A comparison of end and loop colostomy for fecal diversion in gynecologic patients with colonic fistulas

Objective: The purpose of this study was to evaluate the early morbidity, stomal complications, and overall survival rate of gynecologic oncology patients treated with end or loop colostomy for management of colonic fistula. Methods: Seventy-five women with fistulas undergoing fecal diversion procedures from 1983 to 1993 were identified. Information relevant to tumor history, type of fistula, operative procedure, complications, and outcome was extracted from the medical records. Results: A recto- or sigmoid-vaginal fistula was present in 55 patients. The remaining 20 patients had either a colonic fistula involving the bladder, uterus, or skin or multiple fistulas. In 66 (88%) patients, the pelvis had been irradiated. Loop colostomy was performed in 47 (63%) patients and end colostomy in 28 (37%) patients. Cancer was present in 31 (66%) of 47 patients undergoing loop colostomy compared with 12 (43%) of 28 patients undergoing end colostomy (P = 0.05). No significant differences were detected with regard to age at cancer diagnosis or age at the time of colostomy, body habitus, medical condition, type of cancer, history of pelvic irradiation, or incidence of multiple or complex fistulas. Median operative time was increased by 1 hr and median blood loss from 50 to 200 ml with end colostomy. No significant advantage of loop colostomy over end colostomy was detected in the incidence of acute morbidity or hospital stay. Early stomal complications occurred in 8 patients, 7 of whom had loop colostomy (P = 0.25). Delayed complications (prolapse, retraction, stool per vagina, or fistula) following loop colostomies occurred more frequently than following end colostomy: 12 (26%) of 47 compared with 2 (7%) of 28 (P = 0.05), respectively. Five loop stomas were later converted to end stomas. In addition only one of four attempted loop colostomy closures was successful. Overall survival in both groups of patients was similar. Conclusion: There were no statistically significant differences between patients treated with loop or end colostomy with regard to early morbidity or survival. The early and frequent occurrences of stomal complications after loop colostomy overshadow the clinically minor advantages of this method of fecal diversion. Although loop colostomy was performed more often in this setting, these data do not support its routine use for the management of gynecologic oncology patients with fistulas.