An Analysis of Survival and Treatment Failure Following Abdominoperineal and Sphincter-saving Resection in Dukesʼ B and C Rectal Carcinoma

Publisher Website
Google Scholar
Abstract
Abdominoperineal resections for rectal carcinoma are being performed with decreasing frequency in favor of sphincter-saving resections. It remains, however, to be unequivocally demonstrated that sphincter preservation has not resulted in compromised local disease control, disease-free survival, and survival. Accordingly, it is the specific aim of this endeavor to compare local recurrence, disease-free survival, and survival in patients with Dukes' B and C rectal cancer undergoing curative abdominoperineal resection or sphincter-saving resection. For the purpose of this study, 232 patients undergoing abdominoperineal resection and 181 subjected to sphincter-saving resections were available for analysis from an NSABP randomized prospective clinical trial designed to ascertain the efficacy of adjuvant therapy in rectal carcinoma (protocol R-01). The mean time on study was 48 months. Analyses were carried out comparing the two operations according to Dukes' class, the number of positive nodes, and tumor size. The only significant differences in disease-free survival and survival were observed for the cohort characterized by >4 positive nodes and were in favor of patients treated with sphincter-saving resections. A patient undergoing sphincter-saving resection was 0.62 times as likely to sustain a treatment failure as a similar patient undergoing abdominoperineal resection (p = 0.07) and 0.49 times as likely to die (p = 0.02). The inability to demonstrate an attenuated disease-free survival and survival for patients treated with sphincter-saving resection was in spite of an increased incidence of local recurrence (anastomotic and pelvic) observed for the latter operation when compared to abdominoperineal resection (13% vs. 5%). A similar analysis evaluating the length of margins of resection in patients undergoing sphincter-preserving operations indicated that treatment failure and survival were not significantly different in patients whose distal resection margins were < 2 cm, 2–2.9 cm, or ≥3 cm. If any trend was observed, it appeared that patients with smaller resection margins had a slightly prolonged survival (p = 0.10). This observation was present in spite of the fact that local recurrence as a first site of treatment failure was greater in the group with <2 cm than it was in the ≥3 cm category, 22% versus 12%. This increased local recurrence rate in the population with smaller margins was not translated into an increase in overall treatment failure and had absolutely no influence on survival. It is suggested that local recurrence serves as a marker of distant disease. Further, patients undergoing sphincter-saving resections fashioned with mechanical staples fared as well as those subjected to hand-sewn anastomoses. There were no imbalances between the groups for Dukes' class, number of nodes, tumor size, preoperative carcinoembryonic antigen, and type of adjuvant therapy. The data provide the first evidence from a randomized prospective clinical trial that sphincter-saving resection is not associated with an attenuated survivorship or an increased incidence of treatment failure and challenge the prognostic significance of the extent of margins of resection in rectal cancer.