A database on dyspepsia.

A prospective study of dyspepsia was carried out in a primary referral hospital between 1974-1987 including 1540 patients of whom 1433 were seen as outpatients. The study protocol was agreed in advance and a structured questionnaire was used to elicit relevant clinical information: up to three diagnoses were permitted for each patient. The commonest principal diagnoses were duodenal ulcer (26%), functional dyspepsia (22%), and irritable bowel syndrome (IBS) (15%); alcohol related dyspepsia (4%) was as common as gastric carcinoma or symptomatic gall stones. Multiple diagnoses were common (31% given two diagnoses, and 6% given three) so that in all 2111 diagnoses were given to 1540 patients; the functional disorders (IBS and functional dyspepsia) considered together accounted for 39% of all diagnoses made. Whereas organic conditions were diagnosed by clinicians with confidence (63-98% considered 'certain'), even when given as the principal or first diagnosis IBS was considered 'certain' in only 61% and functional dyspepsia 48%. The demographic symptom data, together with information on tobacco and alcohol use, and work lost are described in detail.

Enhanced gastric mucosal haemostasis after upper gastrointestinal haemorrhage.

An endoscopic technique for the measurement of gastric mucosal bleeding time has been developed to study gastric haemostasis in patients with acute upper gastrointestinal haemorrhage. The relation of gastric mucosal bleeding time to skin bleeding time and nonsterodial anti-inflammatory drug usage was examined in 61 control patients and in 47 patients presenting with bleeding peptic ulcers or erosions. Gastric mucosal bleeding time was shorter in patients with haemorrhage (median 2 minutes, range 0-5 minutes) than in the control group (median 4 minutes, range 2-8 minutes) (p less than 0.001). Skin bleeding times were similar in the two groups (medians 4 minutes in patients with haemorrhage and 4.5 minutes in controls). In 21 patients with haemorrhage who were taking non-steroidal anti-inflammatory drugs, the median gastric mucosal bleeding time (2.5 minutes, range 1.0-5.0 minutes) was similar to that in 26 patients with haemorrhage not associated with these drugs (2.0 minutes, range 0.0-5.0 minutes). These results show that gastric mucosal haemostasis is accelerated in response to haemorrhage in the upper gastrointestinal tract, even in patients taking nonsteroidal anti-inflammatory drugs. This enhanced gastric haemostasis probably reflects a local protective response to minimise blood loss from the bleeding lesion.

A comparison of symptoms between non-ulcer dyspepsia patients positive and negative for Helicobacter pylori.

The role of Helicobacter pylori infection in the symptom complex associated with non-ulcer dyspepsia is uncertain, despite the presence of the organism in a high proportion of these patients. In order to exclude physician bias in history taking, 18 patients (9 female) diagnosed as non-ulcer dyspepsia, after endoscopy and gallbladder ultrasonography, underwent computer interrogation using the Glasgow Diagnostic System for Dyspepsia (GLADYS). Five antral and 3 fundal endoscopic biopsies from these patients were also histologically examined for the presence of Helicobacter pylori and quantitatively analysed for polymorph and chronic inflammatory cell densities per mm2 of lamina propria using computer-linked image analysis. In the group of 9/18 patients who were positive for Helicobacter pylori, there were significantly higher antral and fundal inflammatory cell counts than in negative patients. However, analysis of the GLADYS interrogation data showed no significant positive relationships between Helicobacter pylori positivity and any gastrointestinal symptoms. These results confirm a significant association between Helicobacter pylori and superficial gastritis but suggest that non-ulcer dyspepsia in patients with Helicobacter pylori colonisation is probably not a clinically identifiable and distinct syndrome.

Effect of simulated intragastric haemorrhage on gastric acid secretion, gastric motility, and serum gastrin.

The majority of upper gastrointestinal bleeds stop spontaneously despite the low pH and proteolytic activity of gastric juice which inhibit coagulation and platelet aggregation. In order to investigate this paradox six healthy male volunteers received intragastric infusions of 160 ml autologous venous blood or 160 ml egg white acting as control in random order on separate days. Basal acid output was calculated before infusion, net acid secretion and gastric volume emptied were calculated after intragastric infusions. Serum gastrin concentrations were also measured before and after intragastric infusions and expressed as the integrated gastrin response. Basal acid output (mmol/h) was 4.7 (1.9) (mean (SEM)) before egg white infusion and 5.9 (2.6) before venous blood infusion. After egg white infusion net acid secretion (mmol/20 min) increased to 5.6 (3.1) compared with 2.3 (1.3) after venous blood infusion (p less than 0.05). The gastric volume emptied (ml/20 min) was less after venous blood infusion at 105 (28) compared with 321 (66) after egg white infusion (p less than 0.03). Integrated gastrin response was similar after venous blood and egg white infusion. When compared with an equivalent protein meal intragastric blood stimulates less acid secretion and delays gastric emptying. This effect may facilitate haemostasis after gastric bleeding.

Comparison of gastric body and antral pH: a 24 hour ambulatory study in healthy volunteers.

Simultaneous ambulatory records of gastric antral and body pH were made over 24 hours in nine healthy volunteers by means of endoscopically positioned and anchored glass electrodes. Intragastric pH was temporarily raised after the endoscopy with the median pH value 30 minutes after the procedure being 3.9 (range 1.5-7.0) for the antrum and 4.1 (range 1.5-7.0) for the body. Daytime pH (median pH value between 12 00 h and 23 00 h) was lower in the antrum (median = 1.9, range 1.6-2.6) than in the body (median = 2.7, range 1.8-4.5) (p less than 0.05) and this was because of the rise in pH on eating being less marked in the antrum than in the body. The median peak pH recorded during the evening meal was only 4.1 (range 2.4-6.2) in the antrum compared with 6.3 (range 4.4-6.7) in the body (p less than 0.01). Preprandial pH (median value over the hour prior to the evening meal) was similar in the antrum (median = 1.9, range 1.2-2.5) and body (median = 1.9, range 1.3-2.8). Night-time pH (median pH value between 23 00 h and 05 00 h) in six subjects remained low and was similar in the antrum (median = 1.4, range 1.2-1.7) and body (median = 1.3, range 1.1-1.7). In two subjects, however, there were episodes of raised night-time pH which were more marked in the antrum than in the body. Antral biopsies showed gastritis in four of the nine normal volunteers, which in three was associated with the presence of campylobacter-like organisms. This study shows the significant regional variations in day and night-time intragastric pH.

Rebound nocturnal hypersecretion after four weeks treatment with an H2 receptor antagonist.

Daytime intragastric pH, fasting and meal stimulated serum gastrin and nocturnal acid output were studied in eight male duodenal ulcer patients before, during and two days after completing nizatidine 300 mg nocte (20:00 h) for four weeks. Median nocturnal acid output (mmol/10 h) decreased during treatment to 11.6 (range 0.4-26.7) compared with pretreatment value of 39.4 (9.8-91.2); median acid inhibition 77% (p less than 0.01) which was strongest between 24:00 and 04:00 h. Two days after discontinuing treatment, nocturnal acid output increased to 74.1 (11-181). Compared with the pretreatment value this represents median rebound hypersecretion of 77% (p less than 0.05), caused by increased H+ concentration and volume of secretion. Overall median daytime intragastric pH (09:00-21:00 h) was unchanged on the final day of treatment and two days after completing therapy, compared with the pretreatment values. Fasting serum gastrin measured between 09:30 and 10:00 h and the integrated gastrin response to an OXO breakfast taken out at 10:00 h were also similar during and after treatment, compared with pretreatment values. The rebound nocturnal hypersecretion may be relevant to the high ulcer relapse rates after stopping H2 receptor antagonists.

Inhibition of gastric secretion and motility by simulated upper gastrointestinal haemorrhage: a response to facilitate haemostasis?

As gastric acid and pepsin inhibit blood coagulation and platelet aggregation it is surprising that most upper GI haemorrhages stop spontaneously. To investigate this paradox we have studied acid and pepsin secretion, gastric motility and GI hormones after simulated upper GI haemorrhage. In seven healthy volunteers intraduodenal infusion of 160 ml autologous blood decreased pentagastrin stimulated submaximal acid secretion (mmol/h) from 30.0 (3.2) (mean (SE] in the hour preceding infusion to 21.4 (3.7) in the hour following infusion (p less than 0.02), representing a mean reduction in acid output of 30%. Pepsin output (mg/h) was also decreased from 207.5 (67.7) (mean (SE] in the hour preceding blood infusion to 135.7 (54.7) in the hour after infusion (p less than 0.02) representing a mean reduction in pepsin output of 43%. In six volunteers gastric emptying of a liquid meal was delayed after intraduodenal blood infusion compared with intubation alone with the emptying time (min) to half volume (t 1/2) being prolonged at 75.0 (8.2) (mean (SE] after blood infusion compared with 35.5 (6.6) after intubation alone (p less than 0.02). Plasma GIP concentrations (ng/l) increased to peak levels of 127.9 (62.7) (mean (SE] after intraduodenal blood infusion compared with the pre-infusion value of 58.3 (2.3) (p less than 0.02). These changes may represent protective physiological responses to facilitate haemostasis.

Effect of low-dose omeprazole on gastric acid secretion in duodenal ulcer patients.

The effect of 7 days of oral dosing with 5 mg day-1 and 20 mg day-1 omeprazole on basal and pentagastrin-stimulated gastric acid output was studied in nine duodenal ulcer patients. Basal acid output measured 5-6 h post-dosing was decreased by a mean of 75% on 5 mg omeprazole and by 90% on 20 mg omeprazole (P less than 0.05 and P less than 0.01, respectively). Peak acid output measured 6-7 h post-dosing was decreased by a mean of 75% on 5 mg omeprazole and 97% on 20 mg omeprazole (P less than 0.01 for each). There was a wide interindividual variation in response to the 5 mg dose, with five of the nine patients having more than 90% inhibition of peak acid output, but two patients having less than 40% inhibition. This unpredictable response to daily low-dose omeprazole therapy makes it unsuitable for maintenance treatment.

Hepatocellular and gastric carcinoma associated with familial polyposis coli.

A case of familial polyposis coli in association with hepatocellular and gastric carcinoma is reported. No similar case has ever been documented in the world literature. This may be surprising as it is well known that familial polyposis has a potent oncogenicity not only in the colon but also in extracolonic organs.

A comparison of the effects of treatment with either famotidine 40 mg or cimetidine 800 mg nocte on gastric acid secretion and serum gastrin.

The effects on gastric acid secretion and serum gastrin of 4 weeks treatment with famotidine 40 mg nocte or cimetidine 800 mg nocte (at 2200 h) were studied in 16 patients with previous duodenal ulcer. Patients were studied before commencing therapy, on days 5, 12 and 26 of treatment, and on days 3, 10 and 24 after completion of therapy. In contrast to cimetidine, basal and pentagastrin-stimulated gastric acid secretion measured 12 h after dosing was significantly inhibited during treatment with famotidine. In addition, with famotidine there was inhibition of stimulated gastric acid secretion at 3 and 24 days after completion of treatment. Fasting serum gastrin measured 12 h after dosing was not significantly altered by either drug.

Electrodes for 24 hours pH monitoring--a comparative study.

Three pH electrodes in clinical use were examined--(1) antimony electrode with remote reference electrode (Synectics 0011), (2) glass electrode with remote reference electrode (Microelectrodes Inc. MI 506) and (3) combined glass electrode with integral reference electrode (Radiometer GK2801C). In vitro studies showed that both glass electrodes were similar and superior to the antimony electrode with respect to response time, drift, and sensitivity. The effect of the siting of the reference electrode on the recorded pH was examined in five human volunteers. The pH reading using a remote skin reference electrode was higher by a mean of 0.3 pH units (range 0.0-0.6) in the stomach, lower by 0.65 pH units (0.5-0.8) in the duodenum and lower by 0.3 pH units (0.0-0.6) in the oesophagus than that simultaneously obtained with an intraluminal reference electrode. Buccal reference electrodes gave similar readings to skin. Combined reference and glass pH electrodes are recommended for 24-hour ambulatory pH monitoring.

A simple blood sample method for measuring oral zinc absorption in clinical practice.

Tracer activity in blood 14 days after the oral ingestion of zinc-65 has been compared with zinc absorption measured by whole-body counting in 10 controls and 25 subjects with Crohn's disease. Highly significant linear correlations (P less than 0.001) were obtained between blood tracer activity and zinc absorption for both the control group (previously reported) and the Crohn's disease group, thus allowing zinc absorption to be predicted from blood tracer activity with an accuracy of +/- 7.1% for the control group, +/- 17.1% for an active Crohn's disease sub-group and +/- 18.0% for a quiescent Crohn's disease sub-group. Relative to the whole-body counting method, the prediction of zinc absorption from blood tracer activity had a sensitivity of approximately 88%, a specificity of approximately 89% and an overall performance accuracy of approximately 88%. Therefore, where whole-body counting facilities are not available, the blood tracer technique for predicting zinc absorption could be a useful alternative.

An endoscopic procedure for accurate localisation of intraluminal pH electrodes.

An endoscopic procedure has been developed which permits the accurate fixation of pH electrodes to the mucosa of the upper gastrointestinal tract. The electrode is carried to the desired position tethered to the endoscope. A clip-fixing device is passed through the biopsy channel of the endoscope and the electrode is anchored to the mucosa with stainless steel clips. Repeat endoscopy in 20 patients has confirmed that the electrode remains firmly secured to the mucosa for at least 24 h. The electrode can be safely removed at any time by traction on its cable. This procedure should allow, for the first time, accurate 24 h ambulatory pH monitoring of specific regions of the upper gastrointestinal tract.

Effects of nifedipine on gastric acid secretion and gastrin release in man.

As calcium is important in the regulation of gastric acid secretion and gastrin release, we have examined the effect of the calcium antagonist nifedipine on these processes in man. Nifedipine 30 mg orally inhibited basal acid output by 37% (p less than 0.025) and that stimulated by low infusion rates of pentagastrin--that is, 0.031 and 0.062 microgram/kg/h by 44% (p = 0.05) and 39% (p less than 0.02) respectively. On increasing the pentagastrin infusion rate the inhibition was surmounted suggesting it was competitive in type. Nifedipine did not affect basal or Oxo meal stimulated gastrin concentrations in normal volunteers nor did it affect resting serum gastrin or calcium stimulated increase in gastrin in a single patient with Zollinger-Ellison syndrome. These findings are consistent with the transmembrane flux of calcium ions being involved in basal and pentagastrin stimulated acid secretion in man.

A case of eye disease due to dietary vitamin A deficiency in Glasgow.

A 22 year old Caucasian girl living in Glasgow presented with eye disease due to Vitamin A deficiency. There was no evidence of liver disease or malabsorption and the vitamin deficiency was found to be due to her bizarre dietary habit.

Safety evaluation of cimetidine: report at the termination of a seven-year study in dogs.

Cimetidine in tablet form was administered orally daily to eight male and four female beagle dogs at a dose level of 144 mg kg-1 for 385 weeks. Four male and two female control dogs received placebo tablets. Treatment with cimetidine did not affect the clinical condition of the dogs, but was associated with a slightly less rapid weight gain. Five dogs (two cimetidine-treated and three controls) were killed during the course of the study for reasons not related to treatment. Treatment with cimetidine did not affect haematological, clinical chemical, urinalysis or electrocardiographic parameters. Multiple biopsies of gastric mucosa taken at intervals of approximately six months from Week 177 to Week 363 showed no change which could be attributed to treatment with cimetidine. There were no toxic effects on the gastric mucosa of any dog. At necropsy no treatment-related findings were reported except for reduction in prostate size in 6/8 males receiving cimetidine. This was expected in view of previous experience with cimetidine.

Taking a calculated risk: predictive scoring systems in dyspepsia.

The derivation of a predictive index is illustrated using 14 simple questions to assess the chance of the presence of an ulcer. The evaluation and use of such a 'calculated-risk' is briefly described.

The antisecretory effects of zaltidine, a novel long-acting H2-receptor antagonist, in healthy volunteers and in subjects with a past history of duodenal ulcer.

The potency and duration of the antisecretory action of zaltidine, a novel H2-receptor antagonist, have been examined in healthy volunteers and in patients with previous duodenal ulceration. In eight healthy male volunteers single oral doses of 5 mg, 25 mg and 100 mg produced dose-related inhibition of basal and pentagastrin-stimulated acid output (M.A.O.) with an estimated ID50 of 40 mg for the latter. In eight subjects with duodenal ulceration single 100 mg and 200 mg doses produced 85% and 97% inhibition of M.A.O. at peak (3 h post-dose) and 20% and 23% inhibition at 24 h, respectively; inhibition of basal acid output was 97% at 3 h and 50% at 24 h with both doses. The long duration of action of zaltidine is ascribed to its relatively slow elimination from the plasma.

Effect of a single oral dose of enprostil on gastric secretion and gastrin release. Studies in healthy volunteers and patients with pernicious anemia.

In healthy human volunteers, a single oral dose of enprostil (35 micrograms) inhibited basal gastric acid output by a mean of 71 percent, pentagastrin-stimulated output by 46 percent, sham-meal-stimulated output by 48 percent, and histamine-stimulated output by 16 percent. In each case, there was a reduction in both the volume and acidity of the gastric juice. Pepsin output was unchanged. Although enprostil increased the gastric pH, it did not induce basal or post-prandial hypergastrinemia. In patients with hypergastrinemia secondary to achlorhydria, enprostil lowered the basal gastrin level and reduced or abolished the post-prandial gastrin rise in a dose-related fashion. Enprostil reduces basal and stimulated gastric acid secretion and inhibits gastrin release.