Gut: first published as 10.1136/gut.25.11.1217 on 1 November 1984. Downloaded from Gut, 1984, 25, 1217-1220 Does gastric acid release plasma somatostatin in man? M R LUCEY, J A H WASS, P D FAIRCLOUGH, M O'HARE, P KWASOWSKI, E PENMAN, J WEBB, AND L H REES From the Departments ofGastroenterology, Endocrinology, Chemical Endocrinology, St Bartholomew's Hospital, London, Department ofBiochemistry, University ofSurrey, Guildford, Surrey, and the Department ofMedicine, Queen's University, Belfast SUMMARY Food and insulin hypoglycaemia raise plasma concentrations of somatostatin. Both also stimulate gastric acid secretion but it is not clear whether gastric acid itself has any effect on somatostatin secretion. We, therefore, studied the effect on plasma concentrations of somatostatin of infusion of 0.1 N HC1 into the stomach and duodenum of healthy subjects. Plasma somatostatin did not rise with a small dose of HC1 given intragastrically (15 mmol) or intraduodenally (4 mmol). After an intraduodenal infusion of 60 mmol HC1 over 30 minutes, sufficient to reduce intraluminal pH to 2, plasma somatostatin rose moderately in five subjects from a mean value (±SEM) of 32±3 pg/ml to a peak at 10 minutes of 54±11 pg/ml. It is concluded that: (a) intragastric acid infusions do not release circulating somatostatin in man; and (b) that intraduodenal acidification albeit at grossly supraphysiological doses is a moderate stimulus of plasma somatostatin release. Therefore, gastric acid is unlikely to be a major factor mediating postprandial plasma somatostatin release in man. Somatostatin is a tetradecapeptide widely distri- 21-24 years) were within 10% of their ideal body buted in brain, gut, and pancreas of many species weight and taking no medication. None had a http://gut.bmj.com/ with the greatest abundance in stomach, duodenum history of endocrine, gastrointestinal or renal and pancreas.1 2 When administered exogenously in disease. After an overnight fast the subject was pharmacological doses it has many inhibitory intubated by mouth with a fine bore flexible tube actions, both endrocrine and non-endocrine.3 We and its tip was positioned in the stomach or second have shown that circulating concentrations of part of duodenum under radiographic control. Two somatostatin rise in man after oral ingestion of a separate studies were conducted. meal4 or individual nutrients, especially fat or on September 25, 2021 by guest. Protected copyright. protein,5 and after insulin-induced hypoglycaemia.6 INTRAGASTRIC INFUSION Both oral food and hypoglycaemia are stimuli of Six subjects each received on two separate gastric acid secretion.7 It is possible that gastric acid occasions, an intragastric infusion of (a) 15 mmol 0O 1 is one factor stimulating somatostatin release. N isotonic HC1, and (b) a control intragastric Somatostatin is a potent inhibitor of gastric acid infusion of 150 ml 0-15 N NaCl each given over 30 secretion8 and therefore may participate in a minutes. Intragastric pH was continuously negative feedback loop. monitored during the infusions and up to 110 The present study was designed to investigate minutes thereafter by a pH-sensitive radiotelemetry whether intragastric or intraduodenal acid can capsule tethered 5 cm beyond the distal aperture of release somatostatin into peripheral plasma in the infusion tube.9 Experiments were performed in normal subjects. random order separated by at least one week. Methods INTRADUODENAL INFUSION SUBJECTS (a) Three subjects received on one occasion an Fourteen male volunteers, mean age 22 years (range intraduodenal infusion of 4 mmol 0.1 N isotonic Address for correspondence: Dr M R Lucey, Liver Unit, King's College HC1 given over five minutes. (b) Five subjects were Hospital, Denmark Hill, London SE5 8RX. given 60 mmol 0-1 N isotonic HC1 over 30 minutes Received for publication 10 February 1984 into the second part of the duodenum. 1217 Gut: first published as 10.1136/gut.25.11.1217 on 1 November 1984. Downloaded from 1218 Lucey, Wass, Fairclough, O'Hare, Kwasowski, Penman, Webb, and Rees Intraluminal pH was continuously monitored as 30 0 4 mmol HCI intrtduodenal infusion described for intragastric infusions. A o 20 Subject Blood was taken intermittently through an Plasm Subject B somatostatin[ Subject Cu indwelling heparinised needle for estimation of (pg/ml) plasma somatostatin, gastrin, pancreatic poly- 10 peptide, gastric inhibitory polypeptide and motilin. ... .* . Is I Samples for hormone estimation were taken into -15 0 15 30 45 60 90 120 180 240 lithium heparin tubes containing 10 000 KIU Minutes aprotonin, centrifuged at 4°C and separated. The plasma was frozen immediately and stored at -20°C Fig. 1 Plasma somatostatin in three subjects given an until assay. intraduodenal infusion of4 mmol HCI. ASSAYS (b) Five subjects each received an intraduodenal Plasma somatostatin,"° gastrin,"1 pancreatic poly- infusion of 60 mmol 0-1 N HC1 over 30 minutes. See peptide,12 gastric inhibitory polypeptide13 and Figures 2a, 2b. Intraluminal pH fell immediately motilin14 were assayed by radioimmunoassay. from a mean pre-infusion concentration of 6-9±0-4 Before assay somatostatin was extracted using to a nadir of 2-0±0*1 at two minutes and remained Vycor glass; I125 tyrosine-somatostatin (4 pg/tube) low thoughout the infusion. Intraluminal pH rapidly was used as a tracer together with a highly specific returned to its pre-infusion level when the infusion rabbit antisomatostatin serum (final dilution had been completed. Plasma somatostatin rose from 1:150 000) which gave a sensitivity of 10 pg/ml a basal concentration 32±3 pg/ml to a peak of plasma. 54±11 pg/ml at 10 minutes (p<0-05 vs basal); and All subjects gave informed consent in writing. was significantly raised at 20 minutes 46±5 pg/ml These studies were approved by the District Ethical (p<0-01) and 30 minutes 45±5 pg/ml (p<001). Committee of St Bartholomew's Hospital. Plasma somatostatin returned to basal concen- trations by 40 minutes and gradually declined below STATISTICS basal concentrations thereafter. Results are expressed as mean ± 1 SEM. Student's t test for matched pairs was used and p<005 taken as Plasma gastrin, gastric inhibitory polypeptide and significant. pancreatic polypeptide concentrations did not change during this infusion. Plasma motilin fell http://gut.bmj.com/ Results moderately during the acid infusion although this reduction reached significance at one time point INTRAGASTRIC INFUSION only; basal 317±59 pg/ml, 247+57 pg/ml at 20 The basal intragastric pH was 2 and this did not alter minutes (p<0.05). Plasma motilin returned to pre- during or after the infusion of acid alone or saline. infusion levels by 60 minutes after which it declined Basal plasma somatostatin concentrations did not below basal concentrations. differ in either the acid or saline significantly on September 25, 2021 by guest. Protected copyright. experiments and plasma somatostatin concen- Discussion trations did not rise with either infusion. Similarly, neither infusion elicited a significant change in The present study was designed to investigate the serum gastrin concentrations. potential role of gastric acid as a factor mediating the release of circulating somatostatin in man. INTRADUODENAL INFUSION Intragastric and intraduodenal infusion of HC1 in (a) Three subjects each received an intraduodenal dogs raises portal and peripheral plasma somato- infusion of 4 mmol 0.1 N HC1 in five minutes. statin and intraluminal somatostatin concen- Intraduodenal pH fell in all during the infusion to a trations.15 1617 Somatostatin released into the local nadir of 2. In two subjects plasma somatostatin at draining veins after intragastric instillation of a five minutes was slightly raised compared with basal protein meal in dogs is enhanced by prior adjust- concentrations: 12 vs 19 pg/ml; 12 vs 18 pg/ml; while ment of the meal to a pH of 2.18 concentrations were unchanged in the third. See In the present study plasma somatostatin concen- Figure 1. Plasma somatostatin was not increased in trations were not significantly altered by either an any subject thoughout the remaining period of intragastric infusion of 15 mmol 0-1 N HC1 given sampling. Plasma gastrin, pancreatic polypeptide, over 30 minutes, a dose which mimics the estimated gastric inhibitory polypeptide and motilin were maximal postprandial gastric acid output19 or intra- unchanged thoughout each experiment in all three duodenal infusion of 4 mmol 0-1 N HC1 over five subjects. minutes, a stimulus reported to raise plasma concen- Gut: first published as 10.1136/gut.25.11.1217 on 1 November 1984. Downloaded from Does gastric acid release plasma somatostatin in man? 1219 60rnmol HCI 4001- (D - 60mrwnol HCI 8r 300 0-,I 71 Plastrnmotlrin Nadir 6 I (pg/ml) 200 intraduodenal 5 pH 4 100 1 500 Plasma 30 Plasma 400 gastrin 20 I GIP (pg/mi) 10 (pg/mi) 300 70 Plasma 601 T 100 somatostatin 0 (pg/mi) 40r 301 Ptasma - 20 [ PPIng/ml) 401 °l ... -15 0 15 30 60 90 120 180 240 -15 0 15 30 60 90 120 180 240 Minutes Minutes Fig. 2 (a) IntraduodenalpH, plasma gastrin andplasma somatostatin infive subjects given an intraduodenal infusion of60 mmol HCI. (b) Plasma motilin, pancreatic polypeptide, and gastric inhibitory polypeptide infive subjects given an intraduodenal infusion of60 mmol HCI. trations of other putative gut hormones.20 21 22 23 reported to stimulate the release of many other During an intraduodenal infusion of 60 mmol 0O1 N putative gut hormones in man, plasma gastric HC1 over 30 minutes, a grossly supraphysiological inhibitory polypeptide, pancreatic polypeptide and http://gut.bmj.com/ dose which caused a sustained reduction in intra- motilin were measured in addition to somatostatin duodenal pH not found distal to the duodenal bulb and gastrin during the intraduodenal acidification in healthy subjects,24 25 there was a moderate rise in experiments.