Rebound intragastric hyperacidity after abrupt

Transcription

1 Gut, 1991,32, University Department of Medicine, Royal Free Hospital School of Medicine, London C U Nwokolo J T L Smith A M Sawyerr R Pounder Correspondence to: Dr R Pounder, University Department of Medicine, Royal Free Hospital School of Medicine, Pond Street, London NW3 2QG. Accepted for publication 18 March 1991 Rebound intragastric hyperacidity after abrupt withdrawal of histamine H2 receptor blockade C U Nwokolo, J T L Smith, A M Sawyerr, R Pounder Abstract In a series of 24 hour studies, intragastric acidity and plasma gastrin concentration were measured simultaneously in 46 healthy subjects before, during, and 24 to 48 hours after abrupt withdrawal of a histamine H2 receptor antagonist regimen. For 34 days subjects were given either cimetidine 8 mg at night (n=8), ranitidine 15 mg twice daily (n= 1), ranitidine 3 mg at night (n= 12), nizatidine 3 mg at night (n=8), or famotidine 4 mg at night (n=8). All subjects responded to H2 blockade by a decrease in 24 hour intragastric acidity. Withdrawal ofh2 blockade resulted in a significant rise in median nocturnal integrated intragastric acidity in 42 of 46 subjects (+36%; 95% CI +19, +55%) compared with prestudy values, but this rise was not associated with a significant change in the median integrated plasma gastrin concentration (+1%; 95% CI -12, +13%). A statistically significant rise in nocturnal acidity was observed after all regimens, except after dosing with famotidine. After stopping, median daytime integrated acidity and plasma gastrin concentrations in the whole group were raised, but not significantly: values were +15% (95% CI +4, +34%) and +5% (95% CI -2, +12%), respectively. A statistically significant increase in daytime acidity was observed only after dosing with ranitidine. In conclusion, intragastric hyperacidity occurs in most subjects after abrupt withdrawal of a histamine H2 receptor blocker, but this phenomenon is not associated with hypergastrinaemia. When the histamine H2 receptor antagonists were introduced for the management of peptic ulceration, there was concern that they would induce a surge of gastric acid secretion in the days or weeks after their withdrawal.' xperiments were performed to explore this possibility, but most reported that no rebound hypersecretion could be detected.2"'a It came as a surprise when Fullarton et al" reported rebound nocturnal hypersecretion of acid occurring after four weeks of treatment with nizatidine 3 mg at night. They found, in a group of eight duodenal ulcer patients, that mean nocturnal acid output was 39.4 mmol/1 hours before treatment, but rose significantly to 74-1 mmol/1 hours when measured two days after stopping. They observed no significant change in daytime intragastric acidity. The object of this series of experiments was to measure 24 hour intragastric acidity and plasma gastrin concentration before, during, and after five different H2 antagonist regimens. These experiments were performed in conjunction with an extensive research programme investigating the phenomenon of tolerance during continued H2 blockade.'2 '4 Methods 1455 SUBJCTS AND DRUG RGIMNS Forty six of 48 healthy men completed the study. Their median age was 21 years (range 19 to 24 years), median weight was 74 kg (range 6 to 96 kg), and median height was 1 '8 m (range to 1-93 m). Twenty four smoked cigarettes (4-2 cigarettes per day). None had been dosed with an antisecretory drug for at least eight weeks before entry to this study. This study was 'open label,' and all the H2 blockers were supplied by the pharmacy of the Royal Free Hospital, London. The subjects received no antisecretory drug during the first 24 hour simultaneous study of intragastric acidity and plasma gastrin concentration. They were then randomly assigned to receive one of the following regimens: cimetidine 8 mg at night (Smith Kline & French Laboratories Ltd, n=8),'2 either ranitidine 3 mg at night (n=12) or ranitidine 15 mg twice daily (n= 1) (Glaxo Laboratories Ltd); nizatidine 3 mg at night (n=8) (li Lilly and Co Ltd),'2 or famotidine 4 mg at night (n=8) (Thomas Morson Pharmaceuticals). 12 The experiments involving cimetidine, nizatidine, and famotidine were extensions of previously published studies investigating 'tolerance,' but the ranitidine experiments were not part of the earlier 'tolerance' experiments. 12 The subjects were dosed with an H2 receptor antagonist for 34 days, and 24 hour profiles of intragastric acidity and plasma gastrin concentration were measured for assessment of antisecretory activity on day 29 of dosing. The final 24 hour study of acidity and plasma gastrin concentration was begun 24 hours after the last oral dose of each H2 blocker. XPRIMNTAL DSIGN The subjects were studied using the Royal Fzee Hospital protocol,'2"' with minor changes as specified below. The subjects began fasting at 133 hours, and were provided with a standard light supper at 1815 hours. They had nothing to eat or drink until 213 hours, when a 1 FG Salem Sump nasogastric tube (Argyle Medical, Crawley, West Sussex) was positioned in the stomach; its position was checked by a water recovery test. Aliquots (5-1 ml) of intragastric contents were aspirated hourly throughout the study, and the ph of each aliquot was measured immediately to the nearest.1 ph unit by means of a glass electrode and digital ph meter Gut: first published as /gut on 1 December Downloaded from on 26 August 218 by guest. Protected by copyright.

2 1456 Nwokolo, Smith, Sawyerr, Pounder LU r, Day Figure 1: 24 hour median hourly intragastric acidity before dosing (), on day 29 ofdosing (-4--), and 24 hours after stopping cimetidine 8 mg at night ( ), in eight healthy subjects. N=nightcap, B= breakfast, C=coffee, L= lunch, T=tea, D=dinner. Figure 2: 24 hour median hourly acidity before dosing (--), on day 29 ofdosing (-4-), and 24 hours after stopping ranitidine 15 mg twice daily( a, in 1 healthy subjects. Figure 3: 24 hour median hourly intragastric acidity before dosing (--), on day 29 ofdosing (-4), and 24 hours after stopping ranitidine 3 mg at night (* a), in 12 healthy subjects. ci 11 N 81F 6 V 41F C.) a a a a 9 a a a a A a a a a a a a a a (Radiometer, Copenhagen). The electrode was calibrated with standard buffers (ph 7-, 4.1, and 1.9; Radiometer, Copenhagen) before and after every six samples in each hourly batch of aspirates. very hour from 23 hours to 23 hours the next day (apart from 1, 3, 5, and 7 hours) blood was taken via a venous cannula for assay of the plasma gastrin concentration. The blood was collected in lithium heparin tubes which contained -2 ml aprotinin (Bayer UK Ltd, Newbury). The tubes were centrifuged immediately, and the plasma transferred to plastic tubes and frozen to -2 C. All the plasma samples from each subject were analysed for gastrin in one batch, by radioimmunoassay using the antibody GAS 179 in Professor Bloom's laboratory at the Royal Postgraduate Medical School London. 16 The subjects were fully ambulant around the ward during the study. The food and environmental conditions for all studies were identical to N Day... Day...*... Gut: first published as /gut on 1 December Downloaded from on 26 August 218 by guest. Protected by copyright.

3 Rebound intragastric hyperacidity after abrupt withdrawal ofhistamine H2 receptor blockade N A B I <8 l 6 < 4 Figure 4: 24 hour median 2 hourly intragastnc acidity before dosing (-4-), on day 29 ofdosing (-4), and 24 hours after stopping * nizatidine 3mg at night (- ),ineight healthy subjects. Figure 5: 24 hour median hourly intragastric acidity before dosing (-4-), on day 29 ofdosing (@--), and 24 hours after stopping famotidine 4 mg at night (* a), in eight healthy subjects. : a U 12 1oo V 2 h N those used in earlier experiments at the Royal Free Hospital.'5 The standard meals (bedtime snack, breakfast, coffee, lunch, tea and dinner) were eaten at 2245, 815, 145, 1315, 1545, and 1815 hours, respectively. The bedtime doses of medication were taken at 235 hours; the morn- TABL I Median integrated nocturnal (24-8 hours) intragastric acidity (mmol.hll) in 46 healthy subjects before, during, and after dosing with an H2 antagonist regimen Cimetidine Ranitidine Ranitidine Nizatidine Famotidine (8 mg, (15 mg, (3 mg, (3 mg, (4 mg, nocte) bd) nocte) nocte) nocte) No Value before dosing (day ) During dosing (day 29) 112t 153t 33t 5t 133t (-82%) (-69%) (-93%) (-9%) (-74%) After dosing (day 36) 666* 7t 79t 614* 561 (+9%) (+43%) (+67%) (+28%) (+ 1%) p value compared with before dosing *=.5; t=-1 (Wilcoxon rank sum test). TABL II Median integrated daytime (9-23 hours) intragastric acidity (mmol.hll) in 46 healthy subjects before, during, and after dosing with an H2 antagonist regimen Cimetidine Ranitidine Ranitidine Nizatidine Famotidine (8 mg, (15 mg, (3 mg, (3 mg, (4 mg, nocte) bd) nocte) nocte) nocte) No Value before dosing (day ) During dosing (day 29) * 353* 263 (-4%) (- 1 1%) (-9%) (-4%) (-32%) After dosing (day 36) t 526t (-2%) (+ 11%) (+28%) (+13%) (+ 1%) p value compared with before dosing *=.5; t= 1 (Wilcoxon rank sum test). B C L T D N Day _ Day...*.. ing dose of ranitidine 15 mg was taken by one group at 83 hours. STATISTICAL ANALYSS Profiles of intragastric acidity and plasma gastrin concentration and were obtained for each subject. The area under the curve for each profile was calculated by the trapezoid rule, with integrated acidity expressed as mmol.h/l and plasma gastrin concentration as pmol.h/l. Values of integrated acidity and plasma gastrin concentration were calculated for the night time ( to 8 hours), and daytime (9 to 23 hours). The significance of observed differences between groups was assessed using the Wilcoxon matched pair signed rank test. All statistical calculations were made using the Oxstat program (Wallingford Computing Services, Wallingford). THICAL AND SAFTY ISSUS The studies were approved by the thics Committee of the Royal Free Hospital, and written consent was obtained from each subject. Routine laboratory safety studies were performed before and after each study. Results Forty six subjects tolerated all the experiments Gut: first published as /gut on 1 December Downloaded from on 26 August 218 by guest. Protected by copyright.

4 1458 Figure 6: Changes in integrated nocturnal acidity compared with changes in integrated nocturnal plasma gastrin concentration (24-8 hours) in 46 healthy subjects dosed with an H2 antagonistfor 35 days. The acidity or plasma gastrin concentration after dosing is expressed as a percentage of the value before dosing (cimetidine 8 mg at night= l; ranitidine 15 mg twice daily=@; ranitidine 3 mg at night=o; nizatidine 3 mg at night= A;famotidine 4 mg at night=-). Figure 7: Changes in integrated daytime acidity compared with changes in integrated nocturnal plasma gastrin concentration (9-23 hours) in 46 healthy subjects dosed with an H2 antagonist for 35 days. The acidity or plasma gastrin concentration after dosing is expressed as a percentage of the value before dosing (cimetidine 8 mg at night=-; ranitidine 15 mg twice daily=-; ranitidine 3 mg at night=o; nizatidine 3 mg at night= A;famotidine 4 mg at night=-). a) J 151- inn u L 2C CD C: 1 ecn a).. :L, Q 5 5~~~~~~o Gastrin (% of predosing) 3 * :6234 ) _. A.~~~~ 1~~~~~~~~~~~~ so 1C o ~~~~~~~~~~~ - ~ 5 O U 5 1 Gastrin (% of predosing) without any adverse event, but two subjects allocated to receive ranitidine 15 mg twice daily were withdrawn from the study (one had tonsillitis and the other family commitments which precluded continuation in the study). Biochemical and haematological profiles were normal before and after the experiments. Full compliance was reported by all the subjects. 24 HOUR INTRAGASTRIC ACIDITY Figures 1 to 5 show the profiles of 24 hour median hourly intragastric acidity for the five different regimens of H2 blockade before, during, and immediately after abrupt withdrawal of the antisecretory drugs. Compared with values before dosing, each of the five H2 blocker regimens was associated with a significant decrease in median daytime acidity during dosing, and four of the five regimens were followed by a significant rise in median integrated nocturnal intragastric acidity after dosing (Table I). Analysis of variance could detect no significant difference in the change in nocturnal intragastric acidity (expressed as a percentage of pretreatment acidity) between the five H2 antagonist regimens. Two of the five regimens were associated with a significant rise in median integrated intragastric acidity during the daytime 34 to 48 hours after the last dose of an H2 antagonist (Table II), but analysis of variance detected no significant difference between the groups. ' a j a Ll o In p a D Nwokolo, Smith, Sawyerr, Pounder : Ranitidine 3 mg at bedtime (n = Ranitidine 15 mg twice daily (n - Cimetidine 8 mg at bedtime (n = 8) A Nizatidine 3 mg at bedtime (n 2431 Famotidine 4 mg at bedtime (n = - 8) 2 1 Ranitidine 3 mg at bedtime (n = = 12) * Ranitidine 15 mg twice daily (n - = 1) Cimetidine 8 mg at bedtime )n = 8) A Nizatidine 3 mg at bedtime (n = 8) * Famotidine 4 mg at bedtime (n = = 8) PLASMA GASTRIN CONCNTRATION Compared with values before dosing, the 24 hour profiles of plasma gastrin concentration were raised during dosing with all of the antisecretory drug regimens,'2 but there was no significant change in the median integrated plasma gastrin concentration during the 24 hours after stopping the drugs. The individual data points for all 46 subjects, correlating changes in either nocturnal or daytime integrated intragastric acidity with integrated plasma gastrin concentration after withdrawal of H2 blockade, are shown in Figures 6 and 7. Figure 6 shows that, compared with before dosing, the rise (median +36%, 95% CI +19, +55%) in nocturnal intragastric acidity, observed in 42 of the 46 subjects after withdrawal of dosing with an H2 blocker regimen, was not associated with a significant change in the plasma gastrin concentration (median + 1%; 95% CI - 12, + 13%). Figure 7 shows that daytime acidity was increased in 33 of the 46 subjects (median + 15%; 95% CI +4, + 34%), but that the integrated plasma gastrin concentration was unchanged (median +5%; 95% CI -2, + 12%). Discussion The results of these studies confirm and extend Fullarton and colleagues' original report of rebound nocturnal hyperacidity": increased nocturnal intragastric acidity does occur after 12) 1 ) = 8) Gut: first published as /gut on 1 December Downloaded from on 26 August 218 by guest. Protected by copyright.

5 Rebound intragastric hyperacidity after abrupt withdrawal ofhistamine H2 receptor blockade 1459 abruptly stopping short acting histamine H2 receptor antagonists. In retrospect, why was it generally thought that H2 blockers do not induce hypersecretion of acid?8 Some experiments were performed before the expected complete plasma elimination of the H2 antagonist, and others were completed some weeks after withdrawal of treatment'-7 9 1; almost all the original experiments examined maximal acid secretion, stimulated by either histamine or pentagastrin. It is now clear that measurement of nocturnal intragastric acidity represents the prolonged observation of 'spontaneous' gastric function, unaffected by meals or outside stimuli. Reliable studies of basal gastric acid secretion are exceptionally difficult,"7 but remarkable reproducibility can be achieved in studies of nocturnal intragastric acidity.'3 This sensitivity has allowed the detection of rebound hyperacidity, and similarly the detection of nocturnal tolerance to H2 blockade The mechanism of nocturnal rebound hyperacidity remains unclear. A drug induced decrease in intragastric acidity induces a rise in the plasma gastrin concentration,2 and it is possible that continued H2 blockade could induce proliferation of G cells or sustained hypergastrinaemia. The present studies show that when the phenomenon of rebound hyperacidity is occurring, plasma gastrin concentrations have returned to the values before dosing. Hence, rebound hyperacidity is not the result of hypergastrinaemia - but it could be argued that the concentration of gastrin remains 'inappropriately' high, as it is not decreased at a time when intragastric acidity is increased.2 It is possible therefore that part of the phenomenon of rebound acid hypersecretion results from a persisting drug induced change in gastrin release. The original experiments, using pentagastrin or histamine tests to measure maximal acid output,2`79 ' showed that there is no increase in parietal cell mass after short term treatment with an H2 blocker. It remains a possibility that prolonged pharmacological control of gastric acid secretion results in an overactivity of, or increased sensitivity to, vagal drive. It will be very difficult to devise experiments in man that can either prove or disprove whether rebound hypersecretion is due to vagal drive. An alternative mechanism to explain the rebound hypersecretion of acid after withdrawal of an H2 antagonist is that H2 receptors become more sensitive, or 'up regulated' during treatment. Aadland and Berstad21 showed that the mean acid output in response to a low dose of intravenous histamine increased from 6 7 to 1 1 mmol/hour, when measured before and 6-84 hours after stopping four weeks of treatment with cimetidine 1 g/day. The acid output measured one to four weeks later had returned to pretreatment values. This response is consistent with augmented parietal cell sensitivity to histamine stimulation. Jones et a122 assessed the sensitivity of the H2 receptor using impromidine, a specific H2 agonist, before and after a three month course of ranitidine 15 mg at night. Six duodenal ulcer patients were studied 1 hours after the last dose of ranitidine: basal acid output increased from 1F2 to 2 8 mmollhour, and acid output during maximal impromidine infusion increased from 36&9 to 44 2 mmollhour. The antisecretory effect of intravenous ranitidine given after impromidine was also enhanced at the end of treatment. Thus, the hypersecretory response could be a result of 'up regulation,' whether by increased number, affinity, or activity (in coupling to adenylate cyclase) of the H2 receptors. Fullarton et al suggested that rebound hyperacidity could be one explanation for recurrent peptic ulceration after withdrawal of H2 blockade." Recent experiments at the Royal Free Hospital have shown that the phenomenon of nocturnal hyperacidity exists for only six days after 25 days of dosing with ranitidine 3 mg at night - from days 9-21 after dosing there is no significant change in intragastric acidity, compared with values before dosing.23 Miss Doris lliott prepared this manuscript. The study was supported by grants from Glaxo Group Research Limited. nthusiastic technical assistance was provided by Nurse J Sercombe and the following clinical medical students: P Peyser, J Tuckley, J Greening, A mmanuel, A Muir, J Ablett, L Pallis, H Reid, H Seymour. 1 Saunders JHB, Wormsley KG. Long-term effects and aftereffects of treatment of duodenal ulcer with metiamide. Lancet 1977; i: Crean GP, Holden RJ, MacKenzie I, Hearns JR. The effects of 6 weeks' administration of cimetidine on gastric acid secretion. In: Burland WL, Simkins MA, eds. Cimetidine: Proceedings of the Second International symposium on Histamine H2-Receptor Antagonists. Amsterdam: xcerpta Medica, 1977: Gillespie G, Gray GR, Smith IS, et al. Short-term and maintenance cimetidine treatment in severe duodenal ulceration. In: Burland WL, Simkins MA, eds. Cimetidine: Proceedings of the Second International Symposium on Histamine H2-Receptor Antagonists. Amsterdam: xcerpta Medica, 1977: Hetzel DJ, Hansky J, Shearman DJC, et al. Cimetidine treatment of duodenal ulcer: short-term clinical trial and maintenance study. Gastroenterology 1978; 74 (suppl): Binder HJ, Cocco A, Crossley RJ, et al. Cimetidine versus intensive antacid therapy for duodenal ulcer: a multicentre trial. Gastroenterology 1978; 74 (suppl): Domschke W, Domschke S, Demling L. A double-blind study of cimetidine in patients with duodenal ulceration: clinical, kinetic and gastric and pancreatic secretory data. In: Burland WL, Simkins MA, eds. Cimetidine: Proceedings of the Second International Symposium on Histamine H2-Receptor Antagonists. Amsterdam: xcerpta Medica, 1977: Sewing KF, Hagie L, Ippoliti AF, et al. ffect of one-month treatment with cimetidine on gastric secretion and serum gastrin and pepsinogen levels. Gastroenterology 1978; 74 (suppl): Winship DH. Cimetidine in the treatment of duodenal ulcer. Gastroenterology 1978; 74: Brown P, Ceravolo C, Zambelli A. Rebound rise in gastric acid secretion: study of cimetidine vs trithiozine: a preliminary report. Curr Ther Res 1978; 23: Bodemar G, Walan A. Maintenance treatment of recurrent peptic ulcer by cimetidine. Lancet 1978; i: Fullarton GM, McLaughlan G, MacDonald A, Crean GP, McColl KL. Rebound nocturnal hypersecretion after four weeks treatment with an H2 receptor antagonist. Gut 1989; 3: Nwokolo CU, Smith JTL, Gavey C, Sawyerr A, Pounder R. Tolerance during 29 days of conventional dosing with cimetidine, nizatidine, famotidine or ranitidine. Aliment Pharmacol Therap 199; 4 (suppl): Smith JTL, Nwokolo CU, Gavey C, Pounder R. Tolerance during eight days of high-dose H2-blockade: placebocontrolled studies of 24 hour acidity and gastrin. Aliment Pharmacol Therap 199; 4 (suppl): Nwokolo CU, Sawyerr A, Smith JTL, Pounder R. Intravenous pentagastrin can induce tolerance to H2-blockade in man. AlimentPharmacol Therap 199; 4 (suppl): Lanzon-Miller 5, Pounder R, Hamilton MR, Chronos NAF, Ball 5, Mercicca J, et al. Twenty-four hour intragastric acidity and plasma gastrin concentration in healthy subjects and patients with duodenal or gastric ulcer or pernicious anaemia. Aliment Pharmacol Therap 1987; 1: Bryant MG, Adrian T. Gastrin. In: Bloom SR, Long RG, eds. Radioimmunoassay of gut regulatory peptides. London: Saunders, 1982: Sharms B, Axelson M, Pounder R, et al. Acid secretory capacity and plasma gastrin concentration after administration of omeprazole to normal subjects. Aliment Pharmnacol Therap 1987; 1: Gut: first published as /gut on 1 December Downloaded from on 26 August 218 by guest. Protected by copyright.

6 146 Nwokolo, Smith, Sawyerr, Pounder 18 Wilder-Smith CH, rnst T, Gennoni, M, Zeyen B, Varoa L, Roehmel, et al Loss of acid suppression dosing with H2- receptor antagonists. Aliment Pharmnacol Therap 1989; 4 (suppi 1): Rogers MJ, Primrose JN, Holmfield H, Johnston D. The effects of 15 days of dosing with placebo, sufotidine 6 mg noctre, or sufotidine 6 mg twice daily, on 24-hour intragastric acidity and 24-hour plasma gastrin. Aliment Pharmacol Therap 199; 4 (suppl 1): Pounder R, Smith JTL. Drug-induced changes of plasma gastrin concentration. Gastroenterology Clin North Am 199; 19: Aadland, Berstad A. Parietal and chief cell sensitivity to histamine and pentagastrin stimulation before and after cimetidine treatment in healthy subjects. Scand J Gastroenterol 1979; 14: Jones DB, Howden CW, Burget DW, Silletti C, Hunt RH. Alteration of H2-receptor sensitivity in duodenal ulcer patients after maintenance treatment with an H2-receptor antagonist. Gut 1988; 29: Prewett J, Hudson M, Nwokolo CU, Sawyerr AM, Pounder R. Nocturnal intragastric acidity during and after a period of dosing with either ranitidine or omeprazole. Gastroenterology 1991; 1: Gut: first published as /gut on 1 December Downloaded from on 26 August 218 by guest. Protected by copyright.