Famotidine for the prevention of peptic ulcers and oesophagitis in patients taking low-dose aspirin (famous): a phase iii, randomised, double-blind, placebo-controlled trial

Articles
Famotidine for the prevention of peptic ulcers and
oesophagitis in patients taking low-dose aspirin (FAMOUS):
a phase III, randomised, double-blind, placebo-controlled trial
Ali S Taha, Caroline McCloskey, Rakesh Prasad, Vladimir Bezlyak Summary
Background There are few therapeutic options for the prevention of gastrointestinal mucosal damage caused by low- Lancet 2009; 374: 119–25
dose aspirin. We therefore investigated the effi
cacy of famotidine, a well-tolerated histamine H₂-receptor antagonist, Published Online
in the prevention of peptic ulcers and erosive oesophagitis in patients receiving low-dose aspirin for vascular July 6, 2009
protection.
See Comment page 93
Methods Adult patients (aged ≥18 years) from the cardiovascular, cerebrovascular, and diabetes clinics at Crosshouse
Hospital, Kilmarnock, UK, were eligible for enrolment in this phase III, randomised, double-blind, placebo-controlled Gastroenterology Unit,
Crosshouse Hospital,
trial if they were taking aspirin 75–325 mg per day with or without other cardioprotective drugs. Patients without University of Glasgow,
ulcers or erosive oesophagitis on endoscopy at baseline were randomly assigned by computer-generated randomisation Kilmarnock, UK (A S Taha MD,
sequence to receive famotidine 20 mg twice daily (n=204) or placebo twice daily (n=200). Patients had a fi nal C McCloskey BSc,
endoscopic examination at 12 weeks. The primary endpoint was the development of new ulcers in the stomach or R Prasad MRCP); and
Department of Medicine
duodenum or erosive oesophagitis at 12 weeks after randomisation. Analysis was by intention to treat, including all (A S Taha, C McCloskey,
randomised patients who received at least one dose of study drug (famotidine or placebo). This trial is registered as R Prasad), and Robertson
an International Standard Randomised Clinical Trial, number ISRCTN96975557.
Centre for Biostatistics
(V Bezlyak PhD), University of
Glasgow, UK
Findings All randomised patients received at least one dose and were included in the ITT population. 82 patients Correspondence to:
(famotidine, n=33; placebo, n=49) did not have the fi nal endoscopic examination and were assumed to have had Dr Ali S Taha, Crosshouse Hospital,
normal fi ndings; the main reason for participant withdrawal was refusal to continue. At 12 weeks, comparing patients Kilmarnock KA2 0BE, UK
assigned to famotidine with patients assigned to placebo, gastric ulcers had developed in seven (3·4%) of 204 patients ali.taha1@btinternet.com
compared with 30 (15·0%) of 200 patients (odds ratio [OR] 0·20, 95% CI 0·09–0·47; p=0·0002); duodenal ulcers had
developed in one (0·5%) patient compared with 17 (8·5%; OR 0·05, 0·01–0·40; p=0·0045); and erosive oesophagitis
in nine (4·4%) compared with 38 (19·0%; OR 0·20, 0·09–0·42; p<0·0001), respectively. There were fewer adverse
events in the famotidine group than in the placebo group (nine vs 15); four patients in the placebo group were admitted
to hospital with upper gastrointestinal haemorrhage. The other most common adverse event was angina (famotidine,
n=2; placebo, n=4).
Interpretation Famotidine is eff ective in the prevention of gastric and duodenal ulcers, and erosive oesophagitis in
patients taking low-dose aspirin. These fi ndings widen the therapeutic options for the prevention of gastrointestinal
damage in patients needing vascular protection.
Funding Merck Laboratories and Astellas Pharma.
Introduction
related to aspirin or non-steroidal anti-infl ammatory Low-dose aspirin (ie, 75–325 mg), is one of the most drugs (NSAIDs),7,8 but there have been concerns about widely used drugs in the world. Increasingly, it is being their costs,9 safety,10–12 and risk of interaction with bought over-the-counter or prescribed for its anti- clopidogrel, which is frequently prescribed concurrently thrombotic activity in cardiovascular and cerebrovascular with aspirin.13,14 Also, although studies have focused on diseases, and in diabetes mellitus.1–3 Despite the benefi ts peptic ulcers caused by NSAIDs,15 recent evidence of antithrombotic strategies, there has been a rise in the suggests that erosive oesophagitis is frequently seen in incidence of major upper gastrointestinal complications patients taking low-dose aspirin for vascular in patients taking aspirin, such as peptic ulcer bleeding, protection.7,16perforation, and sometimes death.4–6 The prevention of Famotidine is a histamine H -receptor antagonist that these side-eff ects has been hindered by two main factors: has proved to be well tolerated and able to prevent and the scarcity of studies on gastrointestinal mucosal heal peptic ulcers in patients receiving conventional damage in patients with vascular disorders who require NSAIDs.17–19 The FAMOUS trial (Famotidine for the low-dose aspirin, and the few therapeutic options that are Prevention of Peptic Ulcers in Users of Low-dose Aspirin) currently available. Proton-pump inhibitors are known to be eff ective in the treatment and prevention of ulcers standard dose of 20 mg twice daily, in the prevention of www.thelancet.com Vol 374 July 11, 2009
Articles
gastric and duodenal ulcers, and erosive oesophagitis in erosive oesophagitis, oesophageal strictures, or gastric or patients receiving low-dose aspirin for vascular duodenal ulcers.
protection.
To standardise the reporting criteria for the endoscopic fi ndings, the two endoscopists (AST and CMcC) attended each other’s endoscopic sessions before and regularly Participants
This phase III, randomised, double-blind, parallel-group, The trial was managed by the National Health Service of placebo-controlled trial recruited patients from the Greater Glasgow and Clyde (UK) in line with the European cardiovascular, cerebrovascular, and diabetes clinics at Directive on Medical Research. The study was registered Crosshouse Hospital, Kilmarnock, UK (affi liated to the with the Medicines and Healthcare Products Regulatory University of Glasgow, UK), who were receiving anti- Agency of the UK. The study and its data were controlled thrombotic low-dose aspirin (75–325 mg daily).
by an independent data and safety monitoring committee, Adult patients, aged 18 years or over, were eligible for who considered the safety aspects of the trial and decided enrolment if they had an indication for the antithrombotic on the interim analyses and their eff ect on the duration eff ect of aspirin that was stable at the time of randomisation, such as angina, previous myocardial infarction (12 weeks
or more before recruitment), cerebrovascular disease, Randomisation and masking
diabetes, or peripheral vascular disease. Also, the use of Patients were randomly assigned by computer-generated
aspirin had to be likely to continue for 12 weeks or longer,
randomisation sequence to receive one 20 mg tablet of in the presence or absence of mild to moderate dyspeptic famotidine (Pepcid, Merck, Hoddesdon, UK) twice daily or refl ux symptoms, or gastric or duodenal scars or or one placebo tablet twice daily. This sequence was erosions on endoscopy at baseline. Enrolled patients were generated by the Production Unit, Pharmacy Department, allowed to continue to take other antiplatelet agents, such Western Infi rmary, Glasgow, UK. The assignments of as clopidogrel and dipyridamole, and other treatments for participants were concealed in sealed opaque envelopes their basic or coexisting medical conditions. Patients were and without knowledge of the next assignment in the excluded if they had current malignancy, or a history of sequence. The randomisation codes and numbers were oesophageal, gastric, or duodenal surgery, Zollinger- kept at the Pharmacy Department, Western Infi rmary, Ellison syndrome, or primary oesophageal motility on behalf of the sponsors, and at the offi disorder. Women of childbearing potential were required Pharmacist, Crosshouse Hospital, in case of emergencies to maintain eff ective contraception—as judged by the related to the trial. The codes were broken by the investigator—during the study period, and excluded if independent data and safety monitoring committee at they were pregnant or lactating. Patients were also the time of statistical analyses without the involvement excluded if they had contraindications to study drugs, if of the research team. The placebo tablets contained they used proton-pump inhibitors, H -receptor antagonists, lactose and were identical in shape and taste to the or sucralfate within a week of the fi rst endoscopy, or if famotidine tablets. Both the participants and those giving they had ever been treated for Helicobacter pylori, since the interventions were masked to group assignment.
these treatments might interfere with the risk of
ulcer development. The intake of the following agents Procedures
also precluded patients from entering the trial: other Antacid tablets (magaldrate, alginate, and sodium
investigational compounds, anticholinergic drugs, bicarbonate mixture; Bisodol, Forest Laboratories, Bexley,
prostaglandin analogues, warfarin, high-dose steroids Kent, UK) were provided for patients from both treatment
(more than 7·5 mg of prednisolone or its equivalent daily),
groups for the relief of heartburn and dyspeptic cytotoxic drugs, NSAIDs, or bisphosphonates.
symptoms. Endoscopic assessments were done at The recruitment was undertaken by a team of four baseline and 12 weeks, and clinical assessments were researchers, who invited all eligible patients to consider done at baseline, 6 weeks, and 12 weeks after participating in the trial. Patients who accepted the randomisation. The procedures for these assessments invitation had an endoscopic examination after a cooling- are detailed in webappendix pp 1–2.
off period. The study protocol was approved by the local Comorbid conditions were recorded by calculating the ethics committee, and written informed consent was Charlson index (scale 1–6) for each patient.6,23 Three levels obtained from all patients.
of comorbidity were defi ned according to this index: low Endoscopic fi ndings were recorded for the oesophagus, (0 points), medium (1–2) and high (3 or more).
stomach, and duodenum. The defi nitions of ulcers, The primary endpoint was the development of new erosions, scars, Lanza scores, and oesophagitis,17–22 and ulcers (size ≥3 mm) in the stomach or duodenum or method for H pylori detection are all detailed in erosive oesophagitis at 12 weeks after randomisation. See Online for webappendix
The secondary endpoints included the Lanza scores for Patients were not randomised if they had any of the gastric and duodenal erosions, abdominal and vascular following endoscopic fi ndings at baseline: malignancy, symptoms scores, overall treatment assessment, and www.thelancet.com Vol 374 July 11, 2009
Articles
antacid consumption. Premature exit from the trial for any reason and adverse events were recorded as part of 1561 from the diabetes clinic1509 from the cerebrovascular clinic Statistical analysis
In a 12-week study on low-dose aspirin, Laine and
colleagues24 based their sample size calculations on the assumption of an aspirin-related peptic ulcer rate of 10%. They reported that in patients with osteoarthritis who were taking aspirin, 12-week ulcer rates were 7% in all aspirin users, 13% in those with baseline erosions, and 27% in those with previous history of ulcers or their We planned to study the rate of ulcer formation in patients taking low-dose aspirin for vascular protection in the presence or absence of dyspeptic symptoms, previous history of ulcers, H pylori infection, gastric scars or erosions, or duodenal scars or erosions. We assumed that mucosal scars represented an objective evidence of previous peptic ulcers. By contrast with the patients with osteoarthritis who were in fairly good health, erosive or ulcerative lesions were found in about 48% of patients taking low-dose aspirin for vascular protection.25 We also previously found that erosive oesophagitis was present in 27% of patients taking low-dose aspirin and presenting with upper gastrointestinal bleeding.16 Yeomans and colleagues7 reported that oesophagitis developed in 18% of non-bleeding patients on low-dose aspirin. In the current study, and taking these previously reported ulcer rates and Figure 1: Trial profi le
hypotheses into account, we assumed that peptic ulcers or oesophagitis, or both, would develop in 13% of patients in our trial. To show a halving of this rate to 6·5% in the the remaining and already randomised patients had active group compared with placebo, with 80% power and at a 5% signifi cance level, we calculated that a sample size Data were summarised as categorical (the numbers of 700 patients would be needed (350 in each group).
and proportions of patients in each category), and Three interim analyses were planned, with the fi rst continuous variables (mean, median, SD, minimum, scheduled after the randomisation of the fi rst 200 patients. and maximum values). For the analysis of the primary However, we later decided to carry out an interim analysis outcome, frequencies, proportions, and totals are after 100 patients had been recruited because of safety presented. The odds ratios (ORs) and 95% CIs for factors: comorbidity of volunteers, cardiac deaths, and famotidine versus placebo were calculated by logistic admissions with severe chest pain were all higher than regression without adjustment for other covariates. The expected, and there was a higher overall ulcer rate than proportional odds model and 95% CIs were used to originally predicted.
compare the grades of oesophagitis in the famotidine The interim analyses included formal statistical and placebo groups at the completion of the trial. p values comparisons between treatment groups for the outcome for the primary analysis were calculated by use of logistic measures. Additional factors in relation to the review of regression and the Wald test. For the analysis of secondary safety data were also taken into account. For consideration outcomes, categorical variables and the percentage of of early termination of the trial because of overwhelming patients were compared between treatment groups by evidence of benefi t with respect to the primary endpoint, use of logistic regression. For rank variables (Lanza we used the method of Lan and DeMets26 with the scores, cardiovascular and gastrointestinal symptoms), modifi ed O’Brien and Fleming27 alpha spending function. the eff ect of famotidine was estimated by use of the non- After approximately 300 patients had been recruited, a parametric Wilcoxon’s rank sum test. Analysis was by nominal value of p<0·00107 was needed before early intention to treat (ITT) and per protocol (PP). The termination would be considered. On the basis of the intention-to-treat population included all randomised third analysis, the independent data and safety monitoring patients who received at least one dose of study drug committee recommended that no further patients should (famotidine or placebo). Patients who did not have the be recruited and that the study should be stopped once fi nal endoscopic examination were assumed to have had www.thelancet.com Vol 374 July 11, 2009
Articles
for Biostatistics, University of Glasgow. This trial is Famotidine (n=204)
Placebo (n=200)
registered as an International Standard Randomised Clinical Trial, number ISRCTN96975557.
Role of the funding source
This was an investigator-initiated trial. Neither Merck Laboratories nor Astellas Pharma had any input or infl uence on the study design, data collection, data analysis, data interpretation, or writing of the report, or the decision to submit the article for publication. The sponsors’ support was totally unrestricted. However, copies of the protocol and the manuscript were forwarded to them for their records. The corresponding author had full access to all data in the study and had fi nal responsibility for the decision to submit for publication.
Figure 1 shows the trial profi le. All 404 randomised patients received at least one dose and were included in the ITT population. 82 patients (famotidine, n=33; placebo, n=49) did not have the fi nal endoscopic examination and were assumed to have had normal fi ndings. Table 1 shows the baseline characteristics of all study participants. 277 (68·6%) patients were men, and the median age was 63 years. Although only a few patients had gastrointestinal symptoms at baseline (famotidine, n=100; placebo, n=83), a substantial proportion had scars (famotidine, n=108 [52·9%]; placebo, n=94 [47·0%]) or erosions (famotidine, n=118 [57·8%]; placebo, n=97 [48·5%]) in the gastroduodenal mucosa. Also, most patients (famotidine, n=121 [59·3%]; placebo, n=127 [63·5%]) had other medical conditions, refl ected by their medium and high comorbidity scores.
Table 2 and fi gure 2 show the number of patients with gastric and duodenal ulcers, and erosive oesophagitis at 12 weeks. All but one duodenal ulcer in the famotidine group and two gastric ulcers in the placebo group were 5 mm or more. The proportion of patients who developed peptic ulcers of any size or erosive oesophagitis, or both, was lower in the famotidine group than in the placebo group (5·4% vs 32·5%; OR 0·12, 95% CI 0·06–0·23; A subgroup analysis in patients who developed peptic ulcers showed that H pylori infection was present in 16 (42·1%) of 38 patients in the placebo group and in none of eight in the famotidine group.
Data are n (%) or median (range). ACE=angiotensin-converting enzyme. *Recorded by use of the Charlson index (scale 1–6).6,23 The grades of erosive oesophagitis are shown in fi gure 2. The famotidine group had more patients with Table 1: Baseline characteristics of study participants
grade 0, or normal oesophagus (OR 0·20, 0·09–0·42; p<0·0001), and fewer patients with other grades of normal fi ndings. The PP population included all those in oesophagitis than did the placebo group.
the ITT population who took more than 80% of the study Multiple lesions (combinations of erosive oesophagitis, drugs, did not take NSAIDs or other anti-ulcer drugs gastric ulcers, or duodenal ulcers) were present in fi ve (with the exception of antacids), and who underwent the (2·5%) patients assigned to famotidine and in 17 (8·5%) fi nal endoscopic examination. Statistical analyses were patients assigned to placebo (webappendix p 2). The done with SAS version 9.1 and carried out and validated, distribution of the primary endoscopic endpoints by age- independent of the investigators, at the Robertson Centre www.thelancet.com Vol 374 July 11, 2009
Articles
322 patients underwent the fi nal endoscopic assessment Famotidine
Odds ratio
at 12 weeks. At this assessment, the famotidine group had more patients with a Lanza score of 0 (ie, no erosions or ulcers) in the stomach (110 [64·3%] of 171 vs 40 [26·5%] of 151; p<0·0001) and in the duodenum (155 [90·6%] of 171 vs 93 [62·0%] of 150; p<0·0001) than did the control group. 165 (95·9%) of 172 patients assigned to famotidine did not have dysphagia compared with 131 (89·1%) of 147 assigned to placebo (p=0·019). At 12 weeks, patients were invited to provide their overall treatment assessment of Erosive oesophagitis and/or ulcer >5 mm the trial, graded as better, the same, or worse than their Data are n (%). n/a=not applicable. *An ulcer was defi ned as an excavated and deep mucosal lesion measuring 3 mm or assessment at baseline. A better assesment was reported more in diameter. †Grades of erosive oesophagitis are shown in fi gure 2 and defi ned in webappendix p 1.
by 40 (23·5%) of 170 patients in the famotidine group Table 2: Patients with endoscopic lesions at 12 weeks
compared with 21 (14·6%) of 144 in the placebo group (p=0·012).
We investigated the eff ects of several factors on the risk of development of new ulcers in the stomach or duodenum or of erosive oesophagitis. Patients taking β blockers had a higher risk of these lesions: they were found in 54 (23·9%) of 226 patients taking β blockers compared with 24 (13·5%) of 178 patients not taking these agents (OR 2·57, 1·43–4·62; p=0·0017 [multivariate analysis after adjustment for treatment with famotidine]). The presence of all other concurrently prescribed drugs did not aff ect risk of lesion development. Baseline endoscopic abnormalities in the stomach or duodenum (ie, scars or erosions), were also associated with a higher risk of development of ulcers or erosive oesophagitis. These lesions developed in 70 (20·6%) of 339 patients who had scars or erosions at baseline compared with eight (12·3%) of 65 who did not (OR 12·27, 1·63–92·63; p=0·0015 [multivariate analysis, after adjustment for treatment with famotidine]).
More details of the secondary analyses are shown in webappendix p 3. The results of the PP analysis for the primary endpoint were consistent with those of the ITT The reasons for withdrawal are shown in fi gure 1 and the details of adverse events are shown in table 3. Although there were no signifi cant diff erences between the two groups, patients assigned to placebo were more likely to dropout (placebo n=25 vs famotidine n=21) and to have used proton-pump inhibitors (n=11 vs n=3) than were patients assigned to famotidine. Fewer adverse events were reported in the famotidine group (n=9) than in controls (n=15). All adverse events were judged not related to study drug. Four patients from the placebo group developed upper gastrointestinal haemorrhage and were admitted to hospital. Two of these patients needed blood transfusion, three were on β blockers, and one was on clopidogrel in addition to aspirin. More details on these four patients are provided in We tested for an interaction between famotidine and clopidogrel. We found that adverse events were not Figure 2: Primary endoscopic endpoints at 12 weeks
increased by taking the two agents together compared (A) Peptic ulcers. Compared with placebo, patients taking famotidine had signifi cantly fewer cases of gastric
(*p=0·0002), or duodenal ulcers (†p=0·0045) measuring 3 mm or more in diameter. (B) Grades of erosive with taking only one (ie, famotidine alone or placebo oesophagitis. There were fewer cases of erosive oesophagitis and more patients with grade 0 or normal oesophagus plus clopidogrel). Five (55·6%) of the nine patients with (‡p<0·0001) in the famotidine group than in the placebo group. For defi nition of grades see webappendix p 1.
www.thelancet.com Vol 374 July 11, 2009
Articles
been highlighted in a recent multicentre trial by Yeomans Famotidine
and co-workers:7 such rates ranged from 0% to 15% among the participating centres, probably refl ecting the doses of aspirin used, presence or absence of H pylori, and inter- observer variation in reporting the endoscopic fi ndings.7 In our trial, we deemed the presence of scars as indicative of healed or previous history of ulcers. In addition to enrolling patients with scars and erosions, which are known to increase the risk of ulcers,15,24 we also took into account the possible aff ect of drugs concurrently prescribed with aspirin. In particular, we found that β blockers act as an independent factor in increasing the risk of erosive or ulcerative lesions in patients on aspirin, which might suggest a role for mucosal ischaemia in patients taking aspirin with already compromised circulation. However, our analysis of concurrently prescribed drugs was exploratory in nature and might represent a chance Our work diff ers from other similar studies because it Data are n (%). No patient had more than one event. *The patient with cardiac failure and the two patients with myocardial infarction died.
was a single-centre trial, which kept inter-observer variation to a minimum. One cannot exclude the Table 3: Adverse events
possibility that compliance with aspirin intake improved in the study population because of the emphasis by the adverse events in the famotidine group were on research team and the study literature on the importance clopidogrel, compared with six (40·0%) of 15 in the of taking study drugs as well as other concurrently placebo group (OR 0·90, 0·58–1·40; p=0·6412). Also, prescribed agents (including aspirin). Such an improve-no increase in cardiac events was reported in patients ment might have contributed to more ulcer formation assigned to famotidine (table 3).
related to aspirin. It is also possible that patients who were at a greater risk of ulceration were more likely to Discussion
participate in the study. However, these factors do not This study shows that famotidine is eff ective for the aff ect our main fi ndings or conclusions with respect to prevention of peptic ulcers and erosive oesophagitis in the effi patients taking low-dose aspirin. The risk of developing We have shown that famotidine is capable of preventing these lesions is increased in patients treated concurrently oesophagitis in patients taking low-dose aspirin. We with β blockers and in those with gastrointestinal previously identifi ed the disorder in 27% of patients mucosal scarring or erosions at baseline.
taking aspirin who presented with upper gastrointestinal Although our main aim was to investigate the eff ects of bleeding.16 The frequency of oesophagitis in the placebo famotidine on the development of peptic ulcers and group in this study (19%) is similar to that described by erosive oesophagitis, it is still important to compare the Yeomans and colleagues7 in non-bleeding patients taking ulcer rates found in this study with those of previous aspirin (ie, 18%).7 Not unlike peptic ulcers, oesophagitis studies, while being aware that such rates will vary could be related to interference with mucosal prosta-according to the characteristics of the populations glandins by aspirin, or greater vulnerability to acid, studied. The rate of ulcers in our study population hence the benefi cial eff ect of famotidine.16,28(11·4%) is higher than the rates in some recently Gastrointestinal mucosal damage related to low-dose reported trials (5·4%7 and 7·0%24 ), but a clearer view aspirin (similarly to damage associated with NSAIDs) emerges when the individual study populations are is frequently asymptomatic.7,16 Only a small proportion considered.
of patients had upper abdominal complaints at Since most of the patients in our trial had erosions or enrolment, which refl ects their random inclusion. The scars at baseline, our fi ndings can be compared with ulcer individual symptoms scores did not change by the end rates reported by Laine and colleagues24 in patients with of the study. However, patients assigned to famotidine osteoarthritis (ie, 13% in those with baseline erosions, and were less likely to complain of dysphagia and their 27% in those with previous history of ulcers or their overall treatment assessment was better than those complications).24 When the clinical indication for the use assigned to placebo. Also, more patients in the placebo of aspirin is considered, erosive or ulcerative lesions were group dropped out because of the use of proton-pump found in about 48% of patients taking low-dose aspirin for inhibitors than did patients assigned to famotidine.
vascular protection.25 The important eff ect of study Although this trial was not powered to study serious population on ulcer rates in users of low-dose aspirin has ulcer complications, it is noteworthy that all patients www.thelancet.com Vol 374 July 11, 2009
Articles
presenting with upper gastrointestinal bleeding were in 7 Yeomans N, Lanas A, Labenz J, et al. Effi the placebo group. Ulcer complications, such as bleeding, (20 mg once daily) for reducing the risk of gastroduodenal ulcers associated with continuous use of low-dose aspirin. generally depend on the size and depth of mucosal Am J Gastroenterol 2008; 103: 2465–73.
damage. Besides its ulcerogenic eff ect, aspirin brings the Lanas A, Rodrigo L, Marquez JL, et al. Low frequency of upper added risk of its antiplatelet activity, which might gastrointestinal complications in a cohort of high-risk patients taking low-dose aspirin or NSAIDs and omeprazole. aggravate the bleeding potential of gastrointestinal Scand J Gastroenterol 2003; 38: 693–700.
mucosal damage. We found no evidence of an interaction Forgacs I, Loganayagam A. Overprescribing proton pump inhibitors between famotidine and clopidogrel. This could be is expensive and not evidence based. BMJ 2008; 336: 2–3.
because of the fairly small size of our trial, or the fact that 10 Garcia Rodriguez LA, Ruigomez A, Panes J. Use of acid-suppressing drugs and the risk of bacterial gastroenteritis. the metabolism of famotidine plus clopidogrel diff ers Clin Gastroenterol Hepatol 2007; 5: 1418–23.
from that of proton-pump inhibitors plus clopidogrel.13,14 11 Sarkar M, Hennessy S, Yang YX. Proton-pump inhibitor use and the risk for community-acquired pneumonia. Ann Intern Med 2008;
149: 391–98.
and aspirin-related ulcers,7,8 concerns continue to be 12 Targownik LE, Lix LM, Metge CJ, et al. Use of proton pump inhibitors expressed about the overprescribing9 and long-term use and risk of osteoporosis related fractures. CMAJ 2008; 179: 319–26.
of proton-pump inhibitors,10–14 and alternative 13 Ho PM, Maddox TM, Wang L, et al. Risk of adverse outcomes management strategies have been proposed, including associated with concomitant use of clopidogrel and proton pump inhibitors following acute coronary syndrome. JAMA 2009; the use of H -receptor antagonists.29 Thus, famotidine might be a useful alternative for proton-pump inhibitors 14 Juurlink DN, Gomes T, Ko DT, et al. A population-based study of in patients taking low-dose aspirin.
the drug interaction between proton pump inhibitors and
clopidogrel. CMAJ 2009; 180: 713–38.
Contributors
15 Taha AS, Sturrock RD, Russell RI. Mucosal erosions in chronic AST wrote the protocol and the draft manuscript. AST and CMcC NSAID users. Predisposition to ulceration and relationship to undertook the endoscopic assessments. RP helped with the clinical Helicobacter pylori. Gut 1995; 36: 334–36.
assessments. VB helped with the interpretation of the statistical results. 16 Taha AS, Angerson WJ, Knill-Jones RP, Blatchford O. Upper All authors have seen and approved the fi nal version of the report.
gastrointestinal mucosal abnormalities and blood loss complicating low-dose aspirin and anti-thrombotic drugs. Aliment Pharmacol Ther Confl icts of interest
2006; 23: 489–95.
AST previously received research grants and travel expenses from 17 Taha AS, Hudson N, Hawkey CJ, et al. Famotidine for the Astellas, AstraZeneca, Merck Laboratories, and Yamanouchi. AST also prevention of gastric and duodenal ulcers caused by nonsteroidal sat on advisory panels for Ferring and Shire Pharamaceutical. All other antiinfl ammatory drugs. N Engl J Med 1996; 334: 1435–39.
authors declare that they have no confl icts of interest.
18 Hudson N, Taha AS, Russell RI, et al. Famotidine for healing of and maintenance in nonsteroidal anti-infl ammatory drugs-associated Acknowledgments
gastroduodenal ulceration. Gastroenterology 1997; 112: 1817–22.
The study drugs (famotidine and placebo) were donated by Merck 19 Taha AS, Dahill S, Morran C, et al. Neutrophils, Helicobacter pylori, Laboratories, UK. This study was sponsored by Astellas Pharma Inc, and nonsteroidal anti-infl ammatory drug ulcers. Gastroenterology Japan. The data and safety monitoring committee included Ian Ford 1999; 116: 254–59.
(Robertson Centre for Biostatistics, University of Glasgow, Glasgow, 20 Yeomans ND, Naesdal J. Systematic review: ulcer defi nition in NSAID UK), Peter R Mills (Gartnavel General Hospital, Glasgow, and University ulcer prevention trials. Aliment Pharmacol Ther 2008; 27: 465–72.
of Glasgow); and Tom Macdonald (University of Dundee, Dundee, UK).
21 Lanza FL, Graham DY, Davis RE, Rack MF. Endoscopic evaluation References
of cimetidine and sucralfate for prevention of naproxen-induced CAPRIE Steering Committee. A randomised, blinded trial of gastroduodenal injury: eff ects of scoring method. Dig Dis Sci 1990; clopidogrel versus aspirin in patients at risk of ischaemic events 35: 1494–99.
(CAPRIE). Lancet 1996; 348: 1329–39.
22 Lundell LR, Dent J, Bennett JR, et al. Endoscopic assessment of Diener HC, Bogousslavsky J, Brass LM, et al. Aspirin and oesophagitis: clinical and functional correlates and further clopidogrel compared with clopidogrel alone after recent ischaemic validation of the Los Angeles classifi cation. Gut 1999; 45: 172–80.
stroke or transient ischaemic attack in high-risk patients (MATCH): 23 Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of randomised, double-blind, placebo-controlled trial. Lancet 2004; classifying prognostic comorbidity in longitudinal studies: 364: 331–37.
development and validation. J Chronic Dis 1987; 40: 373–83.
Belch J, Macuish A, Campbell I, et al. The prevention of 24 Laine L, Maller ES, Yu C, Quan H, Simon T. Ulcer formation with
progression of arterial disease and diabetes (POPAD) trial: factorial low-dose enteric-coated aspirin and the eff ect of COX-2 selective randomised placebo controlled trial of aspirin and antioxidants in inhibition: a double-blind trial. Gastroenterology 2004; 127: 395–402.
patients with diabetes and asymptomatic peripheral arterial disease. 25 Niv Y, Battler A, Abuksis G, et al. Endoscopy in asymptomatic BMJ 2008; 337: 1030–34.
minidose aspirin consumers. Dig Dis Sci 2005; 50: 78–80.
Taha AS, Angerson WJ, Knill-Jones RP, Blatchford O. Clinical 26 Lan KKG, DeMets DL. Discrete sequential boundaries for clinical outcome in upper gastrointestinal bleeding complicating low-dose trials. Biometrika 1983; 70: 659–63.
aspirin and antithrombotic drugs. Aliment Pharmacol Ther 2006; 27 O’Brien PC, Fleming TR. A multiple testing procedure for clinical 24: 633–36.
trials. Biometrics 1979; 35: 549–56.
Taha AS, Angerson WJ, Prasad R, McCloskey C, Blatchford O. 28 Daneshmend TK, Prichard PJ, Bhaskar NK, Millns PJ, Hawkey CJ. Upper gastrointestinal bleeding and the changing use of COX-2 Use of microbleeding and an ultrathin microscope to assess non-steroidal anti-infl ammatory drugs and low-dose aspirin. gastric mucosal protection by famotidine. Gastroenterology 1989; Aliment Pharmacol Ther 2007; 26: 1171–78.
97: 944–49.
Taha AS, Angerson WJ, Prasad R, McCloskey C, Gilmour D, 29 FKL Chan. Proton-pump inhibitors in peptic ulcer disease. Lancet Morran CG. The incidence and early mortality after peptic ulcer 2008; 372: 1198–200.
perforation, and the use of low-dose aspirin and nonsteroidal
anti-infl ammatory drugs. Aliment Pharmacol Ther 2008;
28: 878–85.
www.thelancet.com Vol 374 July 11, 2009

Source: http://lab420.com/files/Lancet_famotidine_2009.pdf