DNA aneuploidy in ulcerative colitis: reproducibility, topographic distribution, and relation to dysplasia.

Fifty-nine patients with longstanding, total ulcerative colitis were followed up in a prospective colonoscopic surveillance program. Biopsy specimens were sampled from predetermined locations of the colon and rectum at regular intervals. All specimens were assessed for histological dysplasia and, by flow cytometry, for detection of DNA aneuploidy during 8 years of follow-up. Special emphasis was made to correlate the findings of DNA aneuploidy with findings of dysplasia at colonoscopy or, in case proctocolectomy was performed, in the surgical specimen. Fifteen patients (25.4%) had DNA aneuploidy detected at least once during the follow-up. Eight of 10 patients with repeated findings had consistent ploidy level of the aneuploid peaks from one examination to another. Ten patients had multiple peaks. DNA aneuploidy tended to become more widespread in the bowel during the follow-up but persisted in the same part(s) of the colon and rectum. DNA aneuploidy occurred before development of definite dysplasia in 6 patients, simultaneously with development of dysplasia in 6 patients, and after the development of dysplasia in 1 patient only. In 2 patients, single aneuploid peaks were detected once but could not be found again at subsequent examinations. Dysplasia correlated closely topographically to DNA aneuploidy, but the latter finding was more common without concomitant dysplasia. Only in 1 patient, and at one examination, definite dysplasia was recorded without findings of DNA aneuploidy. Detection of DNA aneuploidy in patients with ulcerative colitis is persistent and reproducible and closely correlated to dysplasia. Widespread changes indicate that the entire colorectal mucosa is at increased risk of malignant transformation. Changes in nuclear DNA content appear to be an earlier phenomenon than dysplasia in the malignant transformation of the colorectal mucosa in ulcerative colitis, and the use of flow cytometry in surveillance programs may be of value for selection of patients at high risk of developing colorectal carcinoma.