Five-day proton pump inhibitor-based quadruple therapy regimenis more effective than 7-day triple therapy regimenfor Helicobacter pylori infection

A. NAGAHARA, H. MIWA, T. YAMADA, A. KUROSAWA, R. OHKURA & N. SATO

Department of Gastroenterology, Juntendo University School of Medicine, Tokyo, Japan

Accepted for publication 19 October 2000

INTRODUCTION

Recent trend regarding curative treatments for Helico-

bacter pylori infections is a new triple therapy. Currently,

proton pump inhibitor-based 1 week triple therapy

regimens are widely prescribed, and are now known

to provide 80±90% eradication rates with few adverse

effects.1±4 As a result of their favourable clinical results,

these regimens have become standard. However, these

studies also demonstrate eradication failure rates at a

10±20% rate.

As discussed, many factors can contribute to or cause

eradication failure.5 Recently, the presence of a

clarithromycin-resistant strain of H. pylori was found

to be an important factor in the eradication failures by

proton pump inhibitor, amoxicillin and clarithromycin

regimens.6, 7 However, double resistance strains against

clarithromycin and metronidazole are reported at low

rates.8 Therefore, the addition of metronidazole to the

®rst line regimen, which includes clarithromycin, may

increase eradication rates. In fact, we have reported

that a 5-day quadruple therapyÐinvolving rabeprazole,

amoxicillin, clarithromycin and metronidazoleÐprovi-

ded a higher eradication rate than a 5-day triple

therapy regimen comprised of rabeprazole, amoxicillin

and clarithromycin.9 However, there are no comparable

SUMMARY

Background: There have been no reports that describe

whether 5-day quadruple therapy (rabeprazole + amox-

icillin + clarithromycin + metronidazole; RACM) could

substitute for standard 7-day triple therapy as a ®rst-line

therapy for Helicobacter pylori.

Patients and methods: This study was designed as a

randomized prospective single centre study. A total of

160 H. pylori-positive patients who had not received

therapy were given either a 5-day RACM regimen

(n � 80, rabeprazole 20 mg b.d., amoxicillin 750 mg

b.d., clarithromycin 200 mg b.d. and metronidazole

250 mg b.d.) or a 7-day RAC regimen (n � 80,

rabeprazole 20 mg b.d., amoxicillin 750 mg b.d. and

clarithromycin 200 mg b.d.). Cure of the infection was

assessed by a 13C urea breath test 1 month after the

completion of therapy.

Results: The eradication rates of the 5-day RACM

regimen and the 7-day RAC regimen were 93% (95%

CI: 84±97%) and 81% (95% CI: 71±89%) by intention-

to-treat analysis, 94% (95% CI: 86±98%) and 83%

(95% CI: 73±91%) by all-patients-treated analysis

analysis and 95% (95% CI: 87±98%; P < 0.05) and

82% (95% CI: 72±90%) by per protocol analysis,

respectively. No serious adverse effect was observed,

and 99% of the patients reported complete compli-

ance.

Conclusions: The cure rate of the 5-day RACM regimen

was more effective than the 7-day RAC regimen,

suggesting that this regimen could be preferable as a

®rst-line therapy for H. pylori infection.

Correspondence to: Dr N. Sato, Department of Gastroenterology, Juntendo

University School of Medicine, 2±1-1 Hongo Bunkyo-ku Tokyo 113±8421

Japan.E-mail: nsato@med.juntendo.ac.jp

Aliment Pharmacol Ther 2001; 15: 417±421.

Ó 2001 Blackwell Science Ltd 417

studies into whether short-term quadruple therapy

could replace 7-day proton pump inhibitor-based triple

therapy, which is currently the most popular and

effective regimen.

Therefore, in this study, the 5-day quadruple therapy

regimen consisting of rabeprazole, amoxicillin, clari-

thromycin and metronidazole was compared to the

7-day triple therapy regimen consisting of rabeprazole,

amoxicillin and clarithromycin, in terms of eradication

ef®cacy of H. pylori in our patient population.

PATIENTS AND METHODS

This was a randomized open, prospective single centre

study. All patients were diagnosed with gastric ulcers,

duodenal ulcers or non-ulcer dyspepsia with H. pylori

infections by upper gastrointestinal endoscopy before

receiving curative therapy. The presence of H. pylori

infection was assessed by histology (haematoxylin and

eosin staining), the rapid urease test (pyloritek; Serim

Research Corp., Elkhart, Indiana, USA) and the 13C urea

breath test. Positive H. pylori status was assigned based

upon at least two positive results. Patients who were

diagnosed as being H. pylori-positive and who wished to

receive curative therapy were enrolled into this study.

Written informed consent was obtained from all

patients before receiving the curative therapy.

A total of 160 consecutive H. pylori-positive patients

who visited the out-patient clinic of Juntendo University

Hospital between July 1999 and May 2000 were

enrolled into this study. Sixty-three patients had gastric

ulcers, 50 had duodenal ulcers, 21 had gastroduodenal

ulcers and 26 had non-ulcer dyspepsia. Of the 160

patients, 118 were male and 42 were female; they

were aged between 17 and 73 years (mean age

49.6 � 12.1 years). The exclusion criteria were as

follows: past history of drug allergy to proton pump

inhibitors, amoxicillin, clarithromycin or metronidazole;

previous curative therapy administered at our clinic

or another clinic; severe condition or illness such as

malignancy or hepatic or renal failure; pregnancy or

suspected pregnancy; and previous partial or total

gastrectomy.

Patients were randomly assigned into two groups: a

newly created regimen, which consisted of the 5-day

RACM (rabeprazole 20 mg b.d. + amoxicillin 750 mg

b.d. + clarithromycin 200 mg b.d. + metronidazole

250 mg b.d.); or the 7-day RAC regimen (rabeprazole

20 mg b.d. + amoxicillin 750 mg b.d. + clarithromy-

cin 200 mg b.d.). Patient pro®les in the two groups are

shown in Table 1. Potential cure of the infections was

assessed by the 13C urea breath test 1 month after

completion of therapy.

We adapted the urea breath test as described before.10

Brie¯y, the procedure was as follows: patients took

100 mg of 13C urea with 30 mL of tap water in the

sitting position after at least 5 h of fasting. The breath

samples were collected before and 20 min after 13C urea

administration. Tooth brushing and mouth wash were

performed before and immediately after the dosing of

the urea.

When the cut-off delta 13C value at 20 min was

de®ned as 5 ppm, the sensitivity, speci®city and accu-

racy were found to be 97%, 97%, and 97%, respectively.

In this test, patients whose delta 13C value at 20 min

was less than 5 ppm were considered to be cured; those

whose delta 13C value at 20 min was greater than 10

ppm were not considered to be cured. Patients whose

delta 13C values at 20 min were between 5 and 10 ppm

were classi®ed in the borderline group. Patients in this

borderline group were encouraged to undergo this test

again 2 months later. If at this time patients had a

repeat value which was less than 5 ppm, they were

considered as successfully cured, whilst patients with

repeat values equal to or more than 5 ppm were

considered to be treatment failures. `Borderline' patients

who did not take repeat 13C urea breath tests were

considered treatment failures.

The cure rate was de®ned as the number of success-

fully treated patients divided by the number of success-

fully plus number of unsuccessfully treated patients. The

cure rate was evaluated by intention-to-treat analysis,

all-patients-treated analysis and per protocol analysis.

With intention-to-treat analysis, all enrolled patients

were included, and patients who were lost to follow-up

and therefore did not take the follow-up 13C-urea breath

Table 1. Patient pro®les

5-day RACM 7-day RAC

Number of patients 80 80

Gender (male/female) 60/20 58/22

Mean age � s.d. 50.0 � 12.1 years 49.1 � 12.3 years

Gastric ulcer (scar) (%) 34 (42.5%) 29 (36.3%)

Duodenal ulcer (scar) (%) 22 (27.5%) 28 (35.0%)

Gastroduodenal ulcer

(scar) (%)

12 (15.0%) 9 (11.3%)

Non-ulcer dyspepsia (%) 12 (15.0%) 14 (17.5%)

418 A. NAGAHARA et al.

Ó 2001 Blackwell Science Ltd, Aliment Pharmacol Ther 15, 417±421

test were regarded as `failed to be cured'. With all-

patients-treated analysis, patients who did not take the

follow-up 13C urea breath tests were excluded from

analysis. With per protocol analysis, patients with

compliance rates less than 80% were excluded. The

cure rate was calculated together with a 95% con®d-

ence interval. Patients were interviewed about adverse

effects after completion of the therapy. The compliance

was assessed by interview and the number of pills that

were not taken during the therapy.

For statistical analysis, 95% con®dence intervals, the

Student's t-test, one-way analyses of variance and

Fisher's exact test were used. P-values less than 0.05

was regarded as statistically signi®cant.

RESULTS

Intention-to-treat analysis was performed in these 160

patients. Three patients who did not take follow-up13C urea breath test had withdrawn from the study, and

so 157 patients were available for all-patients-treated

analysis. Given that 10 patients were further excluded

from per protocol analysis because their compliance

rates were less than 80% or they not interviewed, a total

of 147 patients were included in per protocol analysis.

Potential cures of the infections were assessed by the13C-urea breath test 1 month after completion of

therapy. Ten patients showed borderline results at the

1-month follow-up 13C-urea breath test. Out of these

patients, six were assessed as cured by repeated13C-urea breath test, while four patients were consid-

ered treatment failures either by repeated 13C-urea

breath test (two patients) or because they were lost to

follow-up (two patients). The cure rates of each

therapeutic regimen are shown in Table 2 with 95%

con®dence intervals. The 5-day RACM group showed a

higher cure rate than those patients in ITT, APT and TT

analysis. By per protocol analysis, the 5-day RACM

regimen provided a signi®cantly higher eradication rate

than the 7-day RAC group (P < 0.05).

In total, 149 patients were interviewed about adverse

effects because 11 patients had withdrawn or had

visited other physicians who were not involved in this

study. Calculation of the rate of adverse effects included

only those patients who completed an interview. No

adverse effects were reported in 74% (110 out of 149)

of the interviewed patients. Diarrhoea or soft stool and

taste disturbances or glossitis were observed in 18% (27

out of 149) and 7% (11 out of 149), respectively. Skin

rashes were observed in 2% (three out of 149). The

incidence of diarrhoea or soft stool was signi®cantly

higher in the 5-day RACM group than in the 7-day RAC

group (Table 3). However, adverse effects did not

in¯uence compliance for any patient.

Among the 149 patients interviewed, 147 (99%)

reported complete compliance, while two patients

reported 50% and 90% compliance in the 5-day RACM

group and the 7-day RAC group, respectively.

DISCUSSION

When ®rst line proton pump inhibitor-based triple

therapy including clarithromycin has failed, the

remaining bacteria have been reported to frequently

acquire secondary resistance. It is reported that secon-

dary resistance against clarithromycin is as high as

40±50%.11±13 Since clarithromycin resistant strains are

a major obstacle standing in the way of improving

eradication therapy, retreatment for H. pylori becomes

dif®cult.6, 7 Thus, it is important to successfully eradi-

cate H. pylori during the ®rst time therapy.

Table 2. Cure rate for each regimen

ITT (95% CI)% APT (95% CI)% PP (95% CI)%

5-day RACM 93 (84±97) 94 (86±98) 95 (87±98)*

7-day RAC 81 (71±89) 83 (73±91) 82 (72±90)

ITT, intention-to-treat analysis; APT, all-patients-treated analysis;

PP, per protocol analysis.

Cure rates were described with 95% con®dence intervals.* There was a signi®cant difference between the two groups

(P < 0.05).

Table 3. Adverse effects with each regimen

RACM RAC Subtotal

None 52 58 110

Diarrhoea or soft stool 21* 6 27

Taste disturbance or glossitis 4 ,à 7  11

Skin rash 2§,± 1 3

Abdominal pain or nausea 0 3 3

Subtotal 79 75 154

Unknown 5 6 11

Total 84 81 165

* There was a signi®cant difference between the two groups(P < 0.01).

  One patient had both diarrhoea and taste disturbance.

à One patient had taste disturbance, soft stool and skin rash.

§ One patient had both diarrhoea and skin rash.± One patient had taste disturbance, soft stool and skin rash.

FIVE-DAY QUADRUPLE THERAPY ERADICATES H. PYLORI 419

Ó 2001 Blackwell Science Ltd, Aliment Pharmacol Ther 15, 417±421

On the other hand, a quadruple therapy regimen with

proton pump inhibitor plus standard bismuth triple

therapy for 7 days provides a consistently high eradi-

cation rate of approximately 96%.14, 15 However, so far,

quadruple therapy has been utilized in many reports as

a rescue therapy after triple therapy fails.16±18 There-

fore, it is crucial to investigate whether quadruple

therapy could be employed as a ®rst line therapy. In

fact, we have previously reported that 5-day quadruple

therapy provided a higher eradication rate than 5-day

triple therapy.9 Also in this study, the cure rate of the

5-day quadruple therapy regimen was signi®cantly

higher than the cure rate of the 7-day triple therapy

regimen.

Okada et al. reported that quadruple therapy consist-

ing of omeprazole, amoxicillin, roxithromycin and

metronidazole was found to successfully eradicate

even in roxithromycin- and/or metronidazole-resistant

strains.19 Primary resistance strains against amoxicillin,

clarithromycin and metronidazole occur at rates of

0.7%, 6.0% and 6.7%, respectively, and double resistant

strains for clarithromycin and metronidazole are

observed in only 0.7% at our hospital (unpublished

data). With this background, quadruple therapy using

clarithromycin and metronidazole might be effective

against resistant strains, and the low rate of double

resistant strains in our patient population might result

in higher cure rates in this 5-day RACM group than in

the 7-day RAC group.

Regarding the setting dosages of antibiotics for eradi-

cation regimens, prescribing amoxicillin 750 mg b.d.

with clarithromycin 200 mg or 400 mg b.d. is recom-

mended in the guideline which is decided by the

Japanese Society for Helicobacter Research.20 Further-

more, we recently reported that the eradication rate

between 200 mg b.d. and 400 mg b.d. of clarithromy-

cin was similar in proton pump inhibitor-based triple

therapy in our patient population, suggesting that

200 mg of clarithromycin twice daily is considered to

be suf®cient.21

No report has been found which used rabeprazole as

a proton pump inhibitor in a short-term regimen other

than those which we have previously reported.9 It is

well-known that neutralization of the acid is helpful

to maximize the potency of antibiotic action in the

treatment of H. pylori. Therefore, it is important to

consider prescribing rapid-acting proton pump inhibitor

for this kind of a short-term regimen. In this study, we

employed rabeprazole, which has been considered to

have a more potent effect on H. pylori than omepra-

zole.22, 23 Additionally, rabeprazole is a stronger gastric

acid inhibitor and inhibits acid secretion more rapidly

than lansoprazole.24, 25 These speci®c features of rab-

eprazole, especially rapid acid control, would improve

the cure rate of this short-term regimen.

Patient compliance is known to be a major host

predictive factor for treatment failure. Patient compliance

of less than 60% decreased the cure rate from 96% to 69%

and from 72% to 20%.26, 27 Simple regimens, such as a

twice daily quadruple regimen for short-term dosing

(5-day or 4-day) resulted in good eradication rates, that

is, of more than 90% in intention-to-treat analysis.28, 29

Therefore, short-term and simple regimens are recom-

mended in order to maintain good patient compliance. In

this study, complete compliance was reported in 99% of

the interviewed patients in both the 5-day RACM and

7-day RAC groups. Accordingly, short-term and simple

regimens might be responsible for this high compliance.

No adverse effects were observed in 69% and 78% of

those interviewed from the 5-day RACM and 7-day RAC

groups, respectively. This difference was not statistically

signi®cant. Diarrhoea or soft stools and glossitis or taste

disturbances were commonly seen in patients as adverse

effects of the both regimens. In the 5-day RACM group,

diarrhoea or soft stool was observed more frequently

than in the 7-day RAC group. However, these adverse

effects were generally mild and no patient dropped out

because of adverse effects, suggesting the clinical utility

of the 5-day RACM regimen as a ®rst line treatment.

In this study, the 5-day RACM regimen provided

higher cure rates than the 7-day RAC regimen, with few

adverse effects and good compliance, suggesting that

this 5-day RACM regimen can be considered a safe and

well-tolerated ®rst-line regimen.

REFERENCES

1 Miwa H, Ohkura R, Murai T, et al. Impact of rabeprazole, a

new proton pump inhibitor, in triple therapy for Helicobacter

pylori infectionÐcomparison with omeprazole and lansop-

razole. Aliment Pharmacol Ther 1999; 13: 741±6.

2 Miwa H, Nagahara A, Sato K, et al. Ef®cacy of 1 week

omeprazole or lansoprazole±amoxycillin±clarithromycin

therapy for Helicobacter pylori infection in the Japanese popu-

lation. J Gastroenterol Hepatol 1999; 14: 317±21.

3 Miwa H, Ohkura R, Murai T, et al. Effectiveness of omepraz-

ole±amoxicillin±clarithromycin (OAC) therapy for Helicobacter

pylori infection in a Japanese population. Helicobacter 1998;

3: 132±8.

420 A. NAGAHARA et al.

Ó 2001 Blackwell Science Ltd, Aliment Pharmacol Ther 15, 417±421

4 Lind T, van Veldhuyzen Zanten S, Unge P, et al. Eradication of

Helicobacter pylori using one-week triple therapies combining

omeprazole with two antimicrobials: the MACH I Study.

Helicobacter 1996; 1: 138±44.

5 Huang JQ, Hunt RH. Treatment after failure: the problem of

`non-responders'. Gut 1999; 45(Suppl. 1): I40±4.

6 Hoshiya S, Watanabe K, Tokunaga K, et al. Relationship

between eradication therapy and clarithromycin-resistant

Helicobacter pylori in Japan. J Gastroenterol 2000; 35: 10±4.

7 Ducons JA, Santolaria S, Guirao R, Ferrero M, Montoro M,

Gomollon F. Impact of clarithromycin resistance on the

effectiveness of a regimen for Helicobacter pylori: a prospective

study of 1-week lansoprazole, amoxycillin and clarithromycin

in active peptic ulcer. Aliment Pharmacol Ther 1999; 13:

775±80.

8 Mendonca S, Ecclissato C, Sartori MS, et al. Prevalence of

Helicobacter pylori resistance to metronidazole, clarithromycin,

amoxicillin, tetracycline, and furazolidone in Brazil. Helico-

bacter 2000; 5: 79±83.

9 Nagahara A, Miwa H, Ogawa K, et al. Addition of metroni-

dazole to rabeprazole±amoxicillin±clarithromycin regimen for

Helicobacter pylori infection provides an excellent cure rate

with ®ve-day therapy. Helicobacter 2000; 5: 88±93.

10 Miwa H, Murai T, Ohkura R, et al. Usefulness of the 13C urea

breath test for detection of Helicobacter pylori infection in

fasting patients. J Gastroenterol Hepatol 1998; 13: 1039±43.

11 Murakami K, Kimoto M. Antibiotic-resistant H. pylori strains

in the last ten years in Japan. Nippon Rinsho 1999; 57: 81±6.

12 Sakurai K, Takahashi H, Yamaguchi Y, et al. Importance of

drug selection and the use of sensitivity tests for the eradica-

tion therapy for Helicobacter pylori. Nippon Rinsho 1999; 57:

72±5.

13 Suzuki J, Mine T, Kobayasi I, Fujita T. Assessment of a new

triple agent regimen for the eradication of Helicobacter pylori

and the nature of H. pylori resistance to this therapy in Japan.

Helicobacter 1998; 3: 59±63.

14 Borody TJ, Shortis NP, Reyes E. Eradication therapies for

Helicobacter pylori. J Gastroenterol 1998; 33(Suppl. 10): 53±6.

15 van der Hulst RW, Keller JJ, Rauws EA, Tytgat GN. Treatment

of Helicobacter pylori infection: a review of the world literature.

Helicobacter 1996; 1: 6±19.

16 Chan FK, Sung JJ, Suen R, Wu JC, Ling TK, Chung SC. Salvage

therapies after failure of Helicobacter pylori eradication with

ranitidine bismuth citrate-based therapies. Aliment Pharma-

col Ther 2000; 14: 91±5.

17 Gasbarrini A, Ojetti V, Armuzzi A, et al. Ef®cacy of a multistep

strategy for Helicobacter pylori eradication. Aliment Pharmacol

Ther 2000; 14: 79±83.

18 Gomollon F, Ducons JA, Ferrero M, et al. Quadruple therapy is

effective for eradicating Helicobacter pylori after failure of triple

proton-pump inhibitor-based therapy: a detailed, prospective

analysis of 21 consecutive cases. Helicobacter 1999; 4:

222±5.

19 Okada M, Nishimura H, Kawashima M, et al. A new quad-

ruple therapy for Helicobacter pylori: in¯uence of resistant

strains on treatment outcome. Aliment Pharmacol Ther

1999; 13: 769±74.

20 The Japanese Society for Helicobacter Research. The guideline

of diagnosis and eradication for Helicobacter pylori infection:

Japanese J Helicobacter Res 2000 2(Suppl.): 2±12(in Japan-

ese).

21 Miwa H, Murai T, Sato K, et al. Comparison of the ef®cacy of

400 mg and 800 mg of clarithromycin used with lansopraz-

ole and amoxicillin in eradication regimens for Helicobacter

pylori infection in a Japanese population. J Gastroenterol

2000; 35: 536±9.

22 Park JB, Imamura L, Kobashi K. Kinetic studies of Helicobacter

pylori urease inhibition by a novel proton pump inhibitor,

rabeprazole. Biol Pharm Bull 1996; 19: 182±7.

23 Hirai M, Azuma T, Ito S, Kato T, Kohli Y. A proton pump

inhibitor, E3810, has antibacterial activity through binding to

Helicobacter pylori. J Gastroenterol 1995; 30: 461±4.

24 Fujisaki H, Oketani K, Shibata H, et al. Inhibitory action of

E3810 on H+, K+-ATPase and gastric acid secretion in vitro.

Nippon Yakurigaku Zasshi 1993; 102: 389±97.

25 Tomiyama Y, Morii M, Takeguchi N. Speci®c proton pump

inhibitors E3810 and lansoprazole affect the recovery process

of gastric secretion in rats differently. Biochem Pharmacol

1994; 48: 2049±55.

26 Graham DY, Lew GM, Malaty HM, et al. Factors in¯uencing

the eradication of Helicobacter pylori with triple therapy. Gas-

troenterology 1992; 102: 493±6.

27 Wermeille J, Dederding JP, Cunningham M, et al. Failure of

Helicobacter pylori eradication in an ambulatory population: Is

poor compliance the main cause? Digestion 1998; 59

(Suppl. 3): 437.

28 Treiber G, Ammon S, Schneider E, Klotz U. Amoxicillin/

metronidazole/omeprazole/clarithromycin: a new, short

quadruple therapy for Helicobacter pylori eradication. Heli-

cobacter 1998; 3: 54±8.

29 Calvet X, Tito L, Comet R, Garcia N, Campo R, Brullet E. Four-

day, twice daily, quadruple therapy with amoxicillin, clari-

thromycin, tinidazole and omeprazole to cure Helicobacter

pylori infection: a pilot study. Helicobacter 2000; 5: 52±6.

FIVE-DAY QUADRUPLE THERAPY ERADICATES H. PYLORI 421

Ó 2001 Blackwell Science Ltd, Aliment Pharmacol Ther 15, 417±421