Pelvic exenteration: factors associated with major surgical morbidity.

Sixty-nine women underwent pelvic exenteration at Duke University Medical Center from 1970 through 1987. The operative mortality rate was 7.2% with a trend toward a reduction during the course of the study. One or more serious gastrointestinal or genitourinary surgical complication occurred in 26 (38%) patients and 20 (29%) required reoperation for these complications. There was a trend (P less than 0.1) toward an increase in surgical complications among patients who received prior radiation therapy and those requiring urinary diversion, with a decrease among those who underwent gracilis flap pelvic reconstruction. Patients with sigmoid or ileal conduits had a significantly higher incidence of severe surgical complications than those with transverse colon conduits or posterior exenteration alone (P less than 0.05). Those in whom an ileal conduit was constructed without gracilis flap pelvic reconstruction had significantly more surgical morbidity compared to those who underwent pelvic reconstruction or received a transverse colon conduit (P less than 0.05). Multiple changes in technique since 1978 including (1) the routine use of surgical staplers for bowel resection and anastomosis, (2) the introduction of the transverse colon conduit, and (3) the use of gracilis flap for pelvic reconstruction have combined to produce a significant (P less than 0.05) decrease in life-threatening surgical complications.