Search in the Abstract Database

Abstracts Search 2014

* = Presenting author

DOP088. Defunctioning ileostomy does not prevent anastomotic leaks after restorative procto-colectomy with ileal pouch-anal anastomosis in patients treated with anti-TNF and steroids

S. Sahami1, C. Buskens1, T. Young-Fadok2, A. de Buck van Overstraeten3, A. D'Hoore3, W. Bemelman1, 1Academic Medical Center, Surgery, Amsterdam, Netherlands, 2Mayo Clinic, Surgery, Phoenix, United States, 3University of Leuven, Surgery, Leuven, Belgium

Background

Anastomotic leakage (AL) is a serious complication after restorative proctocolectomy with ileal pouch-anal anastomosis (IPAA) that may lead to pelvic sepsis, poor pouch function and ultimately to pouch failure. Previous studies have shown significantly decreased leak rates in diverted patients with less severe clinical consequences. The last decade, a trend has been seen towards more extensive medical treatment in IBD patients, leaving refractory patients in a worse condition when it comes to surgery. Therefore, the aim of this study is to analyse whether a defunctioning ileostomy should be considered as standard care in patients undergoing IPAA.

Methods

In a retrospective study, 621 patients undergoing IPAA for ulcerative colitis, indeterminate colitis or Crohn, were identified from prospectively maintained databases of three large IBD centres. The creation of an ileostomy was left at the discretion of the surgeon. AL was defined as any leak confirmed by either contrast extravasation on imaging or by re-laparotomy (grade A; antibiotics, B; drainage, and C; re-laparotomy).

Results

In 305 patients (49.1%), an ileostomy was created during IPAA. A comparable leak rate was found in the stoma group when compared to non-diverted patients (16.7% vs 17.1%, p = 0.92). As expected, a significantly higher number of patients required re-laparotomy in the non-diverted group (10.4% vs 4.3%, p = 0.003), but there was no difference when combining the clinically relevant leaks (grade B and C; 15.2 vs 13.8, p = 0.65). This unexpected finding of high leak rates despite stoma formation could probably be explained by the increased use of anti-TNF (12.6% versus 4.6%, p < 0.0001), steroids (33.0% vs 12.1%, p < 0.0001), and weight loss (>5% of total bodyweight) (14.6% vs 8.5%, p = 0.02) when compared to non-diverted patients. Multivariate analysis confirmed preoperative anti-TNF and steroid use as independent factors associated with the creation of an ileostomy. Despite having a protective stoma, an extreme high leak rate (40.0% vs 15.1%, p = 0.02) was found in patients treated with a combination of anti-TNF and steroids. The importance of these predictive variables was emphasized by the fact that patients undergoing subtotal colectomy and completion proctectomy with IPAA at a later stage had a significantly decreased leak rate when compared to patients undergoing primary IPAA (11.6% vs 20.7%, p = 0.003).

Conclusion

These results imply that in daily practice surgeons perform loop ileostomy in more fragile and disease affected patients. This strategy seems ineffective in the prevention of AL in these series implicating that a staged procedure, that is subtotal colectomy followed by completion proctectomy and IPAA after weaning of the medication, is more appropriate.