Local recurrence after curative anterior resection with principally blunt dissection for carcinoma of the rectum and rectosigmoid

PURPOSE: The aim of this study was to determine the incidence of local pelv ic recurrence of carcinoma of the rectum and rectosigmoid (tumors where the lower edge is 18 cm or less from the anal verge) in a consecutive series o f patients operated on by a single surgeon. All patients underwent curative anterior resect ion and a formal anatomic dissection of the rectum where m obilization was achieved through a principally careful blunt manual techniq ue along fascial planes, preserving an oncologic package. METHOD: During th e period April 1986 to December 1997, 157 consecutive anterior resections f or carcinoma of the rectum and rectosigmoid were performed by one surgeon ( ALP). One hundred thirty-eight (87.9 percent) were curative, and 19 (12.1 p ercent) were palliative. The mean follow-up period was 46 +/- 31.6 (range, 2-140) months. Data were retrospectively collated and computer coded by an independent contracted medical research team. Follow-up data were available on all patients. RESULTS: Four (3.1 percent) of the 131 patients undergoin g curative anterior resection had local recurrence. Local recurrences occur red between 16 and 38 months from the time of resection, and the cumulative risk of developing local recurrence at five years was 5.2 percent. All tum ors in which pelvic recurrence occurred were high grade, and the probabilit y of developing local recurrence at: five years for this group was 13.9 per cent, which is significantly higher compared with patients who had average or low-grade tumors (P = 0.01). The probability of developing local recurre nce at five pears for Stage I tumors was 0, Stage II was 5.9 percent, and S tage III was 8.9 percent. In addition, there was a significantly higher inc idence of local recurrence in the group of patients undergoing ultralow ant erior resection (between 3 and G cm from the anal verge) as compared with p atients undergoing low or high anterior resection (P = 0.03). Local recurre nce developed in 3 of 28 (10.7 percent) patients having ultralow anterior r esection, 1 of 57 (1.8 percent) patients having low anterior resection (bet ween G and 10 cm from the anal verge), and no patients having high anterior resection (above 10 cm from the anal verge). The clinical anastomotic leak rate for curative anterior resection was 7 of 131 patients (5.3 percent). Thirty-seven of the 131 (28.2 percent) required a proximal defunctioning st oma; 35 (41.2 percent) of these were established for low or ultralow anteri or resections and 2 for high anterior resection. The overall five-year canc er-specific survival rate of the entire group of 131 patients was 81.8 perc ent, and the overall probability of being disease free at five years includ ing both local and distal recurrence was 72.9 percent. Three local recurren ces occurred in the 101 patients (77 percent) who did not receive any form of adjuvant therapy. One local recurrence occurred in the 18 patients (13.7 percent) who had adjuvant chemoradiation. No recurrence occurred in the 12 patients (9.2 percent) who had adjuvant chemotherapy alone. CONCLUSION: Cu rative anterior resection for carcinoma of the rectum and rectosigmoid with principally blunt dissection of the rectum in this study is associated wit h a 3.1 percent incidence and a 5.2 percent probability at five years of de veloping local recurrence. Evidence from this study indicates that, as with sharp pelvic dissection, a low incidence and probability of local recurren ce can be achieved by a principally blunt mobilization technique through ca reful attention to preservation of fascial planes in the pelvis and removal of an oncologic package with selective rather than routine adjuvant or neo adjuvant chemoradiation.