Ulcerative colitis : dysplasia and cancer, surgical treatment, and faecal microbiota transplantation for chronic pouchitis

Background: Ulcerative colitis (UC) is a chronic inflammatory bowel disease of unknown aetiology. UC is associated with increased risk of colorectal cancer (CRC). The risk of CRC has declined over recent decades, but the risk is still approximately 2-fold. CRC is one of the indications for surgery, but most UC patients undergo surgery due to medically refractory disease, suggesting that they often use immunomodulative medication and corticosteroids prior to surgery. The most common operation performed is proctocolectomy with ileal pouch-anal anastomosis (PC-IPAA), avoiding a permanent ileostomy. Controversy exists about whether or not a diverting ileostomy should be constructed. The most common long-term complication of PC-IPAA is pouchitis. Episodes of pouchitis are frequently successfully treated with antibiotics. However, a chronic antibiotic-refractory pouchitis can be difficult to manage. Aims: The dissertation comprises four original studies. The aims of these studies were to determine the incidence and prognosis of UC-associated CRC (Study I), to assess the effect of preoperative anti-tumour necrosis factor (TNF) therapy and corticosteroids on postoperative complications and pouch failure (Study II), to determine the effect of covering ileostomy on postoperative morbidity after IPAA (Study III), and to investigate the efficacy and safety of faecal microbiota transplantation (FMT) in the treatment of chronic pouchitis (Study IV). Patients and methods: All studies were carried out at Helsinki University Hospital, Finland. Studies I, II, and III were retrospective. Study I included all 71 patients with UC-associated CRC who were operated on at Helsinki University Hospital between 1991 and 2018. Moreover, 108 patients with dysplasia in the surgical specimen were analysed. Study III included all 510 consecutive patients who underwent PC-IPAA or proctectomy with IPAA between January 2005 and June 2016. A diverting ileostomy was constructed in 119 of these patients. Study II was a subgroup of Study III that excluded patients who underwent proctectomy or were using tacrolimus or vedolizumab prior to surgery. In total, 445 patients were included. Study IV was a randomized, double-blinded, placebo-controlled trial. We randomly allocated 26 patients in a 1:1 ratio to either donor FMT or autologous transplant (placebo). The recruitment was implemented between December 2017 and August 2018. Results: In Study I, CRC was diagnosed preoperatively in 47 patients (66.2%). Altogether 34 patients (47.9%) had synchronous CRC or dysplasia. The incidence of CRC among patients undergoing surgery has not changed during the 28-year study period (P = 0.113). The overall survival was 71.8%, while the 5-year CRC-specific survival was 81.5%. In Study II, anti-TNF therapy was not associated with postoperative complications. Corticosteroids with a dose equivalent to prednisolone 20 mg or more increased the incidence of anastomotic leak (12.6% vs. 2.5%, P = 0.002) and wound dehiscence (4.2% vs. 0%, P = 0.019), but not pouch failure (2.1% vs. 0%, P = 0.141). Patients with a lower dose of corticosteroids had more pouch failures than patients without corticosteroid treatment (4.4% vs. 0%, P = 0.015). In Study III, patients with a diverting ileostomy had more postoperative complications than patients without ileostomy (55.4% vs. 30.2%, P < 0.0001). Although clinical anastomotic leak was more common in patients without an ileostomy (6.6% vs. 1.7%, P = 0.04), the re-laparotomy rate due to an early complication did not differ between the groups (P = 0.58). Re-admission rate was higher among patients with an ileostomy (42.0% vs. 13.0%, P < 0.0001). Of patients, 43.4% had problems with an ileostomy after discharge. There was no difference in pouch failure rate between the groups (1.7% vs. 2.8%, P = 0.74). In Study IV, relapse-free survival was similar between the groups (log rank P = 0.183; HR 1.90, 95% CI 0.73 to 4.98, P = 0.190). Patients reported no major adverse effects. Conclusions: The incidence of UC-associated CRC has remained constant among patients undergoing surgery. The prognosis of patients with UC-associated CRC was no worse than the prognosis of all CRC patients in the Finnish population. The most important finding in Study I was that one-third of the CRCs were not diagnosed until surgery. Moreover, synchronous lesions were common. These findings are alarming and should be considered before the endoscopic management of UC-associated dysplasia. Anti-TNF agents seem to be safe prior to surgery. By contrast, corticosteroids were associated with a higher incidence of anastomotic leak, wound dehishence, and pouch failure. An ileostomy is associated with a considerable number of complications without any reduction in pouch failure rate. Based on our results, a single-stage PC-IPAA is safe in low-risk UC patients. FMT was not effective in the treatment of chronic pouchitis with the FMT treatment protocol used in our trial. The safety profile of FMT was good with no major adverse effects.