Aim: Many centres now include a segment of transplanted colon with an intestinal graft as International Registry data show this is associated with less dependence on parenteral support and improved graft survival^[1]. We here compare the diagnosis of acute cellular rejection (ACR) based on biopsies (Bx) performed simultaneously in the small bowel (SB) and colonic grafts (paired Bx) after intestinal transplantation (ITx).

Method: We undertook a retrospective review of patients with Intestinal containing grafts  at Addenbrooke’s hospital from 2007 to 2016 who developed ACR. Paired bx were reviewed before and after the ACR by an experienced GI pathologist and were graded based on the VIII International Small Bowel Transplant Symposium Consensus criteria with minor modifications for evaluation of colon biopsies. Each Bx was graded either negative or positive for ACR. If the histology suggested borderline paired with normal, this was counted as normal as it was likely not to be clinically relevant. However, if borderline was paired with mild this was included as a discordance as it was considered clinically relevant. 

Results: During the study a total of 71 ITx were performed in 65 patients. Of these, there were 32 cases of ACR in 30 patients, 18 of which had colon containing grafts. There were a total of 221 paired Bx included in the analysis. Of the 221 paired Bx, the number of concordant normal Bx was 153  (69%) & concordant abnormal Bx was 34 (16%). The total number of discordant Bx was 33 (15%). Among these discordant Bx, the number demonstrating ACR in the small bowel but not the large bowel was 19 and the converse was true in 14 cases. The overall concordance rate was 82.25%. The rate of discordance in ACR was 17%. If the large bowel demonstrated rejection there was a 71% positive predictive value of the small bowel also demonstrating rejection. If the large bowel was normal then the negative predictive value for the small bowel also being normal was 89%. If the small bowel demonstrated rejection then the positive predictive value of the large bowel demonstrating rejection was 64%. If the small bowel was normal then there is a 92% negative predictor value for the large bowel also being normal

Conclusion: Paired Bx in the SB and the colon are usually in agreement regarding the presence or the absence of ACR. However, colonic Bx alone may not suffice to exclude ACR following ITx.

[1] Grant et al. Intestinal Transplant Registry Report: Global Activity and Trends American Journal of Transplantation 2015;15:210-219