Pancolonic Endoscopic and Histologic Evaluation for Relapse Prediction in Patients With Ulcerative Colitis in Clinical Remission

Miyuki Kaneshiro; Kento Takenaka; Kohei Suzuki; Toshimitsu Fujii; Shuji Hibiya; Ami Kawamoto; Maiko Motobayashi; Hiromichi Shimizu; Masakazu Nagahori; Eiko Saito; Ryuichi Okamoto; Kazuo Ohtsuka; Mamoru Watanabe


Aliment Pharmacol Ther. 2021;53(8):900-907. 

In This Article


Mucosal healing assessment is, indeed, a beneficial strategy to guide treatment optimisation in patients with ulcerative colitis. However, the extent of disease activity has not been evaluated often, and the appropriate biopsy sampling methods have not been determined. This prospective study systematically evaluated five colonic segments and explored the colonoscopic strategy for predicting the prognosis of patients with remission. Our results revealed three important findings.

First, the ascending, transverse, or descending colon demonstrated a substantial activity (12.9%) in patients who had no lesions in the sigmoid colon or rectum. In general, inflammation of ulcerative colitis arising from the rectum diffusely or continuously extends to the proximal colon. However, the inflamed mucosa can be atypically distributed, with patchy healing in some patients, especially those on topical treatment,[18] with primary sclerosing cholangitis,[19] and with younger age.[20] Furthermore, in 20%–50% of patients, the extent of endoscopic activity may change during both the natural and the treated course of ulcerative colitis.[21] Based on 482 colectomy results in Horio et al's report, approximately 10% of patients had a rectal-sparing type of ulcerative colitis.[22] Though flexible sigmoidoscopy is often performed, total colonoscopy is the most appropriate examination for ulcerative colitis.[23] Our results demonstrated that evaluating the proximal colon and the disease extent by total colonoscopy is necessary for patients in clinical remission.

Second, pancolonic endoscopic evaluation was strongly associated with relapse prediction in patients in clinical remission. The UCEIS accurately evaluates the mucosal appearance, with high observer agreement, 90% variability in the overall assessment of endoscopic activity,[7] and satisfactory prediction of the outcome.[9] However, the UCEIS, as well as the previously reported endoscopic scores, only macroscopically evaluate the most severely involved area and do not consider the extent of the disease. In our ROC curve analysis, pancolonic UCEIS, which is the sum of five segmental scores, best predicted patient's one-year relapse. A similar concept of evaluating the disease extent was discussed in previous studies;[24,25] however, the feasibility and simplicity of this concept are not yet proven in clinical practice. In our study, we applied the UCEIS, which is a valid scoring system that is widely used as the endoscopic index both in clinical trials and clinical practice. We believe that pancolonic UCEIS can help evaluate the proximal lesions and the extent of the disease.

Third, we systematically collected and scored all colonic biopsies and showed that histological assessment is important in clinical and endoscopic remission. In addition, patients who achieved clinical and endoscopic remission may still have histologically active disease (Geboes >3.0), which is an independent risk for relapse (Table 2). Two recent studies have specifically examined the link between histological remission and the outcome of ulcerative colitis and found that histological activity was associated with an increased chance for clinical relapse.[11,26] Compared with previous studies, our study had a larger sample size and had the strength of taking systematic biopsies. Notably, the accuracy of combining pancolonic UCEIS and histological evaluation was sufficiently high for predicting patient outcomes.

Considering the treatment cost is vital when managing ulcerative colitis.[27] The pancolonic assessment was found to be statistically superior, but its cost was not evaluated in this study. The task of summing the five segments can be cumbersome, but further study is needed to determine its clinical significance. However, the substantial inflammation in the proximal colon and the adjusted HRs of each score found in this study convince us that total colonoscopy should at least be performed in patients with remission. In addition, our study made the protocol of conducting five biopsies; however, future studies should not only examine the cost but also determine how many biopsies should be performed clinically.

This study also has some limitations. Although recommended for the calculation of the UCEIS and especially for clinical trials, video recording of the endoscopies was not performed. We recognise its importance; thus, developing a computer-aided endoscopic video system relating to the present study has been planned.[28] Second, the physicians in charge of the patients were not blinded to the endoscopic findings. We enrolled patients in clinical remission, and none received additional treatment immediately after colonoscopy. Nevertheless, unknown bias was inevitable; hence, the prognosis results should be interpreted carefully. Third, we did not evaluate calprotectin during colonoscopy, and a comparison between biomarker assessment and histological and endoscopic assessments could have strengthened this study. Finally, the generalisability of the data is uncertain because this study was conducted in a single-centre in a tertiary hospital setting, where experts manage patients. A multicentre cohort study is necessary to validate the pancolonic scoring system.

In conclusion, evaluating the extent of the disease by total colonoscopy with histology is relevant in patients in clinical remission. Furthermore, the combination of pancolonic endoscopic and histological evaluations may represent the highest predictive value for the prognosis of these patients.