Value of counting colonic mucosal Ig-containing cells in the differential diagnosis of chronic inflammatory bowel disease.

AIMS To investigate whether counting cells containing immunoglobulin (Ig) subclass in colonic biopsy specimens of patients with chronic inflammatory bowel disease, in addition to conventional histological evaluation, can improve the differentiation of patients with Crohn's disease from those with ulcerative colitis. METHODS The colonic and rectal biopsy specimens of 40 patients with chronic inflammatory bowel disease, comprising 20 patients with Crohn's disease and 20 with ulcerative colitis, were used and sections were stained specifically for IgA, IgM, and IgG heavy chains using an indirect immune peroxidase method. The immunoglobulin subclass containing cells were counted using an ocular grid counting method in a light microscope. A linear stepwise discriminant analysis was performed on Ig subclass containing cell counts in combination with 16 reproducible histological features. The results of this discriminant analysis were compared with the results of the discriminant analyses in which only the histological features were used. RESULTS Applying stepwise discriminant analysis, two histological features (an excess of histiocytes in the lamina propria and the villous or irregular aspect of the mucosal surface) in combination with IgMax were selected as the most discriminatory parameters that distinguish Crohn's disease from ulcerative colitis. IgMmax was defined as the maximum value of the mean percentage of IgM containing cells over all the biopsy locations. The use of this combination resulted in a better classification in 20% of the patients with Crohn's disease and in 9% of the patients with ulcerative colitis compared with the use of histological features alone. CONCLUSIONS Morphometric enumeration of Ig subclass containing cells in colonic mucosal biopsy specimens has diagnostic value as a means of differentiating individual patients with Crohn's disease from those with ulcerative colitis.