Aspects of diverting stoma and ERAS in rectal cancer surgery

Abstract: Annually, nearly 2000 patients are diagnosed with rectal cancer in Sweden. To date, the only known curative treatment is surgery and low anterior resection (LAR) is the operation of choice for tumours in the middle rectum. However, LAR has a high risk for short and long-term morbidity where one of the most severe complications is anastomotic leakage (AL). Since a diverting loop ileostomy has been shown to reduce the risk of early AL after LAR, nearly all patients in Sweden are currently diverted. Yet, a stoma, even temporary, is also associated with significant morbidity. Enhanced Recovery After Surgery (ERAS) is a perioperative care program with the aim to reduce surgical stress and thereby improve postoperative outcome after surgery. The aim of this thesis was to evaluate and optimise the treatment for patients with a diverting stoma following surgery for rectal cancer and to assess if compliance with ERAS influenced clinical outcome after primary diversion in LAR. In paper I, short-term morbidity after LAR in relation to a diverting stoma and ERAS was evaluated. All 287 patients operated on for LAR at Ersta Hospital, Sweden, between 2002-2011, were included. Out of those, 139 had a diverting stoma (S+) at LAR and 148 patients had not (S-), whereas all were treated according to an ERAS program. Most of the diverted patients underwent surgery after 2007 due a change in practice at our institution. Data were prospectively collected in the ERAS database. Postoperative morbidity, including clinically apparent AL, was similar between the two groups, S+ and S-. Total rate of re-laparotomy was comparable but significantly more patients in the S- group underwent re-laparotomy due to AL. However, the total frequency of reinterventions due to AL did not differ. Postoperative recovery was faster among the patients in the S- group but this did not influence the length of stay. In paper II, long-term morbidity within 3 years after LAR depending on whether or not a diverting stoma was fashioned was evaluated. The cohort was the same as in paper I, but data regarding long-term morbidity and permanent stoma were retrospectively collected. Late AL, unexpected readmissions in the late postoperative course, rate of permanent stoma and oncological outcome were comparable between S+ and S-. AL was an independent predictor for a permanent stoma and patients in the S+ group had longer hospital stay during the 3 year follow up. In paper III, complications after closure of a loop ileostomy in relation to the type of anastomosis (hand-sewn or stapled) were analysed. The cohort consisted of 351 patients, operated on for stoma closure, 1999-2006, at three different Swedish hospitals. Data were collected retrospectively. In patients with a stapled anastomosis, the risk of small bowel obstruction after surgery was reduced by 50 percent, operation time was 10 minutes shorter and length of hospital stay was reduced by 1.5 day, compared with patients who received a hand-sewn anastomosis. In paper IV, a total of 29 patients, undergoing rectal cancer surgery between 2008-2013 at Ersta Hospital, were randomised either to oral nutritional supplements (ONS) and rectal enema before surgery or no preoperative nutritional intervention and mechanical bowel preparation with polyethylene glycol (PEG). Bowel cleansing, postoperative morbidity and patients’ nutritional and physiological status were assessed. The bowel was less clean in the right and mid colon but similar in the sigmoid and rectum in the ONS-group. In the interventional arm (ONS), patients gained in percent body fat, from randomisation to 3 days after surgery, and lost less in weight, from randomisation to 4 weeks after surgery, compared to the PEG-group. Postoperative morbidity did not differ. In conclusion, we did not find any benefit of a diverting stoma regarding short and long-term morbidity after LAR among patients treated within an ERAS program. However, there may be an increased risk of symptomatic anastomotic leakage requiring re-laparotomy in those patients who were not diverted. Nonetheless, overall complication rates were similar with and without diversion, which suggests that routine diversion for all patients, undergoing LAR in Sweden, may be called into question. Moreover, a stapled anastomosis during the closure procedure seems preferable, resulting in a reduced frequency of postoperative small bowel obstruction and shortening operative time. Finally, oral nutritional supplements and local rectal cleansing prior to rectal cancer surgery may be a safe alternative to traditional bowel cleansing and in addition improve patients’ nutritional status.