IIB or not IIB: Endometrial biopsy evaluation in horses using the Kenney-Doig scale
Date
2021-07-20
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
ORCID
0000-0002-5843-6611
Type
Thesis
Degree Level
Masters
Abstract
The Kenney-Doig scale is a histopathologic grading system used as the international standard for assessing endometrial pathology and communicating prognostic fertility information for equine breeding prospects. The descriptive modifiers used for the scale are potentially subjective and as such they may not produce repeatable results between different observers and may contribute to certain grades of the scale acting as a ‘catch all’. To investigate the frequencies of Kenney-Doig grades assigned at the Western College of Veterinary Medicine and Prairie Diagnostic Services (hereafter referred to as WCVM/PDS), a retrospective analysis of all equine endometrial submissions was completed from records between 1998 and 2018. Of 726 biopsies, the following grading distribution was found: 46/726 (6.3%) as category I, 307/726 (42.3%) as category IIA, 326/726 (44.9%) as category IIB, and 47/726 (6.5%) as category III. For comparison purposes, a retrospective review of the literature was completed and six different studies reporting Kenney-Doig grading distributions were included. Chi-square analysis showed significant differences between the grading distribution found at WCVM/PDS and each grading distribution reported in the six studies. To account for differences in mare populations, individual grading distributions were generated for five pathologists at the WCVM/PDS. Fisher’s exact test between these five Kenney-Doig grading distributions revealed significant differences in grading tendencies, suggesting the presence of observer variation. While comparisons from the retrospective review were limited by differences in mare populations, a prospective study was then completed to directly test inter-rater and intra-rater agreement of eight American College of Veterinary certified pathologists who were recruited to blindly grade the same set of 63 digitized equine endometrial biopsy slides as well as blindly re-evaluating 21/63 of these slides at a later time point. Cohen’s kappa was used to measure inter-rater agreement and kappa values for pairwise comparison of final Kenney-Doig grades ranged from -0.052 to 0.458 (unweighted) and 0.082 to 0.638 (weighted), with an average Light’s kappa of 0.187 (unweighted) and 0.359 (weighted) across all eight pathologists, 0.143 (unweighted) and 0.330 (weighted) for pathologists practicing at different institutions, and 0.216 (unweighted) and 0.464 (weighted) for pathologists at the same institution. Intra-class correlations measuring intra-rater agreement ranged from 0.116 to 0.774 with an average of 0.553 for all eight pathologists. This study shows only slight to moderate inter-rater agreement and poor to good intra-rater agreement is produced by the Kenney-Doig scale, suggesting that the system may be unreliable and subject to significant observer variability.
Description
Keywords
endometrial biopsy, Kenney-Doig, equine
Citation
Degree
Master of Science (M.Sc.)
Department
Veterinary Pathology
Program
Veterinary Pathology