Gut: first published as 10.1136/gut.15.8.622 on 1 August 1974. Downloaded from Gut, 1974,15, 622-625 The effect of calcium on gastric acid and gastrin secretion in antrectomized subjects J. CHRISTIANSEN, J. F. REHFELD, AND F. STADIL From the Departments of Surgical Gastroenterology F and Clinical Chemistry, Bispebjerg Hospital, and the Department ofSurgical Gastroenterology C, Rigshospitalet, Copenhagen, Denmark SUMMARY Serum gastrin concentrations and gastric acid secretion were measured in nine antrec- tomized subjects during infusion of saline and calcium gluconate. The basal gastrin level was of the same magnitude as in normal individuals and unoperated duodenal ulcer patients. Calcium infusion only induced gastrin release in one patient and did not induce acid secretion in any of the patients. The intravenous infusion of calcium stimulates at least three months before the examination was gastric acid secretion in normal individuals (Reeder, carried out. The patients all gave informed consent. Jackson, Ban, Clendinnen,Davidson, andThompson, Five of the patients had a gastro-duodenostomy 1970) and in duodenal ulcer patients (Barreras and (Billroth I)and fouragastro-jejunostomy(BillrothII). Donaldson, 1967). The mechanism by which calcium The removed specimen measured at least 10 cm induces gastric secretion is unknown. Studies in along the lesser curvature. patients with primary hyperparathyroidism have Four healthy male volunteers aged 22-25 acted as indicated that a release of gastrin may be the ex- controls. planation (Trudeau and McGuigan, 1969; Christian- The patients and control subjects were examined sen and Aagaard, 1972; Dent, James, Wang, Deftos, after an overnight fast. A Levin tube was placed in http://gut.bmj.com/ Talamo, and Fischer, 1972), while a potentiation the stomach under fluoroscopic control. Intermittent between gastrin and calcium probably does not occur mechanical suction producing a subatmospheric in man (Christiansen and Hendel, 1972). pressure of 150 mm of mercury was used in the Studies by Reeder et al (1970) support the view collection of the secretion. Through a thin polyvinyl that calcium exerts its effect on gastric acid secretion tube welded to the Levin tube a marker substance through the release of gastrin. Rosato, Mullen, was instilled into the stomach (51Cr-EDTA, flow Lanciault, Rosato, and Brooks (1973), who in 30 ml/hr) in order to determine recovery. The on September 29, 2021 by guest. Protected copyright. conscious monkeys also found reproducible eleva- opening of the polyvinyl tube was 10 cm proximal tions in serum gastrin concentrations after calcium to the most proximal opening of the Levin tube. infusion, could not, however, demonstrate any Saline was administered by constant intravenous correlation between serum gastrin concentration and infusion (flow 30 ml/hr) for a control period of two gastric acid secretion. hours. Further information on any relationship between Calcium gluconate was infused over the next two calcium concentration and gastrin stimulation of hours with a flow of 30 mlfhr, delivering 4 mg acid secretion could be found by studying the Ca++/kg-hr. The infusions were performed with a effect of an infusion of calcium salts on the gastric constant infusion pump. acid secretion in patients in whom the antral source Thevolume ofthe secretion collected was measured of gastrin had been removed. for each 15-minute period, and the concentration of H+ and 51Cr-EDTA determined. The concentration Material and Methods of H+ was determined by titration with an auto- titrator (Radiometer, Copenhagen) to pH 7.0. The Nine patients were studied, eight men and one concentration of 51Cr-EDTA was determined in a woman, ages ranging from 22 to 60 years, who had well counter. The volume of the secretion collected had an antrectomy for duodenal or prepyloric ulcer and the output of H+ were corrected as to the actual recovery by way of the marker substance Received for publication 23 April 1974. instilled. 622 Gut: first published as 10.1136/gut.15.8.622 on 1 August 1974. Downloaded from Effect of calcium on gastric acid and gastrin secretion in antrectomized subjects 623 Repeated estimations of osmolarity by freezing 30 minutes. Serum calcium concentration was point reduction were taken as an index of the degree determined by atomic absorption spectrophotom- ofduodenal reflux. etry. Blood samples were taken from an indwelling For the statistical evaluation of the results catheter every 15 minutes. All samples were centri- Student's t test for paired and unpaired data was fuged immediately and serum for gastrin measure- applied (Snedecor and Cochran, 1967). ment was stored at - 20C. The concentrations of gastrin in serum were measured radioimmuno- Results chemically (Stadil and Rehfeld, 1971). The antiserum employed, 2604-8, was raised in rabbits against Figure 1 shows the volume and acid secretion as well synthetic human gastrin (2-17) covalently coupled as serum calcium and serum gastrin concentration to bovine serum albumin (Rehfeld, Stadil, and during the intravenous infusion of saline and calcium Rubin, 1972). The antiserum binds four gastrin gluconate in the nine antrectomized subjects. components of different molecular size, of which the The serum calcium concentration rose significantly largest, gastrin component I, appears unpaired during calcium infusion (serum calcium concentra- (Rehfeld, Stadil, and Vikels0e, 1974), while the tion during saline infusion (mean ± SEM): 9.8 ± 0.1 remaining three-gastrin -33, -17, and -13-each mg/100 ml, and during calcium infusion: 12.0 ± 0.4 circulates in sulphated and nonsulphated pairs (for mg/100 ml, p < 0001). The mean basal serum review see Rehfeld, 1973). Monoiodinated synthetic human gastrin (1-17) was used as tracer. The reli- ability parameters of the assay have been described in detail elsewhere (Stadil and Rehfeld, 1973). Blood samples for calcium determination were taken every 100+ sIWe cakium gluconst@ 30o91h 30m1/h * o~~~~~~~~ 90+ http://gut.bmj.com/ 13 80+ 12 cdslum 112 mg/looml 10, 3- 70- sa-gastrln 70- 60+ pgW 60- F-H on September 29, 2021 by guest. Protected copyright. SHO/mI 50- 50-I. s0 40 vlm 201~~30' ~i 40 . 20 - ,nqH/l 10 - 30± 0- 1,0 0,8. 20± nwH/lS1min 0,61 0,4. 0,2 10-i. 60 120 180 240 minnt sal ine calcium Fig 1 Serum calcium concentration (mg/100 ml), serum infusion infusion gastrin concentration (pg equiv SHG/ml), volume secretion (ml/15 min), acidity (m-equiv H+/l), and Fig 2 Serum gastrin concentration (pg equiv SHG/ml) acid output (m-equiv H+/15 min) in nine antrectomized during infusion ofsaline (0-120 min) and during subjects during a controlperiod ofsaline infusion, and infusion ofcalcium (120-240 min) in nine antrectomized during calcium infusion (means + SEM) subjects 2 Gut: first published as 10.1136/gut.15.8.622 on 1 August 1974. Downloaded from 624 J. Christiansen, J. F. Rehfeld, and F. Stadil gastrin concentration was 48 ± 4 pg equiv/ml, and the same radioimmunoassay technique in fasting did not change significantly during saline infusion normal subjects and unoperated duodenal ulcer (49 ± 4 pg equiv/ml, P 0.50). During calcium patients (Stadil and Rehfeld, 1973). The low mean infusion the mean serum gastrin concentration rose fasting gastrin concentration in the four controls in to 55 ± 6 pg equiv/ml (P > 0-20) but analysing the the present study is caused by very low values in data of the individual subjects shows that this rise one of the subjects. Similar findings have recently could be ascribed to a rise in a single patient (fig 2). been reported by Stern and Walsh (1973), who in The volume of secretion was unchanged (during antrectomized patients with a gastro-duodenostomy saline infusion: 35 ± 2 ml/15 min, and during found fasting and postprandial serum gastrin calcium infusion 34 ± 4 ml/15 min, P > 0.40). concentrations similar to those found in nonulcer Acidity decreased slightly during calcium infusion and unoperated duodenal ulcer subjects. In an- (during saline infusion: 13 ± 3 mequiv H+/l, and trectomized patients with a gastrojejunostomy the during calcium infusion 1 ± 3 mequiv H+/l, P > mean fasting and postprandial serum gastrin values 0-10). were lower, but overlapping between the two groups Acid output also decreased slightly during saline occurred. In the present study the serum gastrin infusion (mean ± SEM: 0.5 ± 0-2 m-equiv/15 min) concentration in the four patients with a Billroth IL and during calcium infusion: 04 ± 0-2 m-equiv anastomosis did not differ significantly from the H+/15 min (P > 0.10). group mean during saline infusion. The main source The secretory responses during saline and calcium of this non-antral gastrin is probably the duodenum infusion were not influenced by duodenal reflux as which in man is known to contain large amounts of no significant differences in the osmolarity of the immunoreactive gastrin. gastric juice were found (see table). Calcium infusion in subjects with an intact The patient who exhibited a rise in serum gastrin stomach results in a significant increase in serum concentration during calcium infusion showed the gastrin concentration. In normal subjects Reeder same pattern of volume and acid secretion as shown et al (1970) found a 200% increase in serum gastrin in figure 2. Mean serum gastrin concentration in concentration after calcium infusion and in duodenal the control subjects was 17 + 2 pg equiv/ml during ulcer patients a 135% increase. In patients with saline infusion and 29 ± 3 pg equiv/ml during primary hyperparathyroidism Dent et al (1972) calcium infusion (P < 0-05), ie, a mean percentage found an increase in serum gastrin concentration of http://gut.bmj.com/ increase of 71 (range 48 to 88 %). Acid output rose 100 to 200% when serum calcium concentration from 0.8 ± 0.1 m-equiv/15 min during saline rose 40-45%, which compares with the increase in infusion to 1-2 ± 0.1 m-equiv/15 min during calcium serum calcium concentration observed in the present infusion (50%; P < 0-05).