Basal Serum Gastrin Levels Research Articles

Effect of age and Gender on Fasting and Food-Stimulated Levels of Calcitonin, Parathyroid, and Gastrin Hormones in Healthy Adults

Objective: To investigate the effect of age and gender on basal and food-stimulated serum calcitonin (CT), parathyroid hormone (PTH), and gastrin levels among healthy adults. Methods: Ninety-six healthy adults (76 men and 20 women) aged between 21 and 43 years were recruited. Serum CT, PTH, and gastrin levels were measured after a 9-hour overnight fast, and 1 and 3 hours postprandially. Results: PTH levels decreased early and increased late after feeding. This change was significant in men but not in women. CT levels increased in response to food intake in men but not in women. Gastrin levels were significantly increased after feeding in both men and women. Mean basal and food stimulated CT, PTH, and gastrin levels did not significantly differ between genders. Fasting and post-prandial PTH levels were higher while gastrin levels were lower in older subjects (>30 years old) compared to younger subjects (≤30 years old). Fasting and postprandial CT levels were not significantly different between age groups. Conclusion: Age had a significant effect on fasting and food-stimulated PTH and gastrin hormone levels. The effect of age on PTH levels was independent of baseline vitamin D levels. Men showed significant changes in CT and PTH levels in response to feeding compared to women, although the mean hormone levels were not significantly different between men and women. Abbreviations: CT = calcitonin; MTC = medullary thyroid carcinoma; PTH = parathyroid hormone; SD = standard deviation.

Increased serum gastrin in patients with different clinical forms of Chagas disease coinfected with Helicobacter pylori.

ABSTRACT Trypanosoma cruzi and Helicobacter pylori (HP) are pathogens that cause chronic diseases and have been associated with hypergastrinemia. The aim of this study was to evaluate the fasting gastrin levels in patients with different clinical forms of Chagas disease (CD), coinfected or not by HP. The enrolled individuals were outpatients attending at the university hospital. HP infection was assessed by serology and 13 C-urea breath test. Fasting serum gastrin concentration was measured by chemiluminescence assay. Gastric endoscopic and histological features were also evaluated. Associations between CD and serum gastrin level were evaluated in a logistical model, adjusting for age, gender and HP status. A total of 113 patients were evaluated (45 with Chagas disease and 68 controls). In the multivariate analysis, increasing serum gastrin levels (OR= 1.02; 95% CI= 1.01-1.12), increasing age (OR= 1.05; 95% CI= 1.02 - 1.09) and HP-positive status (OR = 2.88; 95% CI = 1.10 - 7.51) remained independently associated with CD. The serum gastrin levels were significantly higher in the group of patients with the cardiodigestive form ( P = 0.03) as well as with digestive form ( P = <0.001) of Chagas disease than in the controls. In conclusion, patients with cardiodigestive and digestive clinical forms of CD have increased basal serum gastrin levels in comparison with controls. Moreover, we also demonstrated that H. pylori coinfection contributes to the hypergastrinemia shown in CD.

Helicobacter pylori but not gastrin is associated with the development of colonic neoplasms

Recent studies have suggested that Helicobacter pylori (H. pylori) constitutes a risk for the development of colonic neoplasia. Hypergastrinemia can be induced by H. pylori infection, and gastrin can act as putative promoter of colorectal carcinogenesis. Aim of our study was to assess whether H. pylori infection and/or increased serum gastrin levels are associated with the occurrence of colonic neoplasms. For this, we reviewed prospectively collected data of 377 patients with a minimum age of 50 years who underwent colonoscopy. H. pylori and CagA status were determined by serology. Serum gastrin levels were measured in fasting state by commercially available assay. In H. pylori infected patients (n = 138; 36.6%), the overall prevalence of colonic neoplasms was more frequent compared to H. pylori negative patients (n = 239; 63.4%) (OR = 2.73, 95% CI: 1.76-4.24). H. pylori infection occurred more frequently in patients with hyperplastic polyps (OR = 2.66, 95% CI: 1.23-5.74) and adenomas presenting with low grade intraepithelial neoplasia (IEN) (OR = 1.85, 95% CI: 1.14-2.99). Attributable risk for adenomas with high grade IEN and colorectal adenocarcinoma (n = 14) was not assessed due to the low number of cases. The expression of CagA was also associated with an increased risk for colonic neoplasms (OR = 2.25, 95% CI: 1.29-3.94). Hypergastrinemia did not increase the risk for any colonic neoplasms and there was no difference in basal serum gastrin levels between H. pylori positive and negative patients. In conclusion, H. pylori infection, including CagA expression is associated with an increased risk for the development of colonic neoplasm.

Increased gastric acid secretion after massive small bowel resection is related to a decrease in enterogastrones.

The reported increase of gastric secretion after small bowel (SB) resection is controversial. To determine the effect of SB resection on gastric acid secretion we studied basal and dose step pentagastrin-stimulated gastric acid secretion as well as basal serum gastrin, secretin, neurotensin and postprandial gastrin levels in 12 dogs, before and after resection of 60% of the intestine representing both proximal (n = 6) and distal (n = 6) SB. Rat bioassay was also performed to rule out the presence of unknown gastric secretagogues in the blood. Proximal SB resection produced a significant increase in basal and low dose (100 ng/kg/h) pentagastrin-stimulated gastric acid secretion (ED50 = 1,110 vs. 720 ng/kg/h after resection). However, no significant changes in gastric secretion were observed after distal SB resections. Neither proximal nor distal SB resection altered basal or postprandial serum gastrin levels. Proximal SB resection reduced serum secretin levels (229 +/- 38 vs. 134 +/- 16 pg/ml, p < 0.05) but did not alter neurotensin levels. Rat bioassay failed to reveal a circulating secretagogue after SB resections. We conclude that proximal but not distal SB resection increases basal and submaximally stimulated gastric acid secretion. Such an effect may be due to the observed decrease in circulating secretin levels.

A preliminary report on Helicobacter pylori and antral gastrin concentration in patients with duodenal ulcer

This study analyzed the relationship between Helicobacter pylori infection, mucosal gastrin concentration, and basal serum gastrin levels in duodenal ulcer patients with dyspepsia. Patients with duodenal ulcers who were H pylori positive showed higher antral gastrin concentrations than either H pylori-positive patients with dyspepsia or H pylori-negative patients with duodenal ulcers. No correlation was found between serum gastrin and antral gastrin concentrations. The increased antral gastrin concentrations were not correlated with an inhibition of acid-gastrin feedback. These preliminary data confirm that H pylori infection impairs gastrin physiology.

Clinical, anatomical, and evolutive features of patients with the Zollinger-Ellison syndrome combined with type I multiple endocrine neoplasia.

The clinical evolution of type I multiple endocrine neoplasia (MEN I) was studied in 45 patients among a consecutive series of 172 with Zollinger-Ellison syndrome (ZES). These 172 patients were seen in our hospital between 1959 and 1989. Diarrhea was half as frequent in ZES-MEN I as in sporadic ZES cases. At diagnosis, mean basal acid output and serum gastrin levels in MEN I patients (28.8 +/- 6.6 mmol/h and 587 +/- 487 pg/ml, respectively) were not different from those observed in the others with sporadic ZES. Laparotomy was performed in all 36 patients with no diffuse liver involvement to attempt the removal of gastrinomas. Twenty-nine patients had adenomas, located in the pancreas in 21, in the duodenal wall in 3, and in both in 5. Adenomas were multiple in 23 cases (78%). No tumor was found in seven patients. Twenty-nine of the 36 operated patients were tumor-free after surgery; 7 died in the postoperative period between 1959 and 1970. Median follow-up of the 38 other patients was 95 months (range 17-278 months). Among the 24 patients without residual tumor at discharge (group I), biological and/or morphological evidence of a persistent or recurrent source of gastrin was obtained in 22. Among the 14 patients with residual tumor (group II), an increase in tumor size was seen in 5 after a median of 27 months (range 9-36 months), while no change occurred in 9 after 54 months (3-100 months). Actuarial survival curves were not different, either in group I versus group II patients (67 and 72% at 5-year follow-up, respectively) or in MEN I versus sporadic ZES patients. Apparently, complete resection of primary tumor did not reduce the incidence of subsequent liver metastases. In all, 21 of the 45 patients had malignant gastrinomas (47%), consisting of liver metastases in 14 (31%), metastatic lymph nodes in 11 (24%), and lung metastases in 2 (4%). Monitoring of fundic argyrophil cells disclosed hyperplasia in 13 of the 14 MEN I patients (92%), and 5 had invasive carcinoid tumors. Taken together, these results prompt us to recommend that in ZES-MEN I patients, surgery should be avoided and oxyntic mucosa regularly monitored.

Antral gastrin cell hyperfunction in children

Objective: A description of the clinical history, diagnosis and management of a child with antral gastrin cell hyperfunction. Design: Detailed studies of hormonal, endoscopic and morphologic patterns. Setting: Gastrointestinal endocrinology section of a university hospital. Patient: A 14-year-old boy who complained of two episodes of acute upper gastrointestinal bleeding due to duodenal ulcer. Interventions and outcome measures: H 2 -antagonist therapy. Conclusions: Antral gastrin cell hyperfunction can be an underlying cause of peptic ulcer in children. The diagnosis is made by basal and protein meal-stimulated serum gastrin levels. Long-term H 2 -antagonist treatment is suggested

Effect of loxiglumide on basal and gastrinand bombesin-stimulated gastric acid and serum gastrin levels

The effect of the specific cholecystokinin-receptor antagonist loxiglumide on basal and bombesin-, and gastrin 17-I-stimulated gastric acid secretion and serum gastrin levels was studied in 12 healthy subjects. Loxiglumide (10 mg · kg−1 · h−1) significantly augmented basal gastric acid output from 1.8 ± 0.3 to 3.9 ± 0.6 mmol H+/h (P < 0.005) but did not significantly influence integrated basal serum gastrin concentrations (2 ± 21 vs. 32 ± 21 pmol L−1 · h−1). Both gastric acid secretion and integrated serum gastrin concentrations stimulated by bombesin infusion (92.6 pmol · kg−1 · h−1) were significantly augmented by loxiglumide [from 4.0 ± 0.3 to 10.0 ± 1.3 mmol H+/h (P < 0.005) and from 1251 ± 93 to 2558 ± 206 pmol · L−1 · h−1 (P < 0.005), respectively]. Gastric acid output and serum gastrin concentrations during infusion of 5 pmol · kg−1 · h−1 of synthetic human gastrin 17-1 (9.6 ± 2.9 mmol H+/h and 1045 ± 177 pmol · L−1 · h−1) and during infusion of 15 pmol · kg−1 · h−1 of gastrin 17-I (14.5 ± 3.1 mmol H+/h and 2412 ± 312 pmol · L−1 · h−1) were not significantly influenced by loxiglumide (10.3 ± 2.3 mmol H+/h and 1291 ± 257 pmol · L−1 · h−1 for the 5-pmol · kg−1 · h−1 gastrin 17-I infusion dose with loxiglumide and 13.6 ± 3.4 mmol H+/h and 2611 ± 305 pmol · L−1 · h−1 for the 15-pmol · kg−1 · h−1 gastrin 17-I infusion dose with loxiglumide). These data indicate that endogenous cholecystokinin inhibits gastric acid secretion under basal conditions and gastrin release and gastric acid secretion during infusion of bombesin in humans and suggest that the augmented effect of loxiglumide on bombesin-stimulated gastric acid secretion may be explained largely by enhanced gastrin release.

Gastric acid secretion and gastrin release in the baboon.

Significant species differences have been demonstrated in gastric physiology, a factor that limits extrapolation of animal data to man. Primate physiology is thought to be similar to that of man; however, gastric function has not been adequately documented in the primate. In the present study six baboons (body weight 25.5 +/- 1.8 kg) were trained to sit in a chair and gastric acid secretion and gastrin release was studied in conscious animals. Mean basal acid secretion was 1.3 +/- 0.1 mmol (H+)/hr. Maximum output after pentagastrin (12 micrograms/kg/hr) was 9.5 +/- 0.9 mmol (H+)/hr and 11.0 +/- 0.4 mmol (H+)/hr after histamine (40 micrograms/kg/hr). A statistically significant (by cosinor analysis) circadian rhythm was demonstrated for intragastric pH over 24 hr in fasted baboons (P less than 0.001). Mean basal serum gastrin level was 37.7 +/- 8.3 pg/ml. The integrated gastrin response after administration of a protein rich meal was 2.52 +/- 0.07 ng x min/ml and this increased to 5.17 +/- 0.18 ng x min/ml (P less than 0.05) following simultaneous administration of a meal with atropine (0.2 mg/kg) (P less than 0.05). Our results suggest that there is significant basal and stimulated acid secretion in the baboon; the amount of acid secreted is similar to that reported in man. Gastric pH demonstrated a circadian rhythm. Postprandial gastrin release was significantly enhanced by cotreatment with atropine. As the present findings are similar to those previously reported in man, the baboon may be a useful model for further studies in gastric physiology and experimental peptic ulceration.

Role of serum fasting gastrin in screening for hypergastrinemic syndromes in duodenal ulcer disease

Basal serum gastrin levels were measured in 237 patients with endoscopically confirmed duodenal ulcer and were higher than normal in 16 cases. Protein meal gastrin stimulation was performed on this group of 16 patients and on a control group of 48 patients with normal basal gastrin concentrations but high rates of either ulcer recurrence or of complications (e.g., bleeding or perforation); 21 patients from the two groups were also tested for serum gastrin inhibition with secretin. Four cases (25%) of antral G-cell hyperfunction were found in the first group, plus 1 case compatible with Zollinger-Ellison syndrome (6.2%). Only 1 case (2%) of antral G-cell hyperfunction was found among the 48 controls. These results suggest the clinical utility of routine basal gastrin measurement in screening for hypergastrinemic patients with duodenal ulcer disease.

Pyloric Antrum Gastrin Cell Hyperplasia after Jejunum or Colon Transposition: Effects of Jejunal or Colonic Mucosa on G Cells

Jejunal or colonic segments are currently used as esophageal substitutes after resection of intractable peptic stenoses. The present study was carried out in order to investigate the effects of the jejunal or colonic mucosa on antral gastrin (G) cells. Colonic or jejunal patches with intact vascular supply were sutured to the pyloric antrum or the higher portion of the gastric body in 40 rats. Ten further animals were used as controls. Three to 4 months after surgery, the serum gastrin levels were weekly determined in fasted (24 h) and freely fed rats using radioimmunoassay. The pyloric antrum was then removed, and the G cell density was assessed with an immunoperoxidase method. Transposal of the colonic mucosa to the antrum increased G cell density and basal serum gastrin levels, while grafting of the jejunal mucosa did not. G cell proliferation proved to be dependent on the topographic location of the colonic patch with respect to the pyloric antrum. Serum gastrin values in freely fed rats tended to be greater in the colon-to-antrum transposition group than in the other experimental groups, but the difference was not significant. In conclusion, variations between the properties of the jejunal and colonic mucosa would include their uneven effects on antral G cells.

The Role of the Antrum and the Vagus Nerve in the Metabolism of Histamine in the Human Gastric Mucosa

The effects of antrectomy and proximal gastric vagotomy on the metabolism of histamine in the human gastric mucosa were studied in the basal state and during pentagastrin stimulation in patients with duodenal or gastric ulcer disease. Mucosal biopsy specimens were taken from the antral and oxyntic gland areas, whereafter histamine content, histidine decarboxylase activity, and histamine methyltransferase activity were simultaneously assayed. Vagotomy was followed by a decrease in the acid secretory capacity and an increase in basal serum gastrin levels. Histamine content of the oxyntic mucosa increased after vagotomy, but the ability of pentagastrin to form new amounts of the amine was impaired. Antrectomy caused a decrease in acid secretion and a fall in gastrin concentrations. Basal histamine content and rate of amine formation in the remaining oxyntic mucosa were unaffected by antrectomy. Antrectomy impaired the ability of pentagastrin to release histamine. Histamine methyltransferase was not affected by pentagastrin, vagotomy, or antrectomy. In conclusion, both antral gastrin and the vagus nerve seem to exert a regulatory influence on the metabolism of histamine in the human oxyntic mucosa. The withdrawal of these factors either causes impaired ability of pentagastrin to release histamine from its storage site or counteracts the ability of pentagastrin to accelerate histamine synthesis.

Effect of Massive Small Bowel Resection on Gastric Acid Secretion in the Rat

The phenomenon of a transitory gastric acid hypersecretory state after extensive bowel resection is well established. Its time of onset, however, is unknown. The purpose of this study was to determine the immediate effect of massive small bowel resection (MSBR) on gastric acid secretion (GAS). An anesthetized innervated rat model was prepared with gastric and jugular cannulae. Three groups of animals were studied: group I (n = 12), basal unstimulated state; group II (n = 12), pentagastrin (Pg) 16 micrograms/kg h-1 stimulated; and group III (n = 16), 5% liver extract meal (LEM) stimulated. Each group consisted of experimental animals that underwent 95% MSBR from proximal jejunum to terminal ileum and control animals that remained intact. Acid output was determined by extragastric titration with 0.1 M NaOH. Blood was taken for basal and postprandial serum gastrin levels. Basal acid output (mueq/10 min) significantly increased immediately after MSBR in all groups (p less than .01). Ninety minutes following MSBR, acid outputs were significantly elevated in basal and Pg-stimulated but not LEM-stimulated rats. Serum gastrin increased from 30 +/- 1 to 56 +/- 6 pg/mL (p less than .01) in group I rats and from 81 +/- 28 to 129 +/- 13 pg/mL in group III rats (p = NS). We conclude that GAS increases immediately after MSBR in group I and II rats. This increase in GAS may be mediated by gastrin release.

Effect of long-term treatment with omeprazole on serum gastrin and serum group A and G pepsinogens in patients with reflux esophagitis

Twenty-nine nongastrectomized and three partially gastrectomized patients with chronic reflux esophagitis resistant to 12 weeks' treatment with histamine H2-receptor antagonists were treated with a daily oral dose of 20–40 mg of omeprazole for 12–30 months. Basal serum gastrin, serum pepsinogen A, and serum pepsinogen C concentrations were monitored at regular intervals. Serum gastrin levels significantly (P < 0.01) increased threefold to fourfold during the first 1–2 months of the study when all patients ingested 40 mg of omeprazole daily. Dose reduction to 20 mg did not significantly decrease gastrin levels. Serum gastrin levels showed a trend to further increase after the first 3 months of treatment, reaching statistically significant differences for values from the 3–12-month period (P < 0.05) and from the 3–24-month period (P < 0.005). Women and patients with high basal serum gastrin levels before omeprazole treatment were more likely to achieve higher serum gastrin levels during omeprazole treatment. Serum pepsinogen A and C levels were significantly (P < 0.01) increased at all time intervals during long-term treatment with omeprazole. No significant tendency toward higher serum pepsinogen C levels in time was observed. However, serum pepsinogen A levels and the ratio of pepsinogen A to pepsinogen C further increased significantly (P ≤ 0.05) during the initial 3–12-month period. However, this trend was not observed anymore afterward. Antrectomized patients did not show increases in serum gastrin and serum pepsinogen A and C levels, suggesting that hypergastrinemia may be involved in the observed hyperpepsinogenemia.

Protection by oral phenylalanine against gastric carcinogenesis induced by N-methyl-N'-nitro-N-nitrosoguanidine in Wistar rats

The effect of oral administration of L-phenylalanine on the incidence and histology of gastric adenocarcinomas induced by N-methyl-N'-nitro-N-nitrosoguanidine was investigated in inbred Wistar rats. Oral administration of 6% phenylalanine after 25 weeks of treatment with the carcinogen significantly reduced the incidence and number of adenocarcinomas of the glandular stomach at experimental week 52. Oral administration of high dose phenylalanine significantly increased the basal serum gastrin level and significantly decreased the norepinephrine concentration in the antral portion of the gastric wall, as well as the labelling indices of antral mucosa. These findings indicate that orally administered phenylalanine inhibits the development of gastric cancers.

Efficacy and Safety of Omeprazole in the Long-Term Treatment of Peptic Ulcer and Reflux Oesophagitis Resistant to Ranitidine

A total of 143 patients with peptic ulceration of the duodenum, stomach and oesophagus who did not respond to 3 or more months of high-dose treatment with ranitidine, 450 mg or more daily, were treated with oral omeprazole, 40 mg daily. In 94% of the patients, ulcers healed within 2-8 weeks. After healing, 133 patients underwent long-term maintenance treatment with omeprazole, 40 mg daily, for 1-5 years (continuing). During maintenance therapy with omeprazole, no endoscopically verified relapses occurred, and no drug-related adverse effects were seen. There were no significant changes in routine laboratory tests in any patient, including 27 with concomitant liver cirrhosis. Basal serum gastrin levels, which were already elevated by the previous high-dose ranitidine treatment, rose to 4 times normal levels after 4 months of treatment with omeprazole. Thereafter, no further increases in basal serum gastrin levels were observed, even after 5 years of administration. The volume density of argyrophilic cells in the oxyntic mucosa increased during omeprazole treatment, but no dysplasia of the gastric enterochromaffin-like cells was seen. In conclusion, omeprazole was highly effective in healing ranitidine-resistant peptic ulcers, and subsequent maintenance therapy with omeprazole, 40 mg daily, was found to be effective and safe over the period observed.

Prevention of Relapse of Reflux Oesophagitis after Endoscopic Healing: The Efficacy and Safety of Omeprazole Compared with Ranitidine

The prognosis for the acute healing of patients with persistent, erosive or ulcerative reflux oesophagitis depends upon the choice of drug treatment as well as the severity of the initial oesophagitis. The results of this study show that omeprazole, 40 mg once daily (n = 51), is more effective than ranitidine, 300 mg b.i.d. (n = 47), in the acute healing of reflux oesophagitis (90 and 47% healing, respectively; p less than 0.0001). Omeprazole, 20 mg o.m., is also far more effective than ranitidine, 150 mg b.i.d., in the maintenance treatment of these patients; 67% of the omeprazole-treated patients remained in clinical and endoscopic remission throughout the 12-month study period compared with only 10% of those given ranitidine (p less than 0.0001). This therapeutic advantage was achieved without adverse events and without significant abnormalities in the endocrine or exocrine cell population of the oxyntic mucosa. Basal serum gastrin levels were modestly raised after 4 weeks of omeprazole treatment (the change being from 8.6 to 16.9 pmol/l; p less than 0.05), and remained at this level throughout the study period.

Influence of Chronic Rioprostil Treatment on Gastric Endocrine Function

The influence of 4 weeks of treatment with the prostaglandin E1 analogue, rioprostil, 300 micrograms b.i.d., or placebo, on gastric endocrine function is tested in healthy male volunteers. Basal serum gastrin levels and postprandial gastrin output are unchanged after treatment with rioprostil. Similarly, plasma pancreatic polypeptide levels are unaffected. Antral gastrin tissue concentrations as well as antral somatostatin concentration and volume densities of antral G-cells and D-cells are unchanged. Neither basal nor pentagastrin-stimulated gastric acid secretion after rioprostil therapy differ from pretreatment values. Rioprostil, given in a dose of 100 micrograms b.i.d. to rats for 1 week significantly increases antral mucosal height but has no influence on the mucosal concentrations and cell densities of the gastric peptides, gastrin and somatostatin. It is concluded that rioprostil in the dose used does not affect the endocrine stomach after 4-weeks' administration at a dose of 300 micrograms b.i.d.

Antisecretory and antigastrin effects of rioprostil in gastric fistula dogs.

This investigation examined the effect of rioprostil, a primary alcohol prostaglandin E1 analog, on betazole-stimulated gastric acid secretion and on basal and food-stimulated (postprandial) serum gastrin levels in gastric fistula dogs. Rioprostil inhibited betazole-stimulated gastric acid secretion with an ED50 of 16 (10-24) micrograms/kg, intragastrically. A near-maximal gastric antisecretory dose (100 micrograms/kg, intragastrically) had no effect on basal serum gastrin levels but significantly attenuated the rapid rise in serum gastrin which follows feeding, a result different from that reported for other prostaglandin E1 analogs. A nonantisecretory dose of rioprostil (1.0 micrograms/kg, intragastrically) also attenuated the rise in postprandial serum gastrin. An antigastrin effect using a nonantisecretory dose of an antiulcer agent has not been reported previously and may indicate that rioprostil has a direct inhibitory effect on secretion of gastrin. The ability of rioprostil to inhibit gastric acid secretion and decrease postprandial peak serum gastrin levels, coupled with previously established cytoprotective efficacy, makes it an attractive clinical candidate for the treatment and prevention of peptic ulcer disease.

Gastrin and partial gastrectomy in duodenal ulcer patients 25 years postoperatively

Serum gastrin concentrations were measured in 26 patients (Group A) who underwent partial gastrectomy for duodenal ulcer disease at least 25 years previously. The basal serum gastrin level was 31.6 ± 1.5 pg/ml, increasing significantly after food stimulation to 36.8 ± 2.4 pg/ml (p < 0.05). In 16 of these patients (Group B), endoscopic biopsy specimens were taken from the minor and major curvatures of the stomach, from the jejunal part of the gastrojejunostomy, and from the duodenum. Mucosal gastrin concentrations measured were 0, 20.4 ± 3 ng/g, and 30 ± 7.1 ng/g, respectively. Ten of these patients (Group C) had a repeated gastroscopy with food stimulation of the duodenal pouch, using a balloon catheter in the lower duodenum. Serum gastrin concentrations after pouch stimulation did not differ significantly compared with basal values. No changes in the mucosal gastrin concentrations during pouch stimulation were seen. In conclusion, since no gastrin could be detected in the gastric mucosa and the mucosal gastrin concentration in the duodenum was 50 times lower than reported in other studies, the gastrin release measured after food stimulation could have been of extragastric or extraduodenal origin.