clarithromycin

THE AMERICAN JOURNAL OF GASTROENTEROLOGY © 2001 by Am. Coll. of Gastroenterology Published by Elsevier Science Inc.

Vol. 96, No. 5, 2001 ISSN 0002-9270/01/$20.00 PII S0002-9270(01)02334-6

Reduced Symptoms and Need for Antisecretory Therapy in Veterans 3 Years After Helicobacter pylori Eradication With Ranitidine Bismuth Citrate/Amoxicillin/Clarithromycin S. J. Sontag, S. O’Connell, T. Schnell, G. Chejfec, J. Seidel, and A. Sonnenberg Departments of Medicine and Pathology, Veterans Affairs Hospital, Hines, Illinois; Departments of Medicine and Pathology, Veterans Affairs Medical Center in Albuquerque, New Mexico; and Departments of Medicine and Pathology, Loyola University Stritch School of Medicine, Maywood, Illinois

OBJECTIVE: The most effective combination therapy to eradicate Helicobacter pylori has not yet been found. The perfect combination would be effective, relatively free of side effects, and easy to comply with. We studied a 14-day course of three medications taken twice daily by H. pylori– infected patients who were enrolled in the outpatient Veterans Affairs (VA) clinics. The two major objectives were 1) to determine the effectiveness of the combination therapy and 2) to determine the compliance of patients in a VA population. METHODS: Fifty-two male patients were identified with H. pylori infection by positive CLO (Rapid Urease Test) test, positive Giemsa stain, or positive serology. Active infection was confirmed by a positive 13C urea breath test (UBT). Patients were treated for 14 days with open-label triplecombination therapy of ranitidine bismuth citrate (RBC; 400 mg b.i.d.), amoxicillin (1000 mg b.i.d.), and clarithromycin (500 mg b.i.d.). Successful eradication of H. pylori was confirmed by repeat UBT at 6 – 8 wk after the final dose of therapy. RESULTS: Of the 52 enrolled patients, 49 (94.2%) met the criteria for successful completion of the study (per protocol analysis based on compliance with at least 80% of medication and performance of both UBTs). Of the three patients who did not successfully complete, one was cured (after 6 days of treatment), and two remained infected (after 3 days and 9 days of treatment). Of the 49 completed patients, 45 (91.8%) were cured, and four remained infected. Overall, regardless of compliance (intent-to-treat analysis), 46 of the 52 (88.4%) patients had documented cure of H. pylori infection as determined by the posttreatment UBT. By 3 yr after H. pylori eradication, two of 15 (13.3%) patients who were not on baseline medications had developed the need for antisecretory therapy, but 18 of 31 (58.1%) who were on baseline medications were able to stop therapy. Thus, at 3 yr, successful H. pylori eradication decreased the need for antisecretory therapy from 67.4% of the H. pylori–infected

population to 43% of the H. pylori– eradicated population. The effect of H. pylori eradication in improving symptoms at 3 yr was statistically significant in both the ulcer population and the nonulcer population. Adverse events were mild, and included diarrhea (26 patients), bad taste in mouth (24 patients), nausea/upset stomach (nine patients), and headache (two patients). The diarrhea was self-limiting in 25 of the 26 patients. Only two patients discontinued medication because of adverse events. CONCLUSION: The RBC/amoxicillin/clarithromycin combination was, in our VA population, an easily complied with, highly effective, and safe triple therapy with a 90% H. pylori eradication rate. Successful eradication of H. pylori leads to a dramatic decrease in upper-gut symptoms and decreased need for antisecretory therapy. (Am J Gastroenterol 2001; 96:1390 –1395. © 2001 by Am. Coll. of Gastroenterology)

INTRODUCTION Recent studies have clearly linked chronic Helicobacter pylori infection with gastritis, duodenal ulcer, gastric ulcer, and gastric cancer. Indeed, the World Health Organization has labeled H. pylori as a Class 1 carcinogen in the production of gastric cancer (1). Although debate continues regarding the necessity and wisdom of eradicating H. pylori in patients without ulcer disease and whether eradication of the H. pylori results in or exacerbates reflux symptoms, investigation continues into newer combination therapies. The eventual therapeutic combination needs to be not only effective and safe but easy to comply with. Although growing evidence suggests that the more effective therapies involve at least three different agents, administered for a minimum of 10 days (2, 3), therapies involving only two agents for less than 10 days continue to be studied. In the end, however, a treatment that is simple to take and is given for short duration will ultimately fail if it is not effective. With this in mind, we chose to study the overall compliance, tolerability, and effective-

AJG – May, 2001

3 Years After H. pylori Eradication

1391

Table 1. Characteristics and Indications for H. pylori Treatment According to Peptic Ulcer Status

Status Documented past or current peptic ulcer No documented past or current peptic ulcer Total, n (%)

Dyspepsia or Epigastric Pain

Anemia, Weight Loss, Other

Mean Age (yr)

Males (%)

17 (33)

68.5

100

2

12

1

1

1

35 (67)

64.3

100

7

9

5

9

5

52 (100)

65.7

100

9 (17)

Dysphagia

21 (40)

Reflux Symptoms

Barrett’s Follow-Up (No Symptoms)

Number (%)

6 (12)

10 (19)

6 (12)

Of the 35 patients with no past or current evidence of peptic ulcer disease, 21 (60%) received endoscopic exam for upper-gut symptoms. Of the 17 patients with documented past or current peptic ulcer, 15 (88%) received endoscopic exam for upper-gut symptoms.

ness of a full 2-wk course of an RBC-based triple-combination therapy in an H. pylori–infected population that, because of its lower socioeconomic status, is commonly thought to be less compliant than the general population. Thus, we studied patients who were enrolled in our outpatient VA clinics.

MATERIALS AND METHODS Patients Patients were eligible for inclusion if they were 18 yr of age or older and had current H. pylori gastritis. The presence or absence of ulcer disease was not inclusion criteria. Patients were excluded if they 1) had failed previous H. pylori eradication therapy; 2) had a history of hypersensitivity or allergic reaction to macrolides or penicillin compounds, H2 receptor antagonists, or bismuth-containing medications; 3) had received any blood products within the previous 12 wk; 4) were currently pregnant or breastfeeding; or 5) had any significant unstable condition, such as massive ascites, acute hepatitis, unstable angina, and so on. Procedures Three methods were used to identify potential candidates for the study: the rapid urease (CLO) test, the Giemsa stain, and serology. Patients were either newly diagnosed with a positive CLO test or Giemsa stain during routine endoscopic exam or recently diagnosed (within the previous few months) and not yet treated. A positive urea breath test (UBT) was needed to confirm the H. pylori infection. The UBT was performed after a 4-h fast as follows: a baseline breath sample was obtained by exhaling into a plastic bag. Thirty minutes after ingestion of a commercial nutritional supplement (pudding) mixed with 13C-Urea, breath samples were again collected in the plastic bag. The breath samples were processed by a central lab (Meretek). Patients who qualified for the study received a history, physical exam, and routine blood work at baseline and at 4 wk after the last treatment day. At baseline, patients were given a 14-day supply of open-label medications, which began within 10 days of the date of the UBT-confirmed H. pylori infection. All three medications were taken together at the same time: RBC, 400 mg (tablets) b.i.d.; clarithromycin, 500 mg (tablets) b.i.d.; and amoxicillin, 1000 mg (capsules) b.i.d.

To document H. pylori eradication, the UBT test was repeated 6 to 8 wk after the last day of therapy and after at least 6 wk of the last dose of proton pump inhibitors. A minimum of 1 h was spent by the study coordinator explaining to each patient the importance of compliance with the medications. Potential side effects were also explained, and patients were asked in advance to try to “suffer through” diarrhea that was mild to moderate. Compliance was encouraged by weekly calls from the study coordinator. In addition, patients were given an easy-access telephone number to call regarding any questions or any complaints. Compliance was determined by simple patient questioning. Patients were considered compliant with therapy if they took at least 80% of the medication and completed both UBTs. Adverse events were actively sought during the telephone contact at the end of each week by asking specific questions from a standard questionnaire regarding change in bowel habits, taste, or abdominal pain. The protocol was approved by the Hines VA internal review board. Statistical Methods Fisher’s exact test was used to compare the association between two categorical variables of 2 ⫻ 2 tables. In instances with more than two categories, the ␹2 test was used. Proportions were also compared using their 95% CI, calculated according to the binomial distribution. Sample size was calculated to detect a 20% advantage in H. pylori eradication rate with the study regimen, as compared with existing regimens. Assuming a 70% eradication rate for all existing regimens and a 90% eradication rate for the study regimen, 44 patients would be needed to avoid an alpha error of 5% and a beta error of 20%. Because of potential dropouts and protocol violations, 52 patients were enrolled.

RESULTS Fifty-two male patients, of mean age 65.7 yr, entered the study. Table 1 shows the characteristics and indications for H. pylori treatment of the 52 patients according to peptic ulcer status. One third of the patients had a documented past or current peptic ulcer, and two thirds had no history of ulcer

1392

Sontag et al.

AJG – Vol. 96, No. 5, 2001

Table 2. Cure Rates and Confidence Intervals for the Intent-toTreat and Per-Protocol Analyses

Analysis

Total Number Enrolled

N

n/N

Cure Rate (%)

95% CI

Intent to treat Per protocol

52 52

52 49

46/52 45/49

88.5 91.8

79.8–97.2 83.8–99.5

N ⫽ number of patients for analysis; n ⫽ number of patients with successful H. pylori eradication; cure rate ⫽ percentage of patients with successful H. pylori eradication.

disease. Table 2 shows the cure rates and 95% CIs for both the intent-to-treat and the per-protocol analyses. For the intent-to-treat analysis (which disregards compliance and considers all 52 patients who began therapy as failures until proven to be successes), 46 of 52 (88.5%) patients were cured of H. pylori infection as determined by the posttreatment UBT (CI ⫽ 79.8 –97.2). For the per-protocol analysis (which includes only the 49 patients who took at least 80%— or 12 days— of the medication), 45 of 49 (91.8%) patients were cured of H. pylori infection (CI ⫽ 83.8 –99.5). Table 3 shows the adverse events, compliance, and evaluability of all 52 patients. Adverse events were numerous but mild and self-limiting. One evaluable patient stopped therapy after 12 days (successful completion) because of increasing levels of liver enzymes, which resolved after 4 days (H. pylori eradicated). Three patients took therapy for less than 12 days: one stopped therapy on day 3 for emergency coronary artery bypass surgery (H. pylori infection still present), one was noncompliant and took only 6 days of therapy (H. pylori eradicated), and one stopped after 9 days of therapy because of an amoxicillin-induced rash, which resolved after 24 h (H. pylori infection still present). Thus, only one patient was withdrawn from the study because of adverse events: the amoxicillin-induced rash. The remaining adverse events were mild and included diarrhea (26 patients), bad taste in mouth (24 patients), nausea/upset stomach (nine patients), dry mouth (three patients), and headache (two patients). Antidiarrheal therapy was required in only

one of the 26 patients, whereas the diarrhea was self-limiting in the remaining 25 patients. Overall compliance rates were high. Of the 52 patients who entered the study, three did not complete the full 14 days of therapy, but in only one of those cases was the failure to complete due to noncompliance. Figure 1 shows the antisecretory therapy requirements at 8 wk and at 3 yr after H. pylori eradication therapy. Of the 46 patients with successful eradication of H. pylori, 31 (67%) were taking antisecretory therapy comprising either a PPI or H2RA for upper gut symptoms, and 15 (33%) were taking no antisecretory medications. At 3 yr after treatment, 16 of the 31 (52%) still required antisecretory therapy. These 16 patients were joined by two of the 15 (13%) patients who required no therapy at baseline but who did require therapy after H. pylori eradication. Thus, overall, successful eradication of H. pylori decreased the need for antisecretory therapy from 67% to 39% of the total population (95% CI ⫽ 0.08 – 0.49, BL vs 3 yr). Figure 2 shows the long-term effect on upper-gut symptoms after successful eradication of H. pylori. Thirty-eight of the 46 patients with H. pylori eradication responded to a telephone interview. Of the population with past or current ulcer disease, 69% responded that their symptoms were better than they were at baseline. Likewise, 72% of the nonulcer population answered similarly, and only 4% said they felt worse. There was no significant difference between patients with and without a history of peptic ulcer disease. Thus, H. pylori eradication improves upper-gut symptoms in both the ulcer population and the nonulcer population alike.

DISCUSSION Although the benefits of H. pylori eradication in patients with symptoms or complications of peptic ulcer are clear, the benefits of H. pylori eradication in the asymptomatic or dyspeptic nonulcer population remain controversial. Once

Table 3. Adverse Events, Compliance, and Evaluability Symptom

Number of Patients (%)

Diarrhea

26 (51.0)

Bad taste in mouth Nausea/upset stomach Dry mouth Headache Liver enzyme elevation

24 (47.1) 9 (17.6) 3 (5.8) 2 (3.9) 1 (1.9)

Amoxicillin rash

1 (1.9)

Emergency coronary artery bypass

1 (1.9)

Noncompliance

1 (1.9)

Duration Outcome

Mean Days

Range

Self-limiting (n ⫽ 25) Required treatment (n ⫽ 1) Self-limiting Self-limiting Self-limiting Self-limiting Developed after 12 days of therapy; selflimiting; resolved after 4 days Withdrawn: rash after 9 days of therapy; resolved in 24 h; H. pylori not eradicated Withdrawn: H. pylori not eradicated after 3 days of treatment Withdrawn: H. pylori successfully eradicated after 6 days of treatment

8.5

1–30

12.2 6.7

2–28 2–14

10.5 4.0

7–14 4

1

AJG – May, 2001

3 Years After H. pylori Eradication

1393

Figure 1. Antisecretory therapy requirements at 8 wk and 3 yr in the 46 patients with successful H. pylori eradication at 8 wk after eradication of H. pylori, two of 15 (13%) patients who were asymptomatic at baseline (BL) developed symptoms requiring antisecretory therapy, and 15 of 31 (48%) continued to require therapy. Thus, the percentage of those with successful H. pylori eradication that did not require therapy increased from 33% to 63%, and the percentage that did require antisecretory therapy decreased from 67% to 37% (95% CI ⫽ 0.10 – 0.51, BL vs 8 wk). By 3 yr after successful H. pylori eradication, only two of 15 (13%) patients who were asymptomatic at baseline developed symptoms requiring antisecretory therapy; and 16 of 31 (52%) patients who required therapy at baseline continued to require therapy. Overall, at 3 yr, the percentage of those with successful H. pylori eradication that did not require antisecretory therapy increased from 33% to 61%, and the percentage that did require antisecretory therapy decreased from 67% to 39% (95% CI ⫽ 0.08 – 0.49, BL vs 3 yr).

the decision to treat is made, however, the search begins for the optimal combination regimen and the optimal treatment duration. The regimen in our study included RBC, which

possesses the antisecretory activity of the H2RA, as well as the mucosal protective property of certain other bismuth salts (4). The significantly greater solubility of RBC at low

Figure 2. The effects on dyspeptic symptoms 3 yr after successful H. pylori eradication (n ⫽ 38). The question asked was, “Are your symptoms now (3 yr after therapy) better than, the same as, or worse than they were 3 yr ago before surgery?” The results included only 38 of the 46 patients with successful H. pylori eradication. There was no statistically significant difference between patients with and without a history of peptic ulcer disease.

1394

Sontag et al.

pH values provides better antipepsin and anti–H. pylori action compared with the insoluble admixture of bismuth citrate and ranitidine (5). Whether such enhanced solubility and antisecretory and mucosal protective effects translate into clinical efficacy or clinical superiority, however, remains unknown. In clinical trials, several published European studies have shown that RBC in combination with clarithromycin for 2 wk results in H. pylori eradication in 82–94% of patients (6 – 8). Our goal was to determine whether such eradication rates were possible in the United States, especially in a population thought to be noncompliant. Indeed, treatment failure is a result of numerous factors, including antibiotic resistance and compliance, both of which vary from community to community, state to state, and country to country. Treatment success rates in Europe seem in general to be 10% higher than those in the United States (9). There are three major reasons for such variability. First, resistance to the antibiotic components in triple-therapy regimens is a major factor in the failure to eradicate (10). Second, eradication rates from controlled trials published in medical journals may be higher than those in the community setting because of the high motivation of patients who join such studies, the required pill counts, and the motivating factors that accompany financial gain. Third, in almost all clinical trials, including ours, the truly resistant patients—those who had failed prior antibiotic treatment regimens—are almost always excluded from participation, and the study population ends up reflecting an “easier-to-treat” H. pylori–infected group. Indeed, most of the eradication rates reported in the literature reflect the success rates of this easier-to-treat group. Our study used the most stringent assessment of patient adherence to the protocol—the intent-to-treat statistical analysis. In such an analysis, all patients are considered to be failures unless H. pylori eradication is proven. This worse case assessment intentionally underestimates the true efficacy of the treatment regimen, as the burden of proof rests on the investigator and the patient. In our study, the healing rates were almost identical for the intent-to-treat analysis as for the per-protocol analysis, suggesting that this treatment combination is not only theoretically efficacious but practically efficacious as well. In regards to compliance, the VA population is generally thought to be less compliant than the nonveteran population (11). The overwhelming majority of our patients, however, completely finished their 14 days of therapy. Indeed, 49 of the 52 patients were totally compliant for the full 2 wk, and two of the three remaining patients (who took less than 14 days) were complying fully but were stopped by us (one because of an adverse event and one because of the need for coronary artery bypass surgery). Thus, acceptance with the therapeutic regimen translated into a 98.1% compliance rate. Only one patient intentionally took fewer than 14 days of therapy, and although this patient was noncompliant, H. pylori was successfully eradicated. Our compliance rate was

AJG – Vol. 96, No. 5, 2001

as high or higher than those reported in other major studies (12). Only one other U.S. study has used this identical regimen (RBC, 400 mg b.i.d.; amoxicillin, 1000 mg b.i.d.; and clarithromycin, 500 mg b.i.d.) in previously untreated H. pylori–infected patients for a full 14 days (13). In this study, Laine reported a 92% (46/50) intent-to-treat cure rate in 50 patients and a 96% (45/47) per-protocol cure rate. In addition, compliance with the triple therapy was 95%—almost identical to ours. Combining the results of both studies, the intent-to-treat cure rate in the 102 total patients is 90.2% (92/102; 95% CI ⫽ 84.4 –96.0), and the per-protocol cure rate is 89.5% (85/95; 95% CI ⫽ 83.3–95.6). One third of our population had well documented evidence of past or current ulcer disease. Two thirds had no such evidence. Regardless, H. pylori eradication resulted in equally impressive symptom improvement in both populations, with about 70% of respondents agreeing that they felt better. There are a number of theoretical explanations for why the current regimen provided such high cure rates (1): Previous treatment failures were excluded, thus eliminating patients with potentially resistant infections (2); amoxicillin may still be more effective than metronidazole, which has been weakened by growing resistance; (3) RBC cotherapy may be associated with a reduced frequency of antimicrobial resistance, especially to clarithromycin (14); and (4) the current full 14-day combination regimen may actually be the optimal and most reproducible strategy. In summary, our study has produced several clear conclusions: First, the high eradication rate, despite no control group, has established the efficacy of the treatment regimen in compliant patients. Second, serious adverse events are minimal and reversible, occurring in less than 4% of patients. Third, the veteran population (in the appropriate VA outpatient clinic setting) is considerably more compliant than is generally thought. Fourth, eradication of H. pylori decreases GI symptoms and the need for antisecretory therapy in both the peptic ulcer population and the symptomatic nonulcer population and, contrary to previous reports, does not result in an increase in GI symptoms. We conclude that RBC-based therapy taken in combination b.i.d. with amoxicillin (1000 mg) and clarithromycin (500 mg) is safe, easily complied with even in the veteran population, and highly effective in eradicating H. pylori infection. Eradication of H. pylori leads to a dramatic decrease in GI symptoms and need for antisecretory therapy.

ACKNOWLEDGMENT This study was funded by a grant from Glaxo Inc. Reprint requests and correspondence: Stephen J. Sontag, M.D., Veterans Affairs Hospital, Building 1, Room B321, P.O. Box 5000, Hines, IL 60141-5000. Received Aug. 19, 2000; accepted Nov. 27, 2000.

AJG – May, 2001

3 Years After H. pylori Eradication

REFERENCES 1. NIH Consensus Conference. Helicobacter pylori in peptic ulcer disease. JAMA 1994;272:65–9. 2. Dixon JS. Helicobacter pylori eradication: Unravelling the facts. Scand J Gastroenterol 1995;212(suppl 30):48 – 62. 3. Ciociola AA, Webb DD, Turner K. Dual and triple therapy regimens of antisecretory agents and antibiotics for the eradication of Helicobactor pylori: An overview. Scand J Gastroenterol 1996;218(suppl 31):3–9. 4. Stables R, Campbell CJ, Clayton NM, et al. Gastric antisecretory, mucosal protective, anti-pepsin and anti-Helicobacter properties of ranitidine bismuth citrate. Aliment Pharmacol Ther 1993;7:237– 46. 5. McColm AA, McLaren A, Klinkert G, et al. Ranitidine bismuth citrate; a novel anti-ulcer agent with different physicochemical characteristics and improved biological activity to a bismuth citrate-ranitidine admixture. Aliment Pharmacol Ther 1996;10:241–50. 6. Pounder RE, Wyeth JW, Duggan AE, et al. Ranitidine bismuth citrate with clarithromycin for the eradication of Helicobacter pylori and ulcer healing. Helicobacter 1997;2:132–9. 7. Bardhan KD, Dallaire C, Eisold H, et al. Ranitidine bismuth

8.

9. 10. 11.

12.

13.

14.

1395

citrate with clarithromycin for the treatment of duodenal ulcer. Gut 1997;41:181– 6. Axon ATR, Ireland A, Lancaster Smith MJ, et al. Ranitidine bismuth citrate and clarithromycin twice daily in the eradication of H. pylori. Aliment Pharmacol Ther 1997;11:81–7. Heatley R. The treatment of Helicobacter pylori infection. Aliment Pharmacol Ther 1992;6:291–303. Graham DY. Antibiotic resistance in Helicobacter pylori: Implications for therapy. Gastroenterology 1998;115:1272–7. Ashton CM, Peterson NJ, Wray NP, Yu HJ. The Veterans Affairs medical care system: Hospital and clinic utilization statistics for 1994. Med Care 1998;36:793– 803. Hopkins RJ, Girardi LS, Turney EA. Relationship between Helicobacter pylori eradication and reduced duodenal and gastric ulcer recurrence: A review. Gastroenterology 1996;110:1244 –52. Laine L, Estrada R, Trujillo M, Emami S. Randomized comparison of ranitidine bismuth citrate-based triple therapies for Helicobacter pylori. Am J Gastroenterol 1997;12:2213–5. Me´graud F, Pichavant R, Palegry D, et al. Ranitidine bismuth citrate co-prescribed with clarithromycin is more effective in the eradication of Helicobacter pylori than omeprazole with clarithromycin. Gut 1997;41(suppl 1):A92.