PELVIC RESECTION OF RECURRENT RECTAL-CANCER

Objective The authors describe their experience with pelvic resection of recurrent rectal cancer with emphasis on patient selection for cura tive intent based on known tumor risk factors. Summary Background Data Pelvic recurrence is a formidable problem in 30% of patients who have undergone a curative resection of primary rectal cancer. Although rad iation can reduce the development of local recurrence and can provide palliation to many patients with localized disease, it is not curative . The authors and others have used the technique of abdominal sacral r esection (ABSR) with or without pelvic exenteration to resect pelvic r ecurrence and its musculoskeletal extensions in selected patients with satisfactory long-term survival.Methods The technique of ABSR with or without pelvic exenteration or resection of pelvic viscera, which the authors have described previously, was used in 53 patients with recur rent rectal cancer-47 patients for curative intent and 6 for palliatio n. Previous surgeries were abdominal perineal resections (APRs) in 26 patients, anterior resections in 19 patients, and other procedures in 2 patients; original primary Dukes' stage was B in 52% and C in 48%. A lmost all patients had been irradiated previously, generally in the 40 00 to 5900 cGy range. Preoperative carcinoembryonic antigen (CEA) leve ls (before ABSR) were elevated (>5 ng/mL) in 54%. Results Postoperativ e morbidity was encountered in most patients. Mortality was 8.5% in th e curative group. Long-term survival for 4 years was achieved in 14 of 43 patients (33%), and 10 patients were alive with an acceptable qual ity of life after 5 years. Patients who had previous anterior resectio ns or whose preoperative CEA levels were less than 10 ng/mL had a surv ival rate of approximately 45%, whereas patients with previous APRs an d preoperative CEA levels greater than 10 ng/mL had a survival rate of only 15% to 18%. Patients with bone marrow invasion, positive margins , or pelvic node metastases had a median survival of only 10 months. C onclusions Pelvic recurrence of rectal cancer can be resected safely w ith expectation of long-term survival of 33%. Patient selection based on known risk factors can identify patients most likely to benefit fro m resection and eliminate those who should be treated for palliation o nly.