SURGICAL ASPECTS OF PATIENTS WITH ADENOCARCINOMA OF THE STOMACH OPERATED ON FOR CURE

Background and Design: A retrospective study was performed to evaluate our recent results of curative gastric resections for adenocarcinoma. Methods: Between 1979 and 1988, 187 patients fulfilled study entry cr iteria. This group of patients composes 64% of all patients with tumor s arising distal to the gastroesophageal junction. Tumors arising in t he region of the gastroesophageal junction were excluded. Patients wer e classified according to the American Society of Anesthesiologists ph ysical status classification (greater than or equal to 3, 56%) and Eas tern Cooperative Oncology Group performance status (greater than or eq ual to 2, 44%). Histologic characteristics were re-reviewed. Intervent ions: Subtotal and total gastrectomies were performed in 78% and 22% o f the patients, respectively. Extended lymph node dissections were per formed selectively (5%). Adjuvant chemotherapy and radiotherapy were e mployed in 3% and 2% of patients, respectively. Results: Postoperative morbidity and mortality were 27% and 4%, respectively. Synchronous sp lenectomy (P = .06) and type of gastric resection (P = .06) showed a b orderline association with postoperative complications, but did not af fect postoperative mortality. With a median follow-up time of 47 month s in all patients, and a median of 9 years in patients still alive, th e 5- and 10-year overall survival rates (Kaplan-Meier method) were 48% and 32%, respectively. In univariate survival analysis, age, American Society of Anesthesiologists classification, stage, tumor diameter, s erosal extension of tumor, lymph node metastases, and type of resectio n showed prognostic significance. In the Cox multivariate analysis, ho wever, only serosal extension of tumor (P < .001) and lymph node metas tases (P = .02) were independent prognostic factors. Conclusions: Desp ite the older age and comorbid conditions of patients with gastric can cer, 5-year survival was achieved in half the patients by standard rad ical operations. Until appropriate controlled prospective studies are performed, total gastrectomy, splenectomy, and extended lymph node dis section should not be routinely adopted, given their unproven efficacy and potentially increased morbidity and mortality.