The risk of missed gastroesophageal cancer diagnoses in users and nonusers of antisecretory medication

Background & Aims: Some patients with early gastro-esophageal cancer may appear to "heal" because of antisecretory medication, but the risk of a missed diagnosis is unknown. The aim of the study was to estimate the incidence of gastroesophageal cancer with or without pre-endoscopic treatment with antisecretory medication. Methods: We extracted data on use of endoscopies, gastroesophageal cancer diagnoses, death, migration, and use of antisecretory medication (H-2 blockers and proton pump inhibitors) from 5 population-based registries covering 1974-2002. We included all citizens in Funen County (population, 470,000) who between 1993 and 2002 were investigated by endoscopy for the first time. The patients were followed up until death, emigration, or the end of the study period. Results: Among 27,829 patients with a first endoscopy (mean age, 56 years; 48% male, 115,804 person-years of follow-up), 461 had gastroesophageal cancer diagnosed at the first endoscopy and 52 were diagnosed during a median follow-up of 2.7 years after the first endoscopy. The incidence during follow-up was similar to the background population (standardized incidence ratio, 1.24; 95% confidence interval, 0.81-1.91), increased with age, and was higher in male patients. The incidence of gastroesophageal cancer during follow-up was 46 per :100,000 person-years in users of antisecretory medication the last 180 days before the first endoscopy compared with 44 per 100,000 person-years in nonusers (age and sex standardized difference, 4 per 100,000 person-years; 95% confidence interval, -14 to 22). Conclusions: Very few cancers are missed at endoscopy. The risk seems similar in users and nonusers of antisecretory medication before endoscopy.