Is There a Place for Antacids in the Treatment of HeZicobacter pylori Infection? K. BERSTAD, R. WEBERG &L A . BERSTAD Division of Gastroenterology,@edical Dept. A, Haukeland University Hospital, Bergen, Norway
3
Scand J Gastroenterol Downloaded from informahealthcare.com by Queen's University on 12/30/14 For personal use only.
Berstad K, Weberg R, Berstad A . IS there a place for antacids in the treatment of Helicobacterpylori infection? Scand J Gastroenterol 1992;27:10061010. f i w o pilot studies were performed to determine whether aluminium-containing antacids may have a place in the treatment of Helicobacter pylori infection. The urease activity of H . pylori is cytopathic to gastric epithelium, and inhibition of this enzyme may have therapeutic potential. In the first study 24 subjects, 12 of which were infected with H . pylori, were given 1 tablet of chewable aluminium hydroxidecontaining antacids 10 min before a “C-urea breath test. Gastric urease activity was suppressed by 33.3% ( p = 0.02) in the H . pylori-positive subjects (none became negative) within 40 min after administration of the tablet. Gastric H . pylori infection can be effectivelyeradicated by triple regimens containing bismuth salts, tetracycline, and metronidazole. Owing to adverse effects of this treatment and concern for possible neurotoxicity of bismuth, a bismuth substitute is warranted. Hence, in the second study, 20 subjects infected with H . pylori were treated with 1 antacid tablet 4 times daily between meals, plus 500 mg oxytetracycline and 200 mg metronidazole 4 times daily with meals for 2 weeks. Individual H . pylori status was assessed by the “C-urea breath test. Four weeks after cessation of treatment, H. pylori was eradicated in 45% (9 of 20) of the subjects (95% confidence interval, 23.1-68.5%). Thirty per cent (6 of 20) observed one or more adverse effect regarded as moderate or severe, of which loose stools and headache were the most common. We conclude that the suppressive effect of antacids on gastric urease activity may be a mechanism by which aluminium-containing antacids promote peptic ulcer healing and that a low-dose regimen of antacids as a bismuth substitute in the triple treatment for H . pylori eradication is inferior to the triple treatment already established
J
Key words: Antacids; Helicobacter pylori; metronidazole; oxytetracycline; urease K . Berstad, M . D.,Dioision of Gastroenterology, Medical Dept. A , Haukeland University Hospital, N5021 Bergen, Norway
Antacids containing aluminium hydroxide promote healing of duodenal and gastric ulcers in doses that have a rather feeble effect on intragastric acidity (1,2). The mechanisms of action for aluminium-containing antacids beyond acid neutralization are by and large unknown. An in vitro study demonstrated that aluminium-containing antacids inhibit growth of Helicobacter pylori even at low concentrations (3). In previous studies we found reduced gastric H . pylori colonization in biopsy specimens after 4 weeks’ treatment with a low-dose regimen of aluminium hydroxide-containing antacids (4). Moreover, 2 weeks’ treatment with the same antacid regimen suppressed, but did not abolish, an enhanced gastric urease activity as assessed with the I4Curea breath test ( 5 ) . Gastric H . pylori infection is increasingly acknowledged as an important cause of gastritis and peptic ulcer disease (6,7). H . pylori urease, which has a high substrate affinity (8), is observed in all wild strains of H. pylori (9). The urease activity is toxic to human gastric epithelial cells (10, 11) and causes intracellular vacuolization (12). Ammonia generated from the hydrolysis of urea damages the gastric mucosa (13) and enables the bacteria to survive in gastric acid (14). Urease inhibitors may reduce mucosal damage and enhance
the host’s ability to clear the infection (15). Hence, inhibition of H . pylori urease activity may have therapeutic potential (10). To study the prompt effect of antacids on gastric urease activity, a I4C-urea breath test was commenced 10 min after administration of one antacid tablet in subjects with and without H . pylori infection (study 1). Treatment of H . pylori infection with bismuth salts in combination with tetracycline and metronidazole leads to 89-97% eradication of the bacteria ( 1 6 1 9 ) . However, adverse effects like malaise, nausea, and diarrhoea are common, and there is concern about the possible neurotoxicity of bismuth (20). In the second study an aluminium-containing antacid was given as a bismuth substitute together with oxytetracycline and metronidazole in a triple-treatment regimen to 20 subjects infected with H . pylori (study 2). MATERIALS AND METHODS
Subjects Study I . Twenty-four clinically healthy volunteers (15 men) with a mean age of 41.0 years (range, 26-60 years)
Antacids and H. pylori
Scand J Gastroenterol Downloaded from informahealthcare.com by Queen's University on 12/30/14 For personal use only.
were enrolled. All subjects were recruited among hospital employees. None of the participants had a history of peptic ulcer disease. Study 2. Twenty volunteers (8 men) with a mean age of 52 years (range, 35-76 years) were recruited among hospital employees (n = 9) known from previous studies to harbour the bacteria and among patients referred to the clinic ( n = 11). The referred patients were enrolled consecutively among patients in whom upper endoscopy did not reveal ulcer, oesophagitis, or malignancy, and in whom H . pylori infection was diagnosed by the I4C-urea breath test. Procedure and treatment Study 1. Each subject was studied twice after an overnight fast. Before the initial control breath test, 100ml of tap water was given 10 rnin before the base-line breath sample (see below). One to 3 days later the same procedure was followed by one tablet of chewable antacids along with the 100ml of water. The antacid tablet (Link@, ll00mg, A/S Apothekernes Laboratorium, Norway) consisted of aluminium hydroxide and magnesium carbonate in a co-dried gel, with an acid-buffering capacity of 30 mmol per tablet. The subjects were to abstain from tobacco on the mornings before the experiments. Study 2 . Treatment was commenced at an average 19 days (SD, 9; maximum, 38 days) after the initial breath test. The second breath test was performed 4 weeks after cessation of treatment. The subjects were to abstain from tobacco on the mornings before the tests. The subjects were given 500 mg oxytetracycline (Oxytetral@,250-mg tablets, A/S Apothekernes Laboratorium) plus 200 mg metronidazole (FlagylB, 200-mg tablets, RhGnePoulenc Rorer) four times daily at meals and one antacid tablet (Link, 1100 mg) four times daily between meals for 2 weeks. To monitor adverse effects, the subjects were to fill in a self-registration questionnaire handed out at the beginning of treatment and returned at control 6 weeks later. Patient compliance was assessed by the number of returned tablets. No subjects were excluded owing to inadequate compliance. Assessment of H. pylori status with the 14C-ureabreath test 14C-labelled urea, containing 92.5 kBq (2.5 pCi), was given orally with 50 ml of water (0.96 pM) immediately after completion of the base-line breath sample. Double breath samples were obtained every 10th rnin for 30 min. I4C activities of samples and standards were counted in a p-counter. Cumulative expired I4C activity (CA) was calculated in accordance with the trapezoid method with adjustment for assumed individual C 0 2production (9 mmol CO,/kg/h) and body weight. CA is thus an estimate of gastric urease activity, and its unit is percentage recovery of I4C during the first 30 rnin after I4C-urea administration (% recovery 30 min). Biometric evaluation of human gastric urease activity has indicated a cut-off value of 1.47% recovery 30 rnin in random
1007
t
Control
10-40 rnin
a f t e r Antacids Fig. 1. Mean (horizontal bar) and individual gastric urease activities in 12 clinically healthy subjects infected with Heficubacter pylori. Control and 10-40min after one tablet of antacids ( p = 0.02).
subjects (21). Hence, subjects with a CA below 1.47 were regarded as not infected with H . pylori ( H . pylori-negative), whereas those with a CA above 1.47 were regarded as infected with the bacteria ( H . pylori-positive). Ethics Study 1. Informed consent was obtained from each participant. The use of the I4C-urea breath test was approved by the National Institute of Radiation Hygiene, Norway. Study 2 . Written informed consent was obtained from each subject. The study was approved by the Ethics Review Committee (Region 3) and the Norwegian Medicines Control Authority. The study was conducted in accordance with the Good Clinical Practice guidelines issued by the EEC (1990). Statitistics Normality was assessed with the Wilk-Shapiro W test. Student’s t test for paired sample and the Sign test were applied to the effect of treatments (two-tailed). Correlation was assessed by Spearman rank. Figures are given as arith-
1008
K. Berstad ef al.
metic mean and 95% confidence interval (95% CI). P < 0.05 was regarded as statistically significant. Study 1. Effect of antacids on gastric urease activity was calculated as the difference between CA with antacids and control. Study 2 . Relative frequencies of H. pylori status after treatment were assessed in accordance with the intention to treat. Confidence interval for proportion was calculated with Confidence Interval Analysis (22), using the exact method.
Scand J Gastroenterol Downloaded from informahealthcare.com by Queen's University on 12/30/14 For personal use only.
RESULTS Study I There was no significant correlation between CA and weight or age within the two groups in accordance with H . pylori status. CA ranged from 1.63 to 12.8 among the 12 H . pylori-positive (6 male) subjects and from 0.21 to 1.14 among the 12 negative (9 male) subjects. Mean ages were 47.1 years (95% CI, 41 3-52.4) and 34.8 years (95% CI, 30.1-39.6) for the H. pylori-positive and -negative subjects, respectively
...............................
Before Treat ment
4 Weeks after Cessation
Fig. 2. Individual gastric urease activity before and 4 weeks after cessation of treatment with antacids, oxytetracycline, and metronidazole in 20 subjects infected with Helicobacter pylori. Cut-off value at 1.47 indicated with a dotted line.
( p = 0.001). The overall mean relative effect of antacids on gastric urease activity was a reduction in CA by 2Y.8% ( p = 0.035).
Effect of antacids in H. pylori-positive subjects. Compared with control level, gastric urease activity was suppressed in 10 of the 12 H pylori-positive subjects after antacids ( p = 0.039). Mean control CA was 7.36 (95% CI, 5.08-9.66) and 4.90 (95% CI, 3.87-5.94) with antacids. Mean effect of antacids was 2.45 (95% CI, 0.4W.44) ( p = 0.020), which was a 33.3% reduction in mean CA (Fig. 1). None of the H.pylori-positive had a CA less than 1.47 (cut-off value) after treatment. Effect of antacids in H. pylori-negatiue subjects. Compared with control level, gastric urease activity was increased in 11 of the 12 H. pylori-negative subjects after antacids ( p = 0.006). Mean control CA was 0.41 (95% CI, 0.26-0.57) and 0.55 (95% CI, 404.70) with antacids. Mean effect of antacids was -0.13 (95% CI, -0.24 to -0.03) ( p = 0.016), which was a 31.7% increase in mean CA. None of the H. pylori-negative had a CA above 1.47 (cut-off value) after treatment.
Study 2 Effect of treatment on H. pylori infection. Overall mean CA before treatment was 10.3% recovery 30min (range, 2.8-17.0). When a cut-off value of 1.47% recovery 30 min was used, 9 of the 20 H . pylori-positive subjects (45%) were negative (range, 0.12-0.90) 4 weeks after cessation of treatment (Fig. 2). Ninety-five per cent confidence interval for H. pylori eradication was from 23.1% to 68.5%. Mean CA in subjects who became H. pylori-negative was 10.5 (95% CI, 6.6-14.4) before treatment and 0.37 (95% CI, 0.20-0.54) 4 weeks after cessation. In the 11 subjects who remained H . pylori-positive after treatment mean CA was 10.1 (95% CI, 7.9-12.3) before treatment and 6.8 (95% CI, 4.1-9.6) 4 weeks after cessation. Mean reduction in CA in these 11subjects was 3.3 (95% CI, 0.38-6.2) ( p = 0.03), which was a 32.4% reduction in gastric urease activity. Adverse effects. All self-registration questionnaires were returned as agreed (Table I). Of the 20 subjects 6 (30%) had one or more adverse effects that were regarded as moderate or severe. The commonest adverse effects regarded as moderate or severe were loose stools (n = 4) and headache ( n = 3). The three (15%) subjects who noted ‘taste of metal’ regarded it as mild. Seven subjects reported no adverse effects. None of the subjects withdrew owing to adverse effects. There was no difference in adverse effects between subjects in whom H . pylori had been eradicated and those in whom it had not 4 weeks after cessation. Patient compliance. Five subjects (25%) were considered to have inadequate compliance. F~~~ of these subjects returned between 5% and lo%, and the fifth returned 20.1% of the tablets he should have taken.
Antacids and H. pylori Table I. Adverse effects in 20 subjects infected with Helicobacter pylori treated with aluminium-containing antacids, oxytetracycline, and metronidazole Degree of adverse effects* None
Mild
Moderate
Severe
14 13 19 16 15 18 12
4
1 2 0 0 1 1 1
1 2 0 1 2 0 1
~~
Scand J Gastroenterol Downloaded from informahealthcare.com by Queen's University on 12/30/14 For personal use only.
Nausea/vomiting Loose stools Constipation Stomach pain/discomfort Headache Skin rash Otherst
3 1 3 2
1 6
* Mild = does not limit daily activities; moderate = limits to some extent daily activities; and severe = makes daily activities almost impossible. t Includes taste of metal, 3; oral candidiasis, 1; facial itching, 1; gingivitis, 1; flatulence and xerostomia, 1; and fatigue, 1.
DISCUSSION Study I This is the first report of prompt effect of antacids on gastric urease activity. One tablet of antacids reduced the urease activity in subjects infected with H. pylori by 33.3% within 40 min. The result is interesting because urease inhibitors may reduce mucosal damage (15), and inhibition of this enzyme may have therapeutic potential (10). In gastric biopsy specimens we found a reduced H. pylori colonisation after 4 weeks’ treatment with one antacid tablet four times daily (4). After 2 weeks’ treatment with the same regimen gastric urease activity was suppressed 12 h after the last intake of antacids (5). The latter results indicated a suppressive effect beyond the effect of antacids on pH and could be due to reduced colonisation of H. pylori. Owing to the short time elapsed, the prompt suppressive effect on H . pylori urease demonstrated in the present study is probably not due to hampered H. pylori colonization. Whether the prompt suppressive effect is a consequence of acid neutralisation is unknown. The three experiments may indicate that aluminium-containing antacids promote a dual effect on H. pylori: one that reduces H. pylori colonisation (4,5) and another that inhibits H. pylori urease activity (5, this study). In accordance with others (23-25), treatment with antacids alone did not eradicate H. pylori infection in any of our studies. Theoretically, a subject could have been classified as H. pylori-positive owing to a spuriously high initial value, and a lower urease activity at reexamination could have been due to regression towards the mean. Of currently available diagnostic tests for detection of H. pylori infection the urea breath test is recognized as the best (26), and the test is recommended for use in research set-ups to generate more accurate results (27). At present the urea breath tests are the best documented tests for quantitative evaluation of H. pylori urease activity in vivo. The urea breath test dis-
1009
tinguishes well between two populations-one with an increased gastric urease activity and one without (21). Evans et al. (28) found that the urea breath test was concordant in 99.3% with their anti-H. pylori urease (high molecular weight cell-associated proteins) IgG enzyme-linked immunosorbent assay serologic test and that neither test produced false-positive results. Secondly, none of the subjects had their H. pylori status altered by the treatment. Finally, all subjects in this first study (both positive and negative) were treated, and the overall result indicated a reduction similar to that seen in H . pylori-positive subjects. Hence, our result in infected subjects is not likely to be due to regression toward the mean. The increase, although minor, in gastric urease activity in negative subjects is not due to an effect on H. pylori urease, since these subjects were not infected with H. pylori. Thus, besides suppression of H. pylori urease, antacids may actually cause a minor increase in non-H. pylori-related gastric urease. The explanation for this latter effect is uncertain. Our material confirmed the notion that H. pylori-positive subjects are, on average, older than negative subjects (29). Study 2 This is the first report of the effect of antacids in combination with antibiotics in the treatment of H . pylori infection. As assessed by the 14C-urea breath test, the regimen eradicated the infection in 45% of the subjects. In a recent study we confirmed the well-documented effect of bismuth salt in combination with tetracycline and metronidazole in H . pylori eradication and treatment of peptic ulcer disease (30); we achieved eradication rates of 89% and 97% in two groups (31,32). The present study indicates that replacing bismuth with aluminium-containing antacids yields poorer H. pylori eradication than a similar regimen with bismuth (odds ratio, 13.9; 95% CI, 4.8-40.0). The lower rate of H . pylori eradication with the present regimen was probably not due to inadequate patient compliance. Previously, we found that 2 weeks’ treatment with antacids suppressed gastric urease activity 12 h after the last intake of antacids, whereas no effect was seen 2 weeks after cessation of treatment (5). In the present study gastric urease activity in subjects who did not become H. pylori-negative was suppressed by 32.4% 4 weeks after cessation of treatment. Hence, there appears to be a prolonged suppressive effect of the present regimen beyond the effect of antacids alone. To our knowledge, results showing H. pylori eradication after tetracycline and metronidazole without a third remedy have not been published. In one study doxycycline and tinidazole were given for 1 week to 11 H. pylori-infected subjects, 2 of whom (18%) became H. pylori-negative (33). The latter study differed not only in the medication given but also in diagnostic procedure and population investigated. Direct comparison with the present study may not be proper, but it may indicate an additive effect of antacids to the effect
1010
K . Berstad
el
al.
Scand J Gastroenterol Downloaded from informahealthcare.com by Queen's University on 12/30/14 For personal use only.
of oxytetracycline and metronidazole. Whether higher doses of antacids, concomitant administration of antacids and antibiotics, or other dose schedules might improve the results is not known. Thirty per cent of the subjects experienced adverse effects regarded as moderate or severe. In two groups treated with bismuth subnitrate, oxytetracycline, and metronidazole we observed moderate or severe adverse effects in approximately 51% (31, 3 2 ) . Hence, adverse effects did not seem to differ substantially between the two triple regimens (odds ratio, 2.4; 95% CI, 0.9-6.7). Conclusion The prompt suppressive effect of aluminium-containing antacids on gastric urease activity as indicated by the first study may be a mechanism by which antacids exert their clinical effect in peptic ulcer disease. The second study suggests that a low dose of antacids as a bismuth substitute in H . pylori eradication is inferior to the triple treatment already established. Considering the consequences of possible induction of H . pylori resistance to antibiotics, treatment should be aimed at a high rate of eradication, even though it may imply some adverse effects. REFERENCES 1. Weberg R, Berstad A, Lange 0, Schulz T, Aubert E. Duodenal ulcer healing with four antacid tablets daily. Scand J Gastroenterol 1985;20:1041-5. 2. Rydning A, Weberg R, Lange 0, Berstad A. Healing of benign gastric ulcer with low-dose antacids and fiber diet. Gastroenterology 1986;91:5661. 3 . Hirschl A, Stanek G, Potzi R, Rotter M. Wende L. Die Empfindlichkeit von Campylobacter pyloridis gegeniiber antimikrobiellen Chemotherapeutika und Ulcusterapeutika. Z Antimicrob Antineopl Chemother 1986;4:4S-Y. 4. Berstad A, Alexander B, Weberg R, Serck-Hansen A, Hollander S, Hirschowitz BI. Antacids reduce Campylobacter pylori colonization without healing the gastritis in patients with nonulcer dyspepsia and erosive prepyloric changes. Gastroenterology 1988;95:619-24, 5. Berstad K, Weberg R, Berstad A. Suppression of gastric urease activity by antacids. Scand J Gastroenterol 1990;25:496500. 6 . Tytgat GN, Graham DY, Lee A, Marshall BJ, Dixon MF, Axon ATR. Helicobacter pylori: causal agent in peptic ulcer disease'? J Gastroenterol Hepatol 1991;6:103-40. 7. Marshall BJ, McGechie DB, Roger PA, Glancy RJ. Pyloric campylobacter infection and gastroduodenal disease. Med J Aust 1985;142:439-44. 8. Mobley HLT, Cortesia MJ, Rosenthal LE, Jones BD. Characterization of urease from campylobacter pylori. J Clin Microbiol 1988;26:831-6. 9. Blaser MJ, Cover TL, Steele R, Eaton K, Perez-Perez G, Labigne A. Characteristics of a urease-negative H pylori mutant strain. Rev Esp Enf Digest 1990;78 Suppl 1;26, P-40B. 10. Smoot DT, Mobley HLT, Chippendale GR, Lewison JF, Resau JH. Helicobacter pylori urease activity is toxic to human gastric epithelial cells. Infect Immun 1990;58:1992-4. 11. Barer MR, Elliott TSJ, Berkeley D, Thomas JE, Eastham EJ. Received 3 March 1992 Accepted 17 June 1992
Cytopathic effects of Campylobacter pylori urease [letter]. J Clin Pathol 1988;41:597. 12. Xu J-K, Goodwin S , Cooper M, Robinson J . Intracellular vacuolization caused by the urease of Helicobacter pylori. J Infect Dis 1990;161:1302-4. 13. Murakami M, Yo0 JK, Teramura S, et al. Generation of ammonia and mucosal lesion formation following hydrolysis of urea by urease in the rat stomach. J Clin Gastroenterol 1990;12 Suppl l:S104-9. 4. Marshall BJ, Barret LJ, Prakash C, McCallum RW, Guerrant RL. Urea protects Helicobacter pylori from the bactericidal effect of acid. Gastroenterology 1990;99:687-702. 5. Mooney C, Munster DJ, Bagshaw PF, Allardyce RA. Helicobacter pylori acid resistance [letter]. Lancet 1990;335:1232. 6. Borody T. Cole P, Noonan S, et al. Long-term Campylobacter pylori recurrence post-eradication. Gastroenterology 1988;94: A43. 17. Borody T. Lcnnc J , Moore-Jones D, et al. Is doxycycline more effective than tetracycline HCI in triple therapy of Helicobacter pylori? Gastroenterology 1990;98:A24. 18. McNulty CAM, Eyre-Brook IA, Uff IS, Dent IC, Wilkinson SP. Triple therapy is not always 95% effective. In: The Vth International Workshop on Campylobacter infections, February 1989, Puerto Vallarta, Mexico. 19. Truesdale RA, Chamberlain CE, Martin DF, et al. Long-term follow-up and antibody response to treatment of patients with Helicobacter pylori. Gastroenterology 1990;98:A140. 20. Bradley B, Singleton M. Li Wan Po A. Review article. Bismuth toxicity: a reassessment. J Clin Pharm Ther 1989;14:423-41. 21. Berstad K, Wilhelmsen I, Berstad A. Biometric evaluation of gastric urease activity in man. Scand J Gastroenterol 1992; 27 :977-83. 22. Gardner MJ, Altman DG, editors. Statistics with confidence. Confidence intervals and statistical guidelines. 3rd rev ed. London: British Medical Journal, 1990. 23. Graham DY, Klein PD, Opekun AR, et al. In vivo susceptibility of Campylobacter pylori. Am J Gastroenterol 1989;84:233-8. 24. Hirschl AM, Hentschel E, Schiitze K, c t al. The efficacy of antimicrobial treatment in Cumpylohucter-pylor~'-associatedgastritis and duodenal ulcer. Scand J Gastroenterol 1988;23 Suppl 142:7&81. 25. El Nujumi AM, Dorrian CA, Chittajallu RS, Neithercut WD, McColl KEL. Effect of inhibition of Helicobacter pylori urease activity by acetohydroxamic acid on serum gastrin in duodenal ulcer subjects. Gut 1991;32:86670. 26. Veldhuyzen van Zanten SJO, Tytgat KMAJ, Hollingsworth J , et al. "C-urea breath test for the detection of Helicobacter pylori. Am J Gastroenterol 1990;85:399-403. 27. Ching CK. A simplified diagnostic approach to Helicobacter pylori infection-is it adequate? [letter]. Am J Gastroenterol 1991;86:1276. 28. Evans DJ, Evans DG, Graham DY, Klein PD. A sensitive and specific serologic test for detection of campylobacter pylori infection. Gastroenterology 1989;96:1004-8. 29. Kosunen TU, Hook J , Rautelin HI, Myllyla G. Age-dependent increase of Campylobacter pylori antibodies in blood donors. Scand J Gastroenterol 1989;24:110-4. 30. Chiba N , Rademaker JW, Rao BV, Hunt RH. Eradication of Helicobacter pylori-meta-analysis to determine optimal therapy. Gut 1991;32:A1220-1. 31. Wilhelmsen I, Berstad A. Duodenal ulcer recurrence after cognitive therapy and after Helicohacrer pylori eradication. Scand J Gastroenterol 1992;27 Suppl 190:A137. 32. Weberg R , Bang C. Berstad K, et al. Treatment of Helicobacter pylori in peptic ulcer disease. Scand J Gastroenterol 1992;27 Suppl 190:A40. 33. Glupczynski Y, Bourdeaux L, Verhas M, et al. Short-term double or triple oral drug treatment of Helicobacter pylori (Hp) in Central Africa. Gastroenterology 1990;98:A48.
3
Scand J Gastroenterol Downloaded from informahealthcare.com by Queen's University on 12/30/14 For personal use only.
Berstad K, Weberg R, Berstad A . IS there a place for antacids in the treatment of Helicobacterpylori infection? Scand J Gastroenterol 1992;27:10061010. f i w o pilot studies were performed to determine whether aluminium-containing antacids may have a place in the treatment of Helicobacter pylori infection. The urease activity of H . pylori is cytopathic to gastric epithelium, and inhibition of this enzyme may have therapeutic potential. In the first study 24 subjects, 12 of which were infected with H . pylori, were given 1 tablet of chewable aluminium hydroxidecontaining antacids 10 min before a “C-urea breath test. Gastric urease activity was suppressed by 33.3% ( p = 0.02) in the H . pylori-positive subjects (none became negative) within 40 min after administration of the tablet. Gastric H . pylori infection can be effectivelyeradicated by triple regimens containing bismuth salts, tetracycline, and metronidazole. Owing to adverse effects of this treatment and concern for possible neurotoxicity of bismuth, a bismuth substitute is warranted. Hence, in the second study, 20 subjects infected with H . pylori were treated with 1 antacid tablet 4 times daily between meals, plus 500 mg oxytetracycline and 200 mg metronidazole 4 times daily with meals for 2 weeks. Individual H . pylori status was assessed by the “C-urea breath test. Four weeks after cessation of treatment, H. pylori was eradicated in 45% (9 of 20) of the subjects (95% confidence interval, 23.1-68.5%). Thirty per cent (6 of 20) observed one or more adverse effect regarded as moderate or severe, of which loose stools and headache were the most common. We conclude that the suppressive effect of antacids on gastric urease activity may be a mechanism by which aluminium-containing antacids promote peptic ulcer healing and that a low-dose regimen of antacids as a bismuth substitute in the triple treatment for H . pylori eradication is inferior to the triple treatment already established
J
Key words: Antacids; Helicobacter pylori; metronidazole; oxytetracycline; urease K . Berstad, M . D.,Dioision of Gastroenterology, Medical Dept. A , Haukeland University Hospital, N5021 Bergen, Norway
Antacids containing aluminium hydroxide promote healing of duodenal and gastric ulcers in doses that have a rather feeble effect on intragastric acidity (1,2). The mechanisms of action for aluminium-containing antacids beyond acid neutralization are by and large unknown. An in vitro study demonstrated that aluminium-containing antacids inhibit growth of Helicobacter pylori even at low concentrations (3). In previous studies we found reduced gastric H . pylori colonization in biopsy specimens after 4 weeks’ treatment with a low-dose regimen of aluminium hydroxide-containing antacids (4). Moreover, 2 weeks’ treatment with the same antacid regimen suppressed, but did not abolish, an enhanced gastric urease activity as assessed with the I4Curea breath test ( 5 ) . Gastric H . pylori infection is increasingly acknowledged as an important cause of gastritis and peptic ulcer disease (6,7). H . pylori urease, which has a high substrate affinity (8), is observed in all wild strains of H. pylori (9). The urease activity is toxic to human gastric epithelial cells (10, 11) and causes intracellular vacuolization (12). Ammonia generated from the hydrolysis of urea damages the gastric mucosa (13) and enables the bacteria to survive in gastric acid (14). Urease inhibitors may reduce mucosal damage and enhance
the host’s ability to clear the infection (15). Hence, inhibition of H . pylori urease activity may have therapeutic potential (10). To study the prompt effect of antacids on gastric urease activity, a I4C-urea breath test was commenced 10 min after administration of one antacid tablet in subjects with and without H . pylori infection (study 1). Treatment of H . pylori infection with bismuth salts in combination with tetracycline and metronidazole leads to 89-97% eradication of the bacteria ( 1 6 1 9 ) . However, adverse effects like malaise, nausea, and diarrhoea are common, and there is concern about the possible neurotoxicity of bismuth (20). In the second study an aluminium-containing antacid was given as a bismuth substitute together with oxytetracycline and metronidazole in a triple-treatment regimen to 20 subjects infected with H . pylori (study 2). MATERIALS AND METHODS
Subjects Study I . Twenty-four clinically healthy volunteers (15 men) with a mean age of 41.0 years (range, 26-60 years)
Antacids and H. pylori
Scand J Gastroenterol Downloaded from informahealthcare.com by Queen's University on 12/30/14 For personal use only.
were enrolled. All subjects were recruited among hospital employees. None of the participants had a history of peptic ulcer disease. Study 2. Twenty volunteers (8 men) with a mean age of 52 years (range, 35-76 years) were recruited among hospital employees (n = 9) known from previous studies to harbour the bacteria and among patients referred to the clinic ( n = 11). The referred patients were enrolled consecutively among patients in whom upper endoscopy did not reveal ulcer, oesophagitis, or malignancy, and in whom H . pylori infection was diagnosed by the I4C-urea breath test. Procedure and treatment Study 1. Each subject was studied twice after an overnight fast. Before the initial control breath test, 100ml of tap water was given 10 rnin before the base-line breath sample (see below). One to 3 days later the same procedure was followed by one tablet of chewable antacids along with the 100ml of water. The antacid tablet (Link@, ll00mg, A/S Apothekernes Laboratorium, Norway) consisted of aluminium hydroxide and magnesium carbonate in a co-dried gel, with an acid-buffering capacity of 30 mmol per tablet. The subjects were to abstain from tobacco on the mornings before the experiments. Study 2 . Treatment was commenced at an average 19 days (SD, 9; maximum, 38 days) after the initial breath test. The second breath test was performed 4 weeks after cessation of treatment. The subjects were to abstain from tobacco on the mornings before the tests. The subjects were given 500 mg oxytetracycline (Oxytetral@,250-mg tablets, A/S Apothekernes Laboratorium) plus 200 mg metronidazole (FlagylB, 200-mg tablets, RhGnePoulenc Rorer) four times daily at meals and one antacid tablet (Link, 1100 mg) four times daily between meals for 2 weeks. To monitor adverse effects, the subjects were to fill in a self-registration questionnaire handed out at the beginning of treatment and returned at control 6 weeks later. Patient compliance was assessed by the number of returned tablets. No subjects were excluded owing to inadequate compliance. Assessment of H. pylori status with the 14C-ureabreath test 14C-labelled urea, containing 92.5 kBq (2.5 pCi), was given orally with 50 ml of water (0.96 pM) immediately after completion of the base-line breath sample. Double breath samples were obtained every 10th rnin for 30 min. I4C activities of samples and standards were counted in a p-counter. Cumulative expired I4C activity (CA) was calculated in accordance with the trapezoid method with adjustment for assumed individual C 0 2production (9 mmol CO,/kg/h) and body weight. CA is thus an estimate of gastric urease activity, and its unit is percentage recovery of I4C during the first 30 rnin after I4C-urea administration (% recovery 30 min). Biometric evaluation of human gastric urease activity has indicated a cut-off value of 1.47% recovery 30 rnin in random
1007
t
Control
10-40 rnin
a f t e r Antacids Fig. 1. Mean (horizontal bar) and individual gastric urease activities in 12 clinically healthy subjects infected with Heficubacter pylori. Control and 10-40min after one tablet of antacids ( p = 0.02).
subjects (21). Hence, subjects with a CA below 1.47 were regarded as not infected with H . pylori ( H . pylori-negative), whereas those with a CA above 1.47 were regarded as infected with the bacteria ( H . pylori-positive). Ethics Study 1. Informed consent was obtained from each participant. The use of the I4C-urea breath test was approved by the National Institute of Radiation Hygiene, Norway. Study 2 . Written informed consent was obtained from each subject. The study was approved by the Ethics Review Committee (Region 3) and the Norwegian Medicines Control Authority. The study was conducted in accordance with the Good Clinical Practice guidelines issued by the EEC (1990). Statitistics Normality was assessed with the Wilk-Shapiro W test. Student’s t test for paired sample and the Sign test were applied to the effect of treatments (two-tailed). Correlation was assessed by Spearman rank. Figures are given as arith-
1008
K. Berstad ef al.
metic mean and 95% confidence interval (95% CI). P < 0.05 was regarded as statistically significant. Study 1. Effect of antacids on gastric urease activity was calculated as the difference between CA with antacids and control. Study 2 . Relative frequencies of H. pylori status after treatment were assessed in accordance with the intention to treat. Confidence interval for proportion was calculated with Confidence Interval Analysis (22), using the exact method.
Scand J Gastroenterol Downloaded from informahealthcare.com by Queen's University on 12/30/14 For personal use only.
RESULTS Study I There was no significant correlation between CA and weight or age within the two groups in accordance with H . pylori status. CA ranged from 1.63 to 12.8 among the 12 H . pylori-positive (6 male) subjects and from 0.21 to 1.14 among the 12 negative (9 male) subjects. Mean ages were 47.1 years (95% CI, 41 3-52.4) and 34.8 years (95% CI, 30.1-39.6) for the H. pylori-positive and -negative subjects, respectively
...............................
Before Treat ment
4 Weeks after Cessation
Fig. 2. Individual gastric urease activity before and 4 weeks after cessation of treatment with antacids, oxytetracycline, and metronidazole in 20 subjects infected with Helicobacter pylori. Cut-off value at 1.47 indicated with a dotted line.
( p = 0.001). The overall mean relative effect of antacids on gastric urease activity was a reduction in CA by 2Y.8% ( p = 0.035).
Effect of antacids in H. pylori-positive subjects. Compared with control level, gastric urease activity was suppressed in 10 of the 12 H pylori-positive subjects after antacids ( p = 0.039). Mean control CA was 7.36 (95% CI, 5.08-9.66) and 4.90 (95% CI, 3.87-5.94) with antacids. Mean effect of antacids was 2.45 (95% CI, 0.4W.44) ( p = 0.020), which was a 33.3% reduction in mean CA (Fig. 1). None of the H.pylori-positive had a CA less than 1.47 (cut-off value) after treatment. Effect of antacids in H. pylori-negatiue subjects. Compared with control level, gastric urease activity was increased in 11 of the 12 H. pylori-negative subjects after antacids ( p = 0.006). Mean control CA was 0.41 (95% CI, 0.26-0.57) and 0.55 (95% CI, 404.70) with antacids. Mean effect of antacids was -0.13 (95% CI, -0.24 to -0.03) ( p = 0.016), which was a 31.7% increase in mean CA. None of the H. pylori-negative had a CA above 1.47 (cut-off value) after treatment.
Study 2 Effect of treatment on H. pylori infection. Overall mean CA before treatment was 10.3% recovery 30min (range, 2.8-17.0). When a cut-off value of 1.47% recovery 30 min was used, 9 of the 20 H . pylori-positive subjects (45%) were negative (range, 0.12-0.90) 4 weeks after cessation of treatment (Fig. 2). Ninety-five per cent confidence interval for H. pylori eradication was from 23.1% to 68.5%. Mean CA in subjects who became H. pylori-negative was 10.5 (95% CI, 6.6-14.4) before treatment and 0.37 (95% CI, 0.20-0.54) 4 weeks after cessation. In the 11 subjects who remained H . pylori-positive after treatment mean CA was 10.1 (95% CI, 7.9-12.3) before treatment and 6.8 (95% CI, 4.1-9.6) 4 weeks after cessation. Mean reduction in CA in these 11subjects was 3.3 (95% CI, 0.38-6.2) ( p = 0.03), which was a 32.4% reduction in gastric urease activity. Adverse effects. All self-registration questionnaires were returned as agreed (Table I). Of the 20 subjects 6 (30%) had one or more adverse effects that were regarded as moderate or severe. The commonest adverse effects regarded as moderate or severe were loose stools (n = 4) and headache ( n = 3). The three (15%) subjects who noted ‘taste of metal’ regarded it as mild. Seven subjects reported no adverse effects. None of the subjects withdrew owing to adverse effects. There was no difference in adverse effects between subjects in whom H . pylori had been eradicated and those in whom it had not 4 weeks after cessation. Patient compliance. Five subjects (25%) were considered to have inadequate compliance. F~~~ of these subjects returned between 5% and lo%, and the fifth returned 20.1% of the tablets he should have taken.
Antacids and H. pylori Table I. Adverse effects in 20 subjects infected with Helicobacter pylori treated with aluminium-containing antacids, oxytetracycline, and metronidazole Degree of adverse effects* None
Mild
Moderate
Severe
14 13 19 16 15 18 12
4
1 2 0 0 1 1 1
1 2 0 1 2 0 1
~~
Scand J Gastroenterol Downloaded from informahealthcare.com by Queen's University on 12/30/14 For personal use only.
Nausea/vomiting Loose stools Constipation Stomach pain/discomfort Headache Skin rash Otherst
3 1 3 2
1 6
* Mild = does not limit daily activities; moderate = limits to some extent daily activities; and severe = makes daily activities almost impossible. t Includes taste of metal, 3; oral candidiasis, 1; facial itching, 1; gingivitis, 1; flatulence and xerostomia, 1; and fatigue, 1.
DISCUSSION Study I This is the first report of prompt effect of antacids on gastric urease activity. One tablet of antacids reduced the urease activity in subjects infected with H. pylori by 33.3% within 40 min. The result is interesting because urease inhibitors may reduce mucosal damage (15), and inhibition of this enzyme may have therapeutic potential (10). In gastric biopsy specimens we found a reduced H. pylori colonisation after 4 weeks’ treatment with one antacid tablet four times daily (4). After 2 weeks’ treatment with the same regimen gastric urease activity was suppressed 12 h after the last intake of antacids (5). The latter results indicated a suppressive effect beyond the effect of antacids on pH and could be due to reduced colonisation of H. pylori. Owing to the short time elapsed, the prompt suppressive effect on H . pylori urease demonstrated in the present study is probably not due to hampered H. pylori colonization. Whether the prompt suppressive effect is a consequence of acid neutralisation is unknown. The three experiments may indicate that aluminium-containing antacids promote a dual effect on H. pylori: one that reduces H. pylori colonisation (4,5) and another that inhibits H. pylori urease activity (5, this study). In accordance with others (23-25), treatment with antacids alone did not eradicate H. pylori infection in any of our studies. Theoretically, a subject could have been classified as H. pylori-positive owing to a spuriously high initial value, and a lower urease activity at reexamination could have been due to regression towards the mean. Of currently available diagnostic tests for detection of H. pylori infection the urea breath test is recognized as the best (26), and the test is recommended for use in research set-ups to generate more accurate results (27). At present the urea breath tests are the best documented tests for quantitative evaluation of H. pylori urease activity in vivo. The urea breath test dis-
1009
tinguishes well between two populations-one with an increased gastric urease activity and one without (21). Evans et al. (28) found that the urea breath test was concordant in 99.3% with their anti-H. pylori urease (high molecular weight cell-associated proteins) IgG enzyme-linked immunosorbent assay serologic test and that neither test produced false-positive results. Secondly, none of the subjects had their H. pylori status altered by the treatment. Finally, all subjects in this first study (both positive and negative) were treated, and the overall result indicated a reduction similar to that seen in H . pylori-positive subjects. Hence, our result in infected subjects is not likely to be due to regression toward the mean. The increase, although minor, in gastric urease activity in negative subjects is not due to an effect on H. pylori urease, since these subjects were not infected with H. pylori. Thus, besides suppression of H. pylori urease, antacids may actually cause a minor increase in non-H. pylori-related gastric urease. The explanation for this latter effect is uncertain. Our material confirmed the notion that H. pylori-positive subjects are, on average, older than negative subjects (29). Study 2 This is the first report of the effect of antacids in combination with antibiotics in the treatment of H . pylori infection. As assessed by the 14C-urea breath test, the regimen eradicated the infection in 45% of the subjects. In a recent study we confirmed the well-documented effect of bismuth salt in combination with tetracycline and metronidazole in H . pylori eradication and treatment of peptic ulcer disease (30); we achieved eradication rates of 89% and 97% in two groups (31,32). The present study indicates that replacing bismuth with aluminium-containing antacids yields poorer H. pylori eradication than a similar regimen with bismuth (odds ratio, 13.9; 95% CI, 4.8-40.0). The lower rate of H . pylori eradication with the present regimen was probably not due to inadequate patient compliance. Previously, we found that 2 weeks’ treatment with antacids suppressed gastric urease activity 12 h after the last intake of antacids, whereas no effect was seen 2 weeks after cessation of treatment (5). In the present study gastric urease activity in subjects who did not become H. pylori-negative was suppressed by 32.4% 4 weeks after cessation of treatment. Hence, there appears to be a prolonged suppressive effect of the present regimen beyond the effect of antacids alone. To our knowledge, results showing H. pylori eradication after tetracycline and metronidazole without a third remedy have not been published. In one study doxycycline and tinidazole were given for 1 week to 11 H. pylori-infected subjects, 2 of whom (18%) became H. pylori-negative (33). The latter study differed not only in the medication given but also in diagnostic procedure and population investigated. Direct comparison with the present study may not be proper, but it may indicate an additive effect of antacids to the effect
1010
K . Berstad
el
al.
Scand J Gastroenterol Downloaded from informahealthcare.com by Queen's University on 12/30/14 For personal use only.
of oxytetracycline and metronidazole. Whether higher doses of antacids, concomitant administration of antacids and antibiotics, or other dose schedules might improve the results is not known. Thirty per cent of the subjects experienced adverse effects regarded as moderate or severe. In two groups treated with bismuth subnitrate, oxytetracycline, and metronidazole we observed moderate or severe adverse effects in approximately 51% (31, 3 2 ) . Hence, adverse effects did not seem to differ substantially between the two triple regimens (odds ratio, 2.4; 95% CI, 0.9-6.7). Conclusion The prompt suppressive effect of aluminium-containing antacids on gastric urease activity as indicated by the first study may be a mechanism by which antacids exert their clinical effect in peptic ulcer disease. The second study suggests that a low dose of antacids as a bismuth substitute in H . pylori eradication is inferior to the triple treatment already established. Considering the consequences of possible induction of H . pylori resistance to antibiotics, treatment should be aimed at a high rate of eradication, even though it may imply some adverse effects. REFERENCES 1. Weberg R, Berstad A, Lange 0, Schulz T, Aubert E. Duodenal ulcer healing with four antacid tablets daily. Scand J Gastroenterol 1985;20:1041-5. 2. Rydning A, Weberg R, Lange 0, Berstad A. Healing of benign gastric ulcer with low-dose antacids and fiber diet. Gastroenterology 1986;91:5661. 3 . Hirschl A, Stanek G, Potzi R, Rotter M. Wende L. Die Empfindlichkeit von Campylobacter pyloridis gegeniiber antimikrobiellen Chemotherapeutika und Ulcusterapeutika. Z Antimicrob Antineopl Chemother 1986;4:4S-Y. 4. Berstad A, Alexander B, Weberg R, Serck-Hansen A, Hollander S, Hirschowitz BI. Antacids reduce Campylobacter pylori colonization without healing the gastritis in patients with nonulcer dyspepsia and erosive prepyloric changes. Gastroenterology 1988;95:619-24, 5. Berstad K, Weberg R, Berstad A. Suppression of gastric urease activity by antacids. Scand J Gastroenterol 1990;25:496500. 6 . Tytgat GN, Graham DY, Lee A, Marshall BJ, Dixon MF, Axon ATR. Helicobacter pylori: causal agent in peptic ulcer disease'? J Gastroenterol Hepatol 1991;6:103-40. 7. Marshall BJ, McGechie DB, Roger PA, Glancy RJ. Pyloric campylobacter infection and gastroduodenal disease. Med J Aust 1985;142:439-44. 8. Mobley HLT, Cortesia MJ, Rosenthal LE, Jones BD. Characterization of urease from campylobacter pylori. J Clin Microbiol 1988;26:831-6. 9. Blaser MJ, Cover TL, Steele R, Eaton K, Perez-Perez G, Labigne A. Characteristics of a urease-negative H pylori mutant strain. Rev Esp Enf Digest 1990;78 Suppl 1;26, P-40B. 10. Smoot DT, Mobley HLT, Chippendale GR, Lewison JF, Resau JH. Helicobacter pylori urease activity is toxic to human gastric epithelial cells. Infect Immun 1990;58:1992-4. 11. Barer MR, Elliott TSJ, Berkeley D, Thomas JE, Eastham EJ. Received 3 March 1992 Accepted 17 June 1992
Cytopathic effects of Campylobacter pylori urease [letter]. J Clin Pathol 1988;41:597. 12. Xu J-K, Goodwin S , Cooper M, Robinson J . Intracellular vacuolization caused by the urease of Helicobacter pylori. J Infect Dis 1990;161:1302-4. 13. Murakami M, Yo0 JK, Teramura S, et al. Generation of ammonia and mucosal lesion formation following hydrolysis of urea by urease in the rat stomach. J Clin Gastroenterol 1990;12 Suppl l:S104-9. 4. Marshall BJ, Barret LJ, Prakash C, McCallum RW, Guerrant RL. Urea protects Helicobacter pylori from the bactericidal effect of acid. Gastroenterology 1990;99:687-702. 5. Mooney C, Munster DJ, Bagshaw PF, Allardyce RA. Helicobacter pylori acid resistance [letter]. Lancet 1990;335:1232. 6. Borody T. Cole P, Noonan S, et al. Long-term Campylobacter pylori recurrence post-eradication. Gastroenterology 1988;94: A43. 17. Borody T. Lcnnc J , Moore-Jones D, et al. Is doxycycline more effective than tetracycline HCI in triple therapy of Helicobacter pylori? Gastroenterology 1990;98:A24. 18. McNulty CAM, Eyre-Brook IA, Uff IS, Dent IC, Wilkinson SP. Triple therapy is not always 95% effective. In: The Vth International Workshop on Campylobacter infections, February 1989, Puerto Vallarta, Mexico. 19. Truesdale RA, Chamberlain CE, Martin DF, et al. Long-term follow-up and antibody response to treatment of patients with Helicobacter pylori. Gastroenterology 1990;98:A140. 20. Bradley B, Singleton M. Li Wan Po A. Review article. Bismuth toxicity: a reassessment. J Clin Pharm Ther 1989;14:423-41. 21. Berstad K, Wilhelmsen I, Berstad A. Biometric evaluation of gastric urease activity in man. Scand J Gastroenterol 1992; 27 :977-83. 22. Gardner MJ, Altman DG, editors. Statistics with confidence. Confidence intervals and statistical guidelines. 3rd rev ed. London: British Medical Journal, 1990. 23. Graham DY, Klein PD, Opekun AR, et al. In vivo susceptibility of Campylobacter pylori. Am J Gastroenterol 1989;84:233-8. 24. Hirschl AM, Hentschel E, Schiitze K, c t al. The efficacy of antimicrobial treatment in Cumpylohucter-pylor~'-associatedgastritis and duodenal ulcer. Scand J Gastroenterol 1988;23 Suppl 142:7&81. 25. El Nujumi AM, Dorrian CA, Chittajallu RS, Neithercut WD, McColl KEL. Effect of inhibition of Helicobacter pylori urease activity by acetohydroxamic acid on serum gastrin in duodenal ulcer subjects. Gut 1991;32:86670. 26. Veldhuyzen van Zanten SJO, Tytgat KMAJ, Hollingsworth J , et al. "C-urea breath test for the detection of Helicobacter pylori. Am J Gastroenterol 1990;85:399-403. 27. Ching CK. A simplified diagnostic approach to Helicobacter pylori infection-is it adequate? [letter]. Am J Gastroenterol 1991;86:1276. 28. Evans DJ, Evans DG, Graham DY, Klein PD. A sensitive and specific serologic test for detection of campylobacter pylori infection. Gastroenterology 1989;96:1004-8. 29. Kosunen TU, Hook J , Rautelin HI, Myllyla G. Age-dependent increase of Campylobacter pylori antibodies in blood donors. Scand J Gastroenterol 1989;24:110-4. 30. Chiba N , Rademaker JW, Rao BV, Hunt RH. Eradication of Helicobacter pylori-meta-analysis to determine optimal therapy. Gut 1991;32:A1220-1. 31. Wilhelmsen I, Berstad A. Duodenal ulcer recurrence after cognitive therapy and after Helicohacrer pylori eradication. Scand J Gastroenterol 1992;27 Suppl 190:A137. 32. Weberg R , Bang C. Berstad K, et al. Treatment of Helicobacter pylori in peptic ulcer disease. Scand J Gastroenterol 1992;27 Suppl 190:A40. 33. Glupczynski Y, Bourdeaux L, Verhas M, et al. Short-term double or triple oral drug treatment of Helicobacter pylori (Hp) in Central Africa. Gastroenterology 1990;98:A48.
