GASTRIC FUNCTION IN PRIMARY HYPERPARATHYROIDISM IN MAN. Mary G. McGeown and A. M. Connell

Transcription

1 GASTRIC FUNCTION IN PRIMARY HYPERPARATHYROIDISM IN MAN Mary G. McGeown and A. M. Connell The Renal Unit, Belfast City Hospital, Belfast and Department of Internal Medicine, Cincinnati General Hospital, Ohio, U.S.A. Summary i GASTRIC function was studied pre-operatively in 55 patients with primary ohyperparathyroidism. In all patients the diagnosis of hyperparathyroidism was proved by the removal of an enlarged parathyroid gland or glands, following which the serum calcium fell to normal or below. 2. Twenty patients (38 per cent)gave a history of dyspepsia. All patients were given a barium meal and 12 (22 per cent) were found to have a duodenal ulcer. Male patients had dyspepsia and/or an ulcer more frequently than female patients. 3. Kay's augmented histamine test was carried out in all patients. Six patients (11 per cent) had achlorhydria, six had hyperchlorhydria (11 per cent), the remaining patients secreting normal amounts of acid. The acid secretion did not differ significantly from that of a group of euparathyroid patients studied under similar circumstances who did not have duodenal ulcer on barium meal. 4. Antigastric-parietal-cell antibodies and antithyroid antibodies were absent in all 20 patients, including five with achlorhydria, tested. 5. Serum gastrin was slightly elevated in three out of 10 patients preoperatively, and there was no consistent change following parathyroidectomy. 6. Although there is a high incidence of dyspepsia and of duodenal ulcer in patients with primary hyperparathyroidism they do not tend to have increased acid secretion. Introduclion A relationship between the parathyroid glands and gastric function has been suspected since Gutman et al. (1934) drew attention to the fact that patients with hyperparathyroidism often have gastrointestinal symptoms. Occasionally patients with primary hyperparathyroidism have intractable peptic ulceration though free from the much commoner symptoms due to renal stones or hyperparathyroid bone disease (McGeown and Morrison, 1959). Some patients with multiple endocrine adenomas have gastric hypersecretion, and hypercalcaemia due to parathyroid overactivity may occur in the Zoliinger-Ellison syndrome. Ostrow et al. (1960) found an incidence of peptic ulcer of 9.1 per cent in 429 cases of primary hyperparathyroidism collected from the literature and reported it in three of 17 of their own patients. Gastric function was normal or low in the seven patients in which this was tested, and there was no parallelism between the severity of the hyperparathyroidism and the severity of the ulcer symptoms. Donegan and Spiro (1960) studied gastric function in 217

2 218 IRISH JOURNAL OF MEDICAL SCIENCE five patients before and after removal of parathyroid adenomas. In one patient there was a marked reduction in gastric function post-operatively but in the others it remained unchanged. However, Ward et al. (1964) reported raised basal secretion of acid, although the post-histamine secretion was normal, in five of nine patients with primary hyperparathyroidism. They found a significant fall in basal secretion following parathyroidectomy. In view of these somewhat conflicting reports it seemed of importance to study gastric function in our patients with primary hyperparathyroidism. Materials and Methods The patients consisted of 24 males and 31 females with unequivocal hypercalcaemia at the time of study. Six patients had hyperparathyroid bone disease, 44 had renal stones without bone disease, and five had hypercalcaemia discovered during routine screening of the blood while in hospital for some unrelated condition. The patients were not selected in any way and include most, but not all, of the patients with primary hyperparathyroidism seen between January, 1965 and February, The nature and the object of the investigation was fully explained to them. In all the patients the diagnosis of hyperparathyroidism was confirmed by removal of one or more enlarged parathyroid glands which were overactive on histological examination. In seven patients all the parathyroid glands were enlarged and histologically were thought to be hyperplastic, in one patient there were two enlarged glands thought to be multiple adenomata, and in the remaining fortythree patients there was a solitary adenoma. The enlarged glands measured from 0.8 cm. to 3.5 cm. in their bipolar diameter, with the exception of one of the patients with hyperplasia in who each gland was about as large as a lobe of the thyroid, and in whom a total of 29 g. of parathyroid tissue was removed. In all patients the serum calcium fell to normal or below post-operatively and has remained normal since. Gastric function studies were available in 265 patients who had undergone a barium meal and were found to have no abnormality or a hiatus hernia. The patients were closely questioned for dyspeptic symptoms by one of us (M.G.McG.). Even minor heartburn was recorded. All patients had a barium meal, and the barium meals assessed in a batch, without knowledge of the gastric function tests or the history of dyspepsia, by the other one (A.M.C.). A Kay's augmented histamine test was carried out on all the patients by one or other of two nurses, who also carried out the gastric function tests on the subjects used as controls. The patients were fasted from 22.00, and in the morning a No. 16 Levin gastric tube was screened into position after aspiration of the gastric contents. Histamine was given (0.04 mg/kg body weight) subcutaneously after the basal hour secretion had been aspirated. The samples of gastric contents were titrated to ph 7.0 using a Radiometer ph meter. In 20 patients tests for antibodies to gastric parietal cells (Chanarin, 1968) and thyroid (Doniach and Roitt, 1968) were carried out. Serum gastrin in the fasting state was estimated pre- and post-operatively in ten further patients with primary hyperparathyroidism, who were subsequently proved to have parathyroid adenomas. These patients all had normal barium meals and pentagastrin tests.

3 GASTRIC FUNCTION IN PRIMARY HYPERPARATHYROIDISM 219 Gastric function was studied in the patients used as controls in a similar way to the patients for the basal hour, following which they were given pentagastrin (6/~g/kg body weight). Results Dyspepsia Dyspepsia was common, especially amongst the males where it was found in 41 per cent (Table 1). TABLE I Dyspepsia in patients with primary hyperparathyroidism No. of patients with Dyspepsia Total % with Dyspepsia Male Female Total Barium Meal The results of the barium meal are as shown in Table II. Duodenal ulcers, active or healed, were found in 29 per cent of the males and in 16 per cent of the females. Other abnormalities were found in 10 patients, including hiatus hernia in six, and a diverticulum of the duodenum in one patient. None of the patients had a gastric ulcer. TABLE II Barium meal examination in patients with primary hyperparathyroidism No. of patients with Duodenal Ulcer Total % with Ulcer Male Female Total In eight of these patients there was other evidence of a peptic ulcer. Five male patients had previously undergone an operation for a duodenal ulcer and all of these had a previous history of a haematemesis. Two of the female patients had a history of a haematemesis. Gastric secretion, of acid The secretion of acid during the basal hour for the male patients and for the male control subjects is shown in Table III. There were insufficient numbers to permit the patients being divided into groups according to age. Three patients secreted no acid at all,and seven more than 3.1 mmo! HCl. The

4 220 IRISH JOURNAL OF MEDICAL SCIENCE range of acid secretion was 0,0-6.9 mmol HCI per one hour for the hyperparathyroid patients and mmol HCI per one hour for the "control" subjects. There. is no significant difference in distribution of patients between the two groups (~=0.41, D.F. 3, 0.95<P<0.90). TABLE III Basal acid secretion in males (all ages) mmol HCI Hyperparathyroid "Control" subjects No. patients No Total Range mmol HCI Similar values for the female patients, again compared with the "control" subjects, are shown in Table IV. Six patients had achlorhydria and three secreted more than 3.0 mmol per one hour. The range of acid secretion for the hyperparathyroid patients is and for the control subjects mmol per one hour. The numbers of patients with different levels of acid secretion does not differ significantly between the two groups. (~=0.60, D.F. =3, 0.9<P<0.8). TABLE IV Basal acid secretion in females (all ages) mmol HCI Hyperparathyroid "Cont:-ol" subjects No. patients No Total Range mmol HCI The secretion of acid in the post-histamine hour is shown in Table V. Two males and four females had achlorhydria. The upper limit of secretion of acid for normal males has been taken at 35 mmol HCI in one hour, and for normal females has been taken as 30 mmol HCI in one hour, as these levels seem to be generally accepted. By this standard four males and two females therefore excreted excess amounts of acid in response to histamine.

5 GASTRIC FUNCTION IN PRIMARY HYPERPARATHYROIDISM 221 TABLE V Secretion of acid in the post-histamine hour in patients with primary hyperparathyroidism. mmot HCf No. of Patients Males Females Total Range (mmol HCI) There. are no comparable figures for the post-histamine secretion for nonulcer patients in Northern Ireland. The best controls available are the pentagastrin stimulated values for the patients without known peptic ulcer. Since both histamine and pentagastrin are thought to reflect parietal cell mass, it seems reasonable to compare, the numbers of patients who have hypochlorhydria and hyperchlorhydria by the two stimuli. This has been dona for the males in Table VI and for the females in Table VII. The range of acid secretion for the males with hyperparathyroidism is mmol HCI per one hour, and for the male "controls" is mmol HCI per one hour. TABLE Vl Gastrin secretion in response to stimulation in males. mmol HCl No. of Patients Hyperparathyroidism (histamine, stimulation) "Control" subjects (pentagastrin stimulation 0 2 (8.3%) 4 (2.7%) (75.0%) 124 (84.3%) I- 4 (16.6%) 19 (13.0%) Total Range (mmol HCI) The distribution of patients according to their level of acid secretion is similar in the two groups (exact probability : P=0.2032). The range of acid secretion for the females with hyperparathyroidism is mrnol HCl per one hour, and for the female "controls" is mmol HCI per one hour. The distribution of patients according to their level of acid secretion is similar in the two groups (exact probability : P=0.2509).

6 222 IRISH JOURNAL OF MEDICAL SCIENCE TABLE VII Gastric secretion in response to stimulation in females. mmol HCI No. of Patients Hyperparathyroidism (histamine stimulation) "Control" subjects (pentagastrin stimulation) 0 4(12.9%) 8(6.7%) (6.4%) 4(3.3%) (80.6%) 106 (90.0%) Total Range mmol HCI The secretion of acid during the basal hour was variable amongst the eight patients with hyperplasia or multiple adenomata, three had low values ( mmol HCI) and five had elevated values ( mmol HCI). The secretion of acid during the post-histamine hour was also variable in this group of patients, being well within normal limits in five ( mmol HCI) MALE FEMALE $ ~'R UM CALr rag% 14 t (D,, ix4 e~ I~l Q 0 x 11 )~ ,,, I :~ 3 4 S 6 7 El 9 10 It ACID SECRETION MMOL HCI/IHOUg Fig. 1--Basal excretion of acid plotted against the, maximum serum calcium level in patients with hyperp,arathyroidism.

7 GASTRIC FUNCTION IN PRIMARY HYPERPARATHYROIDISM 223 and raised in three ( ). The patient with the huge hyperplastic glands had the highest basal and post-histamine levels. Unfortunately this patient was operated upon before it became possible to estimate the serum gastrin. The basal secretion in one hour is shown with the maximum level of the serum calcium in Fig. 1 for hyperparathyroid subjects of both sexes. There is no relationship between the basal secretion of acid and the maximum serum calcium. There is also no relationship between the post-histamine secretion of acid and the maximum serum calcium. Gastric parietal-cell antibodies; thyroid antibodies These were not detected in any of the 20 patients, including five with achlorhydria, tested. Serum gastrin The fasting levels of serum gastrin were normal in eight and slightly haised in three patients before parathyroidectomy. It was repeated following parathyroidectomy in nine of the patients, in three of whom it rose, and in remainder remaintd unchanged. (Table VIII). TABLE VIII Fasting serum gastrim in patients with hyperparathyroidism. Patient Serum gastrin pg/ml Pre-operative Post-operative J.W. Less than 100 Less than 100 H.P.,, 100,, 100 K.McC.,, 100,, 100 S.K I.S K.R L.M. 1 O0 300 M.J B.G G.K Discussion Dyspepsia was common amongst this group of patients with primary hyperparathyroidism, occurring amongst 41 per cent of the males and 32 per cent of the females. It must be remembered that minor heartburn on close direct questioning was included as evidence of dyspepsia. In their survey of over 6,000 healty normal subjects, Doll and Avery Jones (1951) found that 32 per cent of the males and 30 per cent of the females had dyspepsia in a Southern English population. The incidence of dyspepsia seems therefore to be increased in the males with hyperparathyroidism, but not in the females.

8 224 IRISH JOURNAL OF MEDICAL SCIENCE The incidence of duodenal ulceration is high in both sexes in primary hyperparathyroidism. Doll and Avery Jones (1951) found the maximum incidence of peptic ulcer (between the ages of 45 and 55) in normal males as just under 10 per cent, whereas 28 per cent of the males with hyperparathyroidism had radiological evidence of a duodenal ulcer. Sixteen per cent of the female patients with hyperparathyroidism had an ulcer compared with 6 per cent in Doll and Avery Jones' normal females. The patients in the present study were predominantly of middle age or older, so comparison has been made with Doll and Avery Jones' groups of similar age. There is a geographical variation in the incidence of duodenal ulcer, and it is possible that there may be a higher incidence of duodenal ulcer in people in Northern Ireland than in the London population studied by Doll and Avery Jones. Some of the patients with evidence of ulcer on barium meal did not have dyspepsia, and not all patients with dyspepsia had ulcers. Ellis and Nicoloff (1968) found six patients with duodenal ulcer amongst their 23 patients with primary hyperparathyroidism. It is difficult to find satisfactory values for acid secretion in normal subjects for comparison with the patients with hyperparathyroidism. Comparison has been made with patients in Northern Ireland who had a barium meal (usually for dyspepsia), but who were found to have no abnormality or a hiatus hernia. They relate to Northern Ireland patients not known to have ulcers and on whom gastric function studies were carried out by the same trained personnel and under the same conditions as the patients under study. The acid secretion under basal conditions was similar for both sexes in the hyperparathyroid and "control" groups. Moreover, the basal acid secretion in patients known to have duodenal ulcers is much higher as a group than the values in these patients with hyperparathyroidism. Male ulcer patients had a range of , whereas none of the male hyperparathyroid patients exceeded 6.9 and only seven exceeded 3.1 mmol per one hour. The study on the gastric function of the patien's with hyperparathyroidism extended over a period of six years and histamine stimulation was the usual stimulus to maximal gastric secretion at the beginning of this period. It was felt desirable to continue with the same method throughout the study rather than change to pentagastrin stimulation when this was later introduced. The only "control" group available had received pentagastrin stimulation. However, both histamine and pentagastrin stimulated acid secretion are thought to reflect gastric parietal cell mass. It seems reasonable to compare the numbers of patients with achlorhydria to maximal stimulation, and with hyperchlorhydria in the patient group and the "control" group. The incidence of achlorhydria and hyperchlorhydria in the patients with hyperparathyroidism is similar to that in the "control" group, for both sexes. None of the patients with hyperparathyroidism had very high acid secretion, and they differ from patients known to have duodenal ulcer. As it has been shown that intravenous infusion of calcium increases acid secretion (Murphy et al., 1966) it is perhaps surprising that increased acid secretion is not a feature of primary hyperparathyroidism. However, Smallwood (1987) found that following calcium infusion the increase in acid and pepsin secretion passed off in 3-4 hours, although the serum calcium was still elevated up to 7-8 hours. A second injection again produced a sharp

9 GASTRIC FUNCTION IN PRIMARY HYPERPARATHYROIDISM 225 increase in gastric secretion, so that the stimulus to the gastric mucosa appeared to be the acute rise in the serum calcium rather than a maintained supranormal level. He also found that when calcium was given intravenously quickly it caused nausea and faintness and if these symptoms occurred there was a profound inhibition of gastric secretion. The response appeared to be mediated in the vagus as it was abolished by vagotomy and by atropine. As previous work had suggested that there might be gastric hyper-secretion in primary hyperparathyroidism, the occurrence of achlorhydria in six patients (11 per cent) was unexpected. Achlorhydria is associated with atrophic gastritis but there is no direct evidence as to whether it was present in our patients as gastroscopic examination and mucosal biopsies were not carried out. It seems possible that atrophic gastritis could be a late result of long-continued hyperc.alcaemia in patients with hyperparathyroidism There is a well recognised association of achlorhydria with pernicious anaemia and atrophic gastritis. Five of the six patients were not anaemic, but one patient had pernicious anaemia. It might be expected that antibodies to gastric parietal cells might occur in patients with hyperparathyroidism associated with achlorhydria, but they were not present in five such patients, nor were they present in 15 other patients with hyperparathyroidism without achlorhydria. The fasting level of serum gastrin was normal in seven and only slightly raised in three patients, and there was no consistent change following parathyroidectomy. It has been found to be normal in ten further patients studied after the completion of this study. It is concluded that there is an excess of dyspepsia and of duodenal ulcers amongst patients with primary hyperparathyroidism. Kay's augmented histamine test showed that 11 per cent of the patients had achlorhydria, and 11 per cent had values which might be regarded as elevated. Both basal and stimulated acid secretion in the patients with hyperparathyroidism is similar to that of patients not known to have peptic ulcer and differs in range from that of patient with peptic ulcers. Amongst the patients tested, none was found to have antibodies to gastric parietal cells or to have significantly increased serum gastrin levels. We are grateful to Mr. M. Bell, F.R.C.S., Mr. T. Kennedy, M.S., F.R.C.S., and Mr. E. Morrison, F.R.C.S., for the parathyroid explorations. Mr. Kennedy also gave much helpful advice on presentation of the data. Mr. J. D. Merrett, Ph.D., gave invaluable h~lp with the statistical analysis. We wish to thank staff nurses L. Crawford and M. Lindsay who carried out the Kay's augmented histamine and pentagastrin tests. Mr. J. Henry, A.I.M.LT., carried out the serological tests for antibodies. We wish to thank Dr. K. D. Buchanan, M.D., M.R.C.P.E., for the estimation of the serum gastrin. References Chanarin, I In Clinical Aspects of Immunology, Ed. P.G.H. Gell and R. R.A. Coombs, Blackwell, Oxford, p Donegan, W. L. and Spiro, H. M Parathyroids and gastric secretion. Gastroenterology, 38, 750. Doniach, D. and Roitt, I. M In Clinical Aspects of Immunology. Ed. P. G. H. Gell and R. R. A. Coombs, B~ackwell, Oxford, p. 938.

10 226 IRISH JOURNAL OF MEDICAL SCIENCE Doll, R. and Avery Jones, F Occupational factors in the aetiology of gastric and duodenal ulcers, with an estimate of their incidence in the general population. Medical Research Council Special Report, No. 276, p. 28. Ellis, C. and Nicoloff, D. M Surg., 96, 114. Hyperparathyroidism and peptic ulcer disease. Arch. Gutman, A. B., Swenson, P. C. and Parson, W. B The differential diagnosis of hyperparathyroidism. J. Amer. reed. Ass., 103, 87. McGeown, Mary G. and Morrison, E Hyperparathyroidism. Postgrad. med. J., 35, 330. Murphy, D. L., Goldstein, H., Boyle, J. and Ward, S Stimulation of human gastric secretion by calcium. J. Appl. Physiol., 21, Ostrow, J. D., Blanshard, G. and Gray, S. J Peptic ulceration in primary hyperparathyroidism. Amer. J. Med., 29, 767. Smallwood, R. A Effects of intravenous calcium administration on gastric secretion of acid and pepsin in man. Gut 8, 592. Ward, J. T., Adesola, A. O. and Welbourn, R. B The parathyroids, calcium and gastric secretion in man and the dog. Gut 5, 173.