Interobserver variation among histopathologic evaluations of intestinal tissues from dogs and cats

Abstract

OBJECTIVE: To determine whether substantial interobserver variation exists among diagnostic pathologists for descriptions of intestinal mucosal cell populations and whether histopathologic descriptions accurately predict when a patient does not have clinically evident intestinal disease. DESIGN: Comparative survey. Sample Population-14 histologic slides of duodenal, ileal, or colonic tissue from 10 dogs and 3 cats. PROCEDURE: Each histologic slide was evaluated independently by 5 pathologists at 4 institutions. Pathologists, who had no knowledge of the tissues' origin, indicated whether slides were adequate for histologic evaluation and whether the tissue was normal or abnormal. They also identified the main infiltrating cell type in specimens that were considered abnormal, and whether infiltrates were mild, moderate, severe, or neoplastic. RESULTS: Quality of all slides was considered adequate or superior by at least 4 of the 5 pathologists. For intensity of mucosal cellular infiltrates, there was uniformity of opinion for 1 slide, near-uniformity for 6 slides, and nonuniformity for 7 slides. Five dogs did not have clinical evidence of intestinal disease, yet the pathologists' descriptions indicated that their intestinal tissue specimens were abnormal. CONCLUSIONS AND CLINICAL RELEVANCE: Substantial interobserver variation was detected. Standardization of pathologic descriptions of intestinal tissue is necessary for meaningful comparisons with published articles. Clinicians must be cautious about correlating clinical signs and histopathologic descriptions of intestinal biopsy specimens.