European Journal of Clinical Investigation (1979) 9, 141-145
Inhibition of gastrin secretion by hypertonic solutions in patients w i t h pernicious anaemia and duodenal ulcer M. BRANDSBORG, N. A. LOVGREEN, 0. BRANDSBORG & N. J. CHRISTENSEN, Medical Department 11, Aarhus Amtssygehus, 2nd Clinic of Internal Medicine and Department of Surgical Gastroenterology, Aarhus Kommunehospital, University of Aarhus, and Department of Internal Medicine and Endocrinology, Herlev Hospital, Denmark Received 21 March 1978 and in revised form 28 September 1978
Abstract Serum gastrin increased in patients with perni-
cious anaemia after a beef-meal, but decreased after an oral load of glucose, xylose or sodium chloride. 50 g of glucose and 25 or 75 g of xylose suppressed serum gastrin t o appoximately 4% of basal values at 60 min and were slightly more effective than 10 g of sodium chloride. There was no rise in beef-meal stimulated serum gastrin concentration in vagotomized patients and only a slight rise in two patients with duodenal ulcer when an oral dose of 10 g of sodium chloride was given together with the beef-meal. 25 g of xylose suppressed basal serum gastrin concentration significantly in six vagotomized patients. Nasal administration of small amounts of vasopressin decreased basal serum gastrin significantly in all subjects examined. Further studies indicated, however, that vasopressin was only effective when pharmacological plasma concentrations were attained. The inhibitory effects of 10 g of glucose given orally and intraduodenally were compared in six patients with pernicious anaemia. Serum gastrin concentration decreased approximately to the same extent in both experiments. It is concluded that the inhibitory effect of glucose on gastrin secretion most likely is mediated hormonally via osmo-receptors located in the small intestine.. Key words. Duodenal ulcer, glucose, pernicious anaemia, serum gastrin, sodium chloride, vagotomy, vasopressin, xylose.
Introduction
We have previously shown that oral glucose almost cornpletely inhibited beaf-meal stimulated gastrin secretion [l]. This phenomenon was present in patients who previously had undergone either a selective or a highly selective gastric vagotomy because of duodenal ulcer, Correspondence: Dr M. Brandsborg, Medical Department 11, Aarhus Amtssygehus, DK-8000 Aarhus C, Denmark. 0014-2972/79/04004141$02.00 0 1979 Blackwell Scientific Publications
indicating that the effect of glucose on gastrin secretion was not mediated through the vagal nerve. The aim of the present study was t o examine if this inhibitory effect was due t o the presence of a carbohydrate receptor in the gastrointestinal tract or caused by changes in osmolality. We have therefore examined basal and beef-meal stimulated gastrin secretion in response t o oral glucose, xylose and sodium chloride. The possibility that vasopressin may have an inhibitory effect on gastrin secretion was examined. Inhibitory effects of glucose given orally and intraduodenally were also compared. Most experiments were performed in patients with pernicious anaemia or in vagotomized patients, both groups having relatively high basal serum gastrin concentrations. Patients
The following groups of patients were examined: (1) Twenty-six patients with pernicious anaemia (ten males and sixteen females). The mean age was 62 years and ranged from 34 t o 80 years. The duration from the time of diagnosis to the present examination averaged 9 years (range 1-23 years). (2) Seven patients who had undergone a selective gastric vagotomy for a duodenal ulcer and ten patients with a non-obstructing duodenal ulcer (fifteen males and two females). The mean age was 43 years and ranged from 24 t o 60 years. The experiments were performed at least 3-6 months after the operation. Duodenal ulcer was verified radiologically and in most cases also by gast roduodenoscopy. (3) Four healthy male controls. The mean age was 3 1 years and ranged from 24 t o 34 years. Informed consent t o the procedure was obtained from all subjects examined, Procedure
The experiments were performed in the morning after an overnight fast. Blood was collected through a venous catheter in the antecubital space. Serum gastrin concentration was measured in patients
141
142
M. BRANDSBORG et al.
with pernicious anaemia in the basal state and after: (a) a beef-meal, which consisted of 150 g of beef, containing 30 g of protein and 20 g of fat and consumed in approximately 5 min; (b) 50 g of oral glucose dissolved in 200 ml of water; (c) 25 g or 7 5 g of D-xylose dissolved in 100 and 150 ml of water, respectively; (d) 10 g of glucose dissolved in 100 ml of water; (e) 10 g of sodium chloride dissolved in 100 ml of water, (0 application in each nostril of an aqueous solution of vasopressin (vasopressin Sandoz containing 50 IU/ml). Each spray dose corresponded to a dose of vasopressin of approximately 5 IE; (g) 10 g of glucose dissolved in 100 ml of water was given twice at an interval of a few days. ('1 ucose was given orally at one examination and intraduodenally over 5 min through a catheter at the other examination. The position of the catheter was controlled radiologically. Six or seven patients participated in each experiment. In the vagotomized patients and in two patients with duodenal ulcer the following experiments were performed: (a) Basal and beef-meal stimulated gastrin secretion was measured twice in five patients at an interval of a few days. At one examination the beef was given together with 10 g of sodium chloride dissolved in 100 ml of water. (b) Basal serum gastrin concentration was studied in six patients in the fasting state before and after an oral dose of 25 g of D-xylose dissolved in 100 ml of water. Vasopressin was infused intravenously in a group of patients with duodenal ulcer and healthy controls. In nine subjects the dose was 0.001-0.002 IU/min (Vasopressin Sandoz, 10 IE/ml) dissolved in saline 0.9%) and six subjects received 0.008 IU/min. Control experiments were performed in each subject by infusion of saline without vasopressin. A dose of 0.008 IU/min of vasopressin was considered to result in high and unphysiological plasma vasopressin levels [2]. Serum gastrin was measured three times at 15 min intervals in the basal state, at 5 min intervals for 20 min during vasopressin infusion and at 15 and 30 min after a beef-meal had been given. Methods
Serum gastrin concentration was measured by radioimmunoassay as described previously [3, 41. Wilcoxon tests for pair and two samples and the pair Student's ttest were used for statistical analysis [ 5 ] . Regression analysis was performed according to Bailey [ 6 ] . Results
Basal serum gastrin concentration averaged 1039 pg/ml in patients with pernicious anaemia and ranged from 137 t o 2600 pg/ml. Fig. 1 shows results obtained in the basal state and after a beef-meal, 50 g of oral glucose and 25 or 75 g of xylose had been given. In each subject all values were expressed in per cent of a calculated mean basal value. The number of basal samples obtained in each subject in each experiment appears from the
Serum gastrin 70
/
250-
200-
150-
100-
5 1
I
I
1
-I5
0
30
Serum gastrln
Glucose
b
1
Serum gastrin
I
1
I
-10 0
30
I
60 Min
Xylose
I
-60
I
60 Min
-30
I
I
0
30
I
60 Min
Figure 1. Serum gastrin in patients with pernicious anaemia expressed in per cent of a calculated mean basal value in each subject before and after a beef-meal (upper curve), an oral dose of 50 g of glucose (middle curve) and 25 or 75 g of xylose (lower curve).
figures. After beef-meal serum gastrin concentration increased in all subjects from a mean basal value of 11 18 to 2 185 pg/ml at 60 min (u' < 0.05). After 50 g of oral glucose serum gastrin decreased in all of six subjects to a mean value at 60 min of 40% of basal values (2P < 0.05). Serum gastrin concentration decreased in all subjects after xylose and averaged at 30-60 min and at 60 min 5 1% and 43% of basal values, respectively (2P < 0.05). Blood glucose remained unchanged after xylose. Results obtained after 10 g of glucose or 10 g of sodium chloride given orally are presented in Fig. 2. After 10 g of glucose serum gastrin decreased and was
INHIBITION OF GASTRIN SECRETION BY GLUCOSE Serum gastrin
143
Basal serum gastrin PSI mi
%
150,
Glucose
20000
.
0
0
1500-
i 1000-
0 A A
A
80 500-
150-
0
0
100-
I
500 lboo A Serum gastrin pgl ml
50-
I
I
I
I
1500
Figure 3. Decreases in serum gastrin concentration (abscissa) after 50 g of glucose (A), 25 or 75 g of xylose ( 0 ) and I0 g of sodium chloride ( 0 ) in patients with pernicious anaemia plotted versus basal serum gastrin concentrations (ordinate).
I
Figure 2. Serum gastrin in patients with pernicious anaemia expressed in per cent of a calculated mean basal value before and after an oral dose of 10 g of glucose (upper curve) and 10 g of sodium chloride (lower curve).
on the average 71% of basal values at 30-60 min ( 2 P < 0.05) and 73% of basal values at 60 min after glucose (not significant). Corresponding values for 10 g of sodium chloride were 65% and 57% of basal values (u'
Inhibition of gastrin secretion by hypertonic solutions in patients w i t h pernicious anaemia and duodenal ulcer M. BRANDSBORG, N. A. LOVGREEN, 0. BRANDSBORG & N. J. CHRISTENSEN, Medical Department 11, Aarhus Amtssygehus, 2nd Clinic of Internal Medicine and Department of Surgical Gastroenterology, Aarhus Kommunehospital, University of Aarhus, and Department of Internal Medicine and Endocrinology, Herlev Hospital, Denmark Received 21 March 1978 and in revised form 28 September 1978
Abstract Serum gastrin increased in patients with perni-
cious anaemia after a beef-meal, but decreased after an oral load of glucose, xylose or sodium chloride. 50 g of glucose and 25 or 75 g of xylose suppressed serum gastrin t o appoximately 4% of basal values at 60 min and were slightly more effective than 10 g of sodium chloride. There was no rise in beef-meal stimulated serum gastrin concentration in vagotomized patients and only a slight rise in two patients with duodenal ulcer when an oral dose of 10 g of sodium chloride was given together with the beef-meal. 25 g of xylose suppressed basal serum gastrin concentration significantly in six vagotomized patients. Nasal administration of small amounts of vasopressin decreased basal serum gastrin significantly in all subjects examined. Further studies indicated, however, that vasopressin was only effective when pharmacological plasma concentrations were attained. The inhibitory effects of 10 g of glucose given orally and intraduodenally were compared in six patients with pernicious anaemia. Serum gastrin concentration decreased approximately to the same extent in both experiments. It is concluded that the inhibitory effect of glucose on gastrin secretion most likely is mediated hormonally via osmo-receptors located in the small intestine.. Key words. Duodenal ulcer, glucose, pernicious anaemia, serum gastrin, sodium chloride, vagotomy, vasopressin, xylose.
Introduction
We have previously shown that oral glucose almost cornpletely inhibited beaf-meal stimulated gastrin secretion [l]. This phenomenon was present in patients who previously had undergone either a selective or a highly selective gastric vagotomy because of duodenal ulcer, Correspondence: Dr M. Brandsborg, Medical Department 11, Aarhus Amtssygehus, DK-8000 Aarhus C, Denmark. 0014-2972/79/04004141$02.00 0 1979 Blackwell Scientific Publications
indicating that the effect of glucose on gastrin secretion was not mediated through the vagal nerve. The aim of the present study was t o examine if this inhibitory effect was due t o the presence of a carbohydrate receptor in the gastrointestinal tract or caused by changes in osmolality. We have therefore examined basal and beef-meal stimulated gastrin secretion in response t o oral glucose, xylose and sodium chloride. The possibility that vasopressin may have an inhibitory effect on gastrin secretion was examined. Inhibitory effects of glucose given orally and intraduodenally were also compared. Most experiments were performed in patients with pernicious anaemia or in vagotomized patients, both groups having relatively high basal serum gastrin concentrations. Patients
The following groups of patients were examined: (1) Twenty-six patients with pernicious anaemia (ten males and sixteen females). The mean age was 62 years and ranged from 34 t o 80 years. The duration from the time of diagnosis to the present examination averaged 9 years (range 1-23 years). (2) Seven patients who had undergone a selective gastric vagotomy for a duodenal ulcer and ten patients with a non-obstructing duodenal ulcer (fifteen males and two females). The mean age was 43 years and ranged from 24 t o 60 years. The experiments were performed at least 3-6 months after the operation. Duodenal ulcer was verified radiologically and in most cases also by gast roduodenoscopy. (3) Four healthy male controls. The mean age was 3 1 years and ranged from 24 t o 34 years. Informed consent t o the procedure was obtained from all subjects examined, Procedure
The experiments were performed in the morning after an overnight fast. Blood was collected through a venous catheter in the antecubital space. Serum gastrin concentration was measured in patients
141
142
M. BRANDSBORG et al.
with pernicious anaemia in the basal state and after: (a) a beef-meal, which consisted of 150 g of beef, containing 30 g of protein and 20 g of fat and consumed in approximately 5 min; (b) 50 g of oral glucose dissolved in 200 ml of water; (c) 25 g or 7 5 g of D-xylose dissolved in 100 and 150 ml of water, respectively; (d) 10 g of glucose dissolved in 100 ml of water; (e) 10 g of sodium chloride dissolved in 100 ml of water, (0 application in each nostril of an aqueous solution of vasopressin (vasopressin Sandoz containing 50 IU/ml). Each spray dose corresponded to a dose of vasopressin of approximately 5 IE; (g) 10 g of glucose dissolved in 100 ml of water was given twice at an interval of a few days. ('1 ucose was given orally at one examination and intraduodenally over 5 min through a catheter at the other examination. The position of the catheter was controlled radiologically. Six or seven patients participated in each experiment. In the vagotomized patients and in two patients with duodenal ulcer the following experiments were performed: (a) Basal and beef-meal stimulated gastrin secretion was measured twice in five patients at an interval of a few days. At one examination the beef was given together with 10 g of sodium chloride dissolved in 100 ml of water. (b) Basal serum gastrin concentration was studied in six patients in the fasting state before and after an oral dose of 25 g of D-xylose dissolved in 100 ml of water. Vasopressin was infused intravenously in a group of patients with duodenal ulcer and healthy controls. In nine subjects the dose was 0.001-0.002 IU/min (Vasopressin Sandoz, 10 IE/ml) dissolved in saline 0.9%) and six subjects received 0.008 IU/min. Control experiments were performed in each subject by infusion of saline without vasopressin. A dose of 0.008 IU/min of vasopressin was considered to result in high and unphysiological plasma vasopressin levels [2]. Serum gastrin was measured three times at 15 min intervals in the basal state, at 5 min intervals for 20 min during vasopressin infusion and at 15 and 30 min after a beef-meal had been given. Methods
Serum gastrin concentration was measured by radioimmunoassay as described previously [3, 41. Wilcoxon tests for pair and two samples and the pair Student's ttest were used for statistical analysis [ 5 ] . Regression analysis was performed according to Bailey [ 6 ] . Results
Basal serum gastrin concentration averaged 1039 pg/ml in patients with pernicious anaemia and ranged from 137 t o 2600 pg/ml. Fig. 1 shows results obtained in the basal state and after a beef-meal, 50 g of oral glucose and 25 or 75 g of xylose had been given. In each subject all values were expressed in per cent of a calculated mean basal value. The number of basal samples obtained in each subject in each experiment appears from the
Serum gastrin 70
/
250-
200-
150-
100-
5 1
I
I
1
-I5
0
30
Serum gastrln
Glucose
b
1
Serum gastrin
I
1
I
-10 0
30
I
60 Min
Xylose
I
-60
I
60 Min
-30
I
I
0
30
I
60 Min
Figure 1. Serum gastrin in patients with pernicious anaemia expressed in per cent of a calculated mean basal value in each subject before and after a beef-meal (upper curve), an oral dose of 50 g of glucose (middle curve) and 25 or 75 g of xylose (lower curve).
figures. After beef-meal serum gastrin concentration increased in all subjects from a mean basal value of 11 18 to 2 185 pg/ml at 60 min (u' < 0.05). After 50 g of oral glucose serum gastrin decreased in all of six subjects to a mean value at 60 min of 40% of basal values (2P < 0.05). Serum gastrin concentration decreased in all subjects after xylose and averaged at 30-60 min and at 60 min 5 1% and 43% of basal values, respectively (2P < 0.05). Blood glucose remained unchanged after xylose. Results obtained after 10 g of glucose or 10 g of sodium chloride given orally are presented in Fig. 2. After 10 g of glucose serum gastrin decreased and was
INHIBITION OF GASTRIN SECRETION BY GLUCOSE Serum gastrin
143
Basal serum gastrin PSI mi
%
150,
Glucose
20000
.
0
0
1500-
i 1000-
0 A A
A
80 500-
150-
0
0
100-
I
500 lboo A Serum gastrin pgl ml
50-
I
I
I
I
1500
Figure 3. Decreases in serum gastrin concentration (abscissa) after 50 g of glucose (A), 25 or 75 g of xylose ( 0 ) and I0 g of sodium chloride ( 0 ) in patients with pernicious anaemia plotted versus basal serum gastrin concentrations (ordinate).
I
Figure 2. Serum gastrin in patients with pernicious anaemia expressed in per cent of a calculated mean basal value before and after an oral dose of 10 g of glucose (upper curve) and 10 g of sodium chloride (lower curve).
on the average 71% of basal values at 30-60 min ( 2 P < 0.05) and 73% of basal values at 60 min after glucose (not significant). Corresponding values for 10 g of sodium chloride were 65% and 57% of basal values (u'
