Three hundred and six (94%) of all consecutive patients with prepyloric or duodenal ulcer disease undergoing highly selective vagotomy at a District General Hospital were followed up after 1 to 10 completed years of observation. The 5- and 10-year cumulative recurrence rates were 11.6% and 16.8%, respectively. These figures were not related to age, sex, duration of ulcer disease, or preoperative peak acid output. Prepyloric ulcers had a significantly higher recurrence rate than duodenal ulcers during the first 5 years but this difference was eliminated at 10 years. The recurrence rates varied highly significantly between different surgeons. The 5-year recurrence rate in patients operated during the first 5-year period amounted to 13.4% and was steadily increasing. A corresponding figure for those operated during the second 5-year period was 5.3% and remained constant after 3 years of observation. This difference might reflect an improved surgical technique initiated by repeated evaluation of the clinical results.

Bleeding is a potential but rare complication after endoscopic biopsies of the gastric mucosa in non-operated patients. The resected stomach, however, seems to run an increased risk of this complication. In a retrospective study comprising 1843 consecutive biopsy examinations of the gastric remnant, 15 bleeding episodes were recorded (0.8%). The frequency of bleeding was 1.1% for the Billroth-II-resected stomach and 0.2% for the Billroth-I-resected stomach. In two patients the bleeding was excessive, requiring emergency surgery. Six other patients required blood transfusions. The observed frequency of bleeding was probably a minimal estimate, suggested by a 20% occult bleeding found in a small prospective study. When planning follow-up examination of patients who have had a gastrectomy, the risk of bleeding should be taken into account, and the patients should be appropriately informed and supervised.

The aim of this study was to assess the clinical results after highly selective vagotomy (HSV) when used routinely at a district general hospital. A total of 229 patients with chronic duodenal or prepyloric ulcers was included in a 92 per cent complete follow-up 1-6 years after an elective HSV without drainage. There was no postoperative mortality, and the frequency of postoperative complicatons was low. Transient dysphagia and early fullness each occurred in about a quarter of the patients, whereas persistent dumping was found only in 1.3 per cent and diarrhea in 2.2 per cent of the patients. Recurrent ulcer was diagnosed in 12.7 per cent of the patients, but this figure decreased to 9.3 per cent after exclusion of one of the 24 surgeons involved. The overall clinical results according to the modified Visick classification were recorded as excellent or very good in 70 per cent and unsatisfactory in 20 per cent. The failures were almost exclusively due to a proved or suspected recurrence or to gastric retention, and further operations have been performed on 27 of these patients. After reoperation only 8 patients (3.5 per cent) remained failures according to the patient's own judgement at the time of follow-up. The recurrence ratein this study was considered disquietingly high with regard to the short observation time. However, other obvious advantages of the method and the possibility of improved results after adjustment of the surgical technique were considered to justify continued use of HSV as a routine procedure.

Fasting plasma levels of immunoreactive gastrin, somatostatin, and pancreatic polypeptide (PP) were determined in 67 patients with an endoscopically proven duodenal, pyloric, or prepyloric ulcer. Pretreatment gastrin (45.6 +/- 53.6 pmol/l, mean +/- S.D.) and somatostatin (54.5 +/- 27.5 pg/ml) did not differ significantly from those in 22 healthy controls (21.8 +/- 20.2 pmol/l and 64.8 +/- 24.7 pg/ml, respectively). The gastrin and somatostatin levels were not changed by 3 weeks of treatment either with propantheline and antacids or with cimetidine and antacids. The mean PP value before treatment was significantly (p less than 0.01) higher in duodenal ulcer patients (0.76 +/- 0.55 ng/ml) than in healthy subjects of similar age (0.36 +/- 0.26 ng/ml). The increased PP level was not lowered significantly by medical treatment even when this resulted in healing of the ulcer.

The bacteriology of the gastric remnant contents was examined in 150 patients 15-20 years after Billroth II resection for duodenal ulcer disease. Samples of gastric contents were aspirated through a gastroscope, and aerobic and anaerobic microflora were cultured and analyzed semiquantitatively. All patients but one had bacterial growth, with a mean number of 6.8 strains per remnant. One third of the bacterial strains were anaerobic, and many of them had their normal habitat in the colon. The role played by bacteria of the gastric remnant in the etiology of mucosal pathology and gastric remnant disease is discussed.