Long-term failure after restorative proctocolectomy for ulcerative colitis

Abstract

Objective: To establish the incidence and causes of late failure in patients undergoing restorative proctocolectomy for a preoperative diagnosis of ulcerative colitis was the objective of this investigation.

Summary background data: Restorative proctocolectomy is the elective surgical procedure of choice for ulcerative colitis. Most patients have a satisfactory outcome but failures occur. The reasons and rates of early failure are well documented, but there is little information on long-term failure.

Methods: A series of 634 patients (298 females, 336 males) underwent restorative proctocolectomy for inflammatory bowel disease between 1976-1997, with a mean follow-up of 85 +/- 58 months. Failure was defined as removal of the pouch or the need for an indefinite ileostomy. It was divided into early and late, occurring within 1 year or more than 1 year postoperatively. RESULTS There were 3 (0.5%) postoperative deaths, leaving 631 patients for analysis. Of these, 23 subsequently died (disseminated large bowel cancer, 12; unrelated causes, 9; related causes, 2). There were 61 (9.7%) failures (15 early [25%], 46 late [75%]) due to pelvic sepsis (32 [52%]: 7 early, 25 late), poor function (18 [30%]: 2 early, 16 late), pouchitis (7 [11%]: 2 early, 5 late) and miscellaneous (4, all early). A final diagnosis of Crohn's disease, type of reservoir (J,S), female gender, postoperative pelvic sepsis and a one-stage procedure were significantly associated with failure. Failure rate rose with time of follow-up from 9% at 5 years to 13% at 10 years.

Conclusions: Pelvic sepsis and poor function were the main reasons for later failure. Failure rates should be reported based on the duration of follow-up.

FIGURE 1. Number of operations and failures by year, 1976–1997.

FIGURE 2. Incidence of failure expressed as a percentage by year, 1976–1997.

FIGURE 3. Life-table analysis of failure (Kaplan–Meier).