Multivariate analysis of the risk of stomach cancer after ulcer surgery in an Amsterdam cohort of postgastrectomy patients.

Although gastric cancer incidence is decreasing in the western world, it remains an important cause of death, and there has been debate about screening persons who have undergone gastrectomy for benign ulcers. The authors analyzed risk factors for stomach cancer mortality in an Amsterdam cohort of 2,633 postgastrectomy patients, followed from their initial surgery between 1931 and 1960 until 1975, with 99.7% complete follow-up. Increased stomach cancer mortality was observed in the study population (compared with the general Dutch population) among males 25 years or more after surgery (observed/expected, 5.0; 95% confidence interval (Cl) 2.8-8.3), and among females 15-24 years postoperatively (observed/expected, 3.5; 95% Cl 1.0-9.0). A multivariate Poisson regression analysis showed that after control for age at time of surgery and calendar year of operation, the most important risk factors were time since surgery (0-4 years, relative risk (RR) = 1.0; 5-14 years, RR = 4.1, 95% Cl 0.93-18.5; 15-24 years, RR = 9.4, 95% Cl 2.1-42.3; and 25-46 years, RR = 55.6, 95% Cl 11.7-265.4) and ulcer location (gastric versus duodenal ulcer, RR = 2.6, 95% Cl 1.4-4.8). Surveillance for postgastrectomy cancer could be considered 15-25 years after a patient undergoes surgery for gastric ulcer disease.