it was prospective. In the second half of the study patients diagnosed on initial biopsy specimen as having type III intestinal metaplasia were followed with endoscopy and biopsy at six to 12 month intervals.A total of 718 gastrectomies for gastric cancer were performed at Guy's Hospital between 1976 and 1987. The clinical notes and histology reports were reviewed in all cases. From this group two subgroups of patients were identified -one (n=24) with early gastric cancer and the other (n=694) with advanced disease. Early gastric cancer was defined as malignancy confined to mucosa or submucosa, with or without regional lymph node involvement.'2 The histology of patients with early gastric cancer was reexamined and survival data were obtained and analysed.All material was fixed in 10% formol saline and routinely embedded in paraffin wax. Serial

SUMMARYThis double blind randomised study tested the effectiveness of colloidal bismuth subcitrate (De-Nol)

No abstract

We studied 20 patients with typical symptoms of gastro-oesophageal reflux, to determine whether the diagnostic accuracy of 24-h pH monitoring might be retained in a test using a shorter time. The duration and number of reflux episodes were used to calculate a frequency and duration score (FDS) for the 24-h period, the daytime period, the nighttime period, a 3-h postprandial period after eating a test meal (provocation test), and a 21-h period that excluded this 3-h postprandial period. The daytime FDS was significantly higher than nighttime FDS (p less than 0.02), and there was good correlation between the 21-h FDS and the 3-h postprandial FDS (rs = 0.695; p less than 0.01). We conclude that the 3-h postprandial testing after a standard meal provocation test is a practical, accurate, and well-tolerated method of diagnosing gastro-oesophageal reflux. The results also demonstrated the relative importance of daytime as opposed to nocturnal reflux in producing symptoms and oesophagitis.

No abstract