P716 Clinical significance of residual non-rectal inflammation in ulcerative colitis patients with clinical remission

J. Shin, G. Seong, J.H. Song, S.M. Kong, T.J. Kim, E.R. Kim, S.N. Hong, D.K. Chang, Y.H. Kim

Samsung Medical Center, Internal Medicine Gastroenterology and Hepatology, Seoul, Republic of Korea

Background

The advancement of therapeutic agents has made it possible to achieve endoscopic remission in inflammatory bowel disease. Consequently, the treatment goal of ulcerative colitis (UC) also has been changed to achieve endoscopic remission (ER). However, there was insufficient clinical evidence of whether a step-up treatment should be performed to achieve ER in clinical remission (CR) without ER. And there is inadequate data on the need to consider the distribution and severity of residual inflammation in UC. This retrospective study aimed to evaluate the prognostic significance (such as step-up therapy, hospitalisation, and colectomy) of the distribution and severity of residual inflammation in UC patients with CR.

Methods

A total of 134 UC patients who underwent endoscopic evaluation in CR and underwent colonoscopy more than 3 times between January 2000 and December 2018 were retrospectively reviewed. Patients were allocated by endoscopic healing state and distribution of inflammation to an ER (n = 33, 24.6%), residual non-rectal inflammation with patchy distribution (NRI) (n = 17, 12.7%) or residual rectal involvement with continuous or patchy distribution (RI) (n = 84, 62.7%). We reviewed the patient’s characteristics, endoscopic findings and ascertain poor outcome-free survival (PFS) until June 2019.

Results

In UC patient with CR, the PFS was significantly better in ER and NRI (p = 0.003). ER and NRI had similar PFS (p = 0.683). The baseline clinical characteristics of NRI and RI were not significantly different except for the pattern of residual inflammation (p < 0.001). Multivariate analysis showed that NRI was a good prognostic factor of PFS for UC with CR Like ER (hazard ratio 0.53 (0.05–6.30), p = 0.615).

Table 1. Significant predictors of poor outcome

VariablesUnivariate analysisMultivariate analysis
Distribution of residual inflammationHR (95%CI) pHR (95%CI)p
ER11
NRI0.63 (0.60–6.51)0.6940.53 (0.05–6.30)0.615
RI6.15 (1.74–21.82)0.0055.16 (1.13–23.61)0.034

Conclusion

There was no statistically significant difference in the PFS between ER and NRI in the CR state of UC patients. Therefore, we propose selective escalation of treatment modality in CR patients, even if they do not reach ER.