is accepted as the reference standard to diagnose endometritis in mares. The objective of this study was to determine the distribution of PMNs within different sampling locations of the uterus by cytological examinations and to compare it with PMN numbers in endometrial biopsies of the corresponding location. Cytological and endometrial samples were obtained from 37 uteri within 2 ± 1 hours after slaughter through small incisions from five different, predefined locations of each uterus. The cytological samples were smeared on microscopic slides, stained, and classified as negative (<2% PMNs) or positive (≥2% PMNs) for endometritis. Histologically, the numbers of PMNs were counted in three high power fields by an experienced pathologist and classified as positive for this type of endometritis if ≥5 PMNs occurred in three high power fields (×40 magnification). The biopsies were also evaluated for lymphoplasmacellular endometritis, periglandular fibrosis (endometrosis), and angiosclerosis. The prevalence of positive cytological and histological samples was 14.6% and 17.8%, respectively. A fair agreement between the two diagnostic methods could be detected (k = 0.29; P < 0.01). The numbers of PMNs differed between the sampling locations, resulting in positive and negative locations within a positive scored uterus, in both cytologically positive scored uteri (8/10) and histologically positive scored uteri (13/14). No significant differences were found in PMN numbers in the different locations, either the cytological (P = 0.78) or histological (P = 0.79) examination. Additionally, no significant differences were observed in the assessment of endometrosis (P = 0.96) and angiosclerosis (P = 0.67) within the locations. In conclusion, PMN numbers of a cytological examination of the endometrium showed fair agreement to the occurrence of PMN in the stratum compactum of the histological examination at the same sampling location. Although variations were found in the number of PMN using both methods (cytology and histology), statistically significant differences were not detected within the different locations (P = 0.78; P = 0.79), implyingies that the decision to take more than one sample should be critically considered. Additional research is warranted to determine the number of sampling locations necessary for reliable examination results.