Ethnic survival differences after gastrectomy for gastric cancer are better explained by factors specific for disease location and individual patient comorbidity

Introduction: Different outcomes after resection of gastric cancer between various ethnic patient groups have been described. It remains unclear whether disparity of treatment forms, disease-related variables, or individual patients accounts for this effect. Methods: In the 10 years between 1989 and 1999, 75 patients with gastric adenocarcinoma underwent gastrectomy at a single institution, with constant surgical standards during this time period, including complete (RO) resection attempt and extended lymphadenectomy. Ethnicity, disease characteristics, and treatment variables were analysed for their impact on survival. Results: There were 40 males and 35 females, with a median age of 67 years (range 31-97). The gastrectomy extent was total (n = 25), proximal (n = 18), subtotal (n = 17), distal (n = 14), and segmental (n = 1). The mean lymph-node count was 25 +/- 17 (SD). There was one post-operative death, and an overall complication rate of 27%; the median hospital stay was I I days. Overall actuarial 5-year survival was 33% (95% Cl: 19-47); potentially curable disease (stage IA-IIIB) led to a median survival of 49 months. Asian (n=18) and Hispanic patients (n=20) had significantly better survival than Caucasian (n=31) or other patients (n=6) (P=0.01). Ethnicity was linked to the location of the primary tumour (P=0.002), the gastrectomy extent (P=0.003), and the patient's prior abdominal operation (P=0.01) or tobacco history (P=0.03), but not to resection extent parameters (such as number of lymph nodes retrieved) or differences in pathologic characteristics. When controlling for differences of disease site, stage, R status, and patient comorbidity, ethnicity did not retain an independent prognostic impact on survival. Conclusions: Obvious survival differences after gastrectomy for gastric adenocarcinoma favouring Asian and Hispanic patients in this experience can be explained by different disease patterns (distal location), the related need for fewer extensive procedures (such as total gastrectomy), and diminished patient risks (tobacco, prior operations, noncancer deaths). Our therapeutic approach remains an aggressive gastrectomy/lymphadenectomy combination for potentially curable gastric cancer, irrespective of ethnic patient factors. (C), 2002 Elsevier Science Ltd.