Number of Biopsies Required for Accurate Histological Assessment of Inflammation in Active Ulcerative Colitis

Multiple scoring systems have been developed for histological inflammatory activity of ulcerative colitis and Crohn’s disease. The importance of these validated, quantitative measures has increased as it has become clear that histological analyses are superior to endoscopic assessments in evaluating therapeutic responses and predicting patient outcomes. Similar to studies seeking to determine how many biopsies are needed for diagnosis or dysplasia surveillance, several groups have reported that at least two or three biopsies are required for accurate assessment of inflammatory activity within a single colonic segment. McBride et al now use a unique approach to assess the density of sampling required within a short 2-cm length of colonic mucosa.

Rather than actual biopsies, McBride et al defined 21 contiguous 1-mm diameter “virtual” biopsies within each of two 2-cm regions from each of 11 colectomy specimens. Inflammatory activity of each of the 21 virtual biopsies within each region was assessed histologically using several accepted scoring systems. McBride et al then determined how many biopsies within a 2-cm region were required to accurately determine a single inflammatory score for that region. Agreement statistics improved with increasing numbers of virtual biopsies. The greatest improvement was gained after addition of a third biopsy; however, the benefits of each additional biopsy decreased thereafter. McBride et al therefore concluded that three biopsies are necessary to overcome microscopic heterogeneity and ensure accurate histological activity grading within a 2-cm region.

How these observations can be translated to practice is uncertain, as it would require 150 1-mm biopsies to evaluate a 100-cm colon. Moreover, the relevance to clinical practice is not clear, as the diameter of real-world biopsy specimens is typically approximately 2–3 mm, or four- to nine-fold larger in cross-sectional area, than 1-mm diameter biopsies, and only a small number of specimens were studied. Finally, although interobserver agreement is a key determinant of clinical applicability, it cannot be assessed here because two pathologists worked together to create a single core for each virtual biopsy. Despite these weaknesses, the data do add to the growing chorus indicating that multiple biopsies are required for accurate quantitative analysis of inflammatory activity that can be relied on to guide therapeutic intervention and provide prognostic data. Future work combining strict sampling protocols, multi-observer histological analysis, and patient outcomes will be critical before these can be effectively used for clinical trials and patient management.