 |
 |
 |
 |
 |
 |
 |
|
|
|
|||||||||
|
|||||||||||
PharmD Student, Faculty of Pharmaceutical Sciences, University of British Columbia, Vancouver, British Columbia, Canada
Clinical Practice Leader, Regional Pharmacy Services, Capital Health, Edmonton, Alberta, Canada
Reprints: Dr. Ackman, Regional Pharmacy Services, Capital Health, 0G1.01 WMC 8440-112 St., Edmonton, AL, Canada T6G 2B7, fax 780/407-7690, margaret.ackman{at}capitalhealth.ca
 |
Abstract |
|---|
 Top Abstract RequestResponseSummaryReferences |
|---|
DATA SOURCES: A literature search was conducted using EMBASE (1980–January 2008), PubMed (1966–January 2008), Google, and a manual search of the reference lists using the search terms gastrointestinal bleed, gastrointestinal hemorrhage, peptic ulcer hemorrhage, ASA, aspirin, Plavix, clopidogrel, and PPI. The search, limited to human and English studies, yielded 110 returns.
STUDY SELECTION AND DATA EXTRACTION: Randomized trials that compared aspirin with clopidogrel, involved patients who had previously experienced a GIB, and provided detailed information on the type and dose of drugs used were included. Studies were required to provide information on the recurrence of GIB.
DATA SYNTHESIS: Two randomized trials were reviewed to assess the safety of secondary prevention of cardiovascular disease with respect to previous GIB. These noninferiority trials compared aspirin plus a PPI with clopidogrel over 12 months following confirmed healing of an aspirin-induced ulcer. In both trials, the majority of the GIB recurrences were in the clopidogrel group (8.6% vs 0.7%; difference 7.9%; 95% CI 3.4 to 12.4; p = 0.001 and 13.6% vs 0%; difference 13.6%; 95% CI 6.3 to 20.9; p = 0.0019) and the difference in recurrence rates exceeded the a priori selected upper boundary.
CONCLUSIONS: Findings reported in the limited literature available support that clopidogrel is not equivalent to the combination of aspirin plus a PPI in the patient population studied. Aspirin plus a PPI would be considered clinically superior and should be used in medically managed patients who require single antiplatelet therapy but have had a prior GIB while on aspirin. Further research regarding dual antiplatelet therapy and a PPI is required.
Key Words: aspirin, clopidogrel, gastrointestinal bleed, proton pump inhibitor
Published Online, April 15, 2008. www.theannals.com, DOI 10.1345/aph.1K566
|
Request |
|---|
|
TopAbstractRequestResponseSummaryReferences |
|---|
|
Response |
|---|
|
TopAbstractRequestResponseSummaryReferences |
|---|
Antiplatelet therapy for secondary prevention carries certain risks. A meta-analysis conducted by Derry et al.3 reported rates of 2.47% for any GIB in 65,987 patients taking aspirin for primary and secondary prevention, compared with 1.42% with placebo over 28 months (OR 1.68; 95% CI 1.51 to 1.88). Even when only trials using aspirin doses of 50–162.5 mg/day were analyzed, gastrointestinal hemorrhage still occurred in 2.3% of patients taking aspirin compared with 1.45% of those taking placebo (OR 1.59; 95% CI 1.40 to 1.81). A meta-analysis in a similar population used a more stringent definition of GIB and calculated a number-needed-to-harm (NNH) of 883 for low-dose aspirin over a one-year period.4 This study showed that 883 people would have to be treated with aspirin rather than no antiplatelet therapy for one year for one extra person in the aspirin-treated group to experience a GIB. Overall, the majority of studies found a twofold increase in the risk of major GIB with low-dose aspirin.5 Despite this significant increase in relative risk, it is important to remember that the increase in absolute risk in most patient populations is small, as reflected by the NNH.
The CAPRIE study showed a statistically significant decrease in GI hemorrhage in patients receiving clopidogrel (1.99%) compared with patients receiving aspirin 325 mg (2.66%).2 This represents an NNH of 149, so the clinical significance of this difference is modest. In addition, the use of aspirin 325 mg as the comparator, rather than a lower dose, should be considered.
Observational data support a similar GIB risk between clopidogrel and aspirin. A case–control study determined a relative risk of 2.8 with thienopyridines, which was similar to the 2.7 reported for aspirin 100 mg.6 A cohort study of high-risk patients, such as those with non–aspirin-related peptic ulcer disease or a history of aspirin-related GI complications, demonstrated a GIB recurrence rate of 14% with clopidogrel.7
Neither the current American College of Cardiology/American Heart Association (ACC/AHA) guidelines for stable angina/coronary artery disease8 nor the guidelines of the American College of Chest Physicians (ACCP) from 20049 provide specific recommendations for patients with a history of GIB. However, in the recently released ACC/AHA Non–ST-Segment Elevation Myocardial Infarction guidelines, concomitant drug therapy (eg, PPI) is recommended in patients with a history of GIB to minimize the risk of recurrent bleeding when aspirin and clopidogrel are administered alone or in combination.10 In contrast, the ACCP recommendation for post-MI and postacute coronary syndromes (ACS) patients with a history of aspirin-induced bleeding or with risk factors for bleeding is aspirin at doses of less than or equal to 100 mg daily.9
The purpose of this paper is to review, from a GIB risk perspective, the secondary prevention options of using aspirin and a PPI compared with clopidogrel in patients who have experienced a GIB while on aspirin.
LITERATURE REVIEW
A literature search using EMBASE (1980–January 2008), PubMed
(1966–January 2008), Google, and a manual search of the reference lists
with the search terms gastrointestinal bleed, gastrointestinal hemorrhage,
peptic ulcer hemorrhage, ASA, aspirin, Plavix, clopidogrel, and proton pump
inhibitor, limited to human and English studies, yielded 110 returns.
The randomized trials reviewed here compared the use of aspirin with clopidogrel in patients who had previously experienced a GIB and provided detailed information on the type and dose of drugs used, as well as the characteristics of the patients included. Studies were required to provide information on the recurrence of GIB. Interventions involved the administration of PPIs with the goal of preventing GIB. Studies that excluded aspirin or clopidogrel from either group were not considered. Two trials were identified and are reviewed.11,12
Chan et al.11 conducted a randomized, double-blind, controlled noninferiority trial of patients who took aspirin 325 mg or less for prevention of vascular diseases and presented with upper GIB during therapy. The 320 patients included in this trial were predominantly male (66%) and, on average, 72 years of age. Indication for aspirin was coronary heart disease (52%), cerebrovascular insufficiency (38%), and peripheral vascular disease (4%) for the population overall, with 6% having multiple ischemic conditions. Aspirin was withheld during the eradication period and patients received a PPI plus Helicobacter pylori treatment as appropriate. Forty-six percent of the patients had previous H. pylori infection.
Following endoscopically confirmed ulcer healing for all patients and successful eradication of H. pylori, patients were randomized to receive either aspirin 80 mg daily with esomeprazole 20 mg twice daily (n = 159) or clopidogrel 75 mg daily with matching esomeprazole placebo twice daily (n = 161). The primary outcome was adjudicated recurrent ulcer bleeding within 12 months, defined as hematemesis or melena with endoscopically confirmed ulcers or bleeding erosions, or a drop in hemoglobin of at least 2 g/dL with endoscopically confirmed ulcers or bleeding erosions. Only events that occurred during treatment or within 28 days after discontinuation were included in the analysis.11
The median follow-up was 12 months and was complete for 99% of the patients. The investigators selected a 4% absolute difference as the upper boundary of the 95% confidence interval for the difference in recurrent bleeding rates. Fourteen cases of recurrent ulcer bleeding were identified by the adjudication committee and none of these patients had a recurrent H. pylori infection. Thirteen of these were in the clopidogrel group, resulting in a cumulative incidence of 8.6% compared with 0.7% for patients receiving aspirin plus esomeprazole. The absolute difference in recurrence rates was 7.9% (95% CI 3.4 to 12.4; p = 0.001), exceeding the 4% a priori selected upper boundary. A per protocol analysis found a similar difference of 6.8% (2.3 to 11.3; p = 0.005).11 Therefore, clopidogrel was not shown to be equivalent to the combination of aspirin and esomeprazole. In fact, given the differences in recurrence rates, aspirin plus esomeprazole would be considered clinically superior to clopidogrel monotherapy from the perspective of the prevention of GIBs.
Unfortunately, this trial did not report use at baseline of other antithrombotic agents or nonsteroidal antiinflammatory drugs (NSAIDs) that may have contributed to the initial GIB. They did, however, report that 2 of the patients with recurrent bleeding were receiving clopidogrel and reported using concomitant NSAIDs.11 The 80-mg dose of aspirin may have been a lower dose than that previously used by some patients; that alone may have resulted in a reduction in GIBs. This limitation is based on indirect comparisons of doses, ranging from less than 75 mg/day to 1500 mg/day, that have shown dose-dependent increases in bleeding.13 The generalizability of this study may be limited by its predominantly male and Asian population.
Lai et al.12 conducted a study of similar design, with an a priori boundary of 5% to demonstrate noninferiority of clopidogrel. Patients who presented with an upper GIB while receiving aspirin at a dose of less than 325 mg daily and who had endoscopically confirmed gastric and/or duodenal ulcer were enrolled. Patients with H. pylori were treated with a 1-week course of H. pylori therapy followed by 5 weeks of famotidine. All other patients were treated with famotidine for 6 weeks. Treatments were repeated or extended as required to achieve endoscopically confirmed healing. Patients were then randomized to receive 52 weeks of aspirin 100 mg plus esomeprazole 20 mg daily (n = 86) or clopidogrel 75 mg daily (n = 84), which were administered in a blinded fashion. These patients were predominantly male (60%), with a mean age of 76 years and 53% with initial H. pylori. Patients' indications for antiplatelet therapy were ischemic heart disease (52%), stroke/transient ischemic attack (45%), and peripheral vascular disease (3%). The majority (82%) of patients had been receiving between 100 and 150 mg of aspirin daily and the duration of previous aspirin use ranged from 2 weeks to 6 years.
Endoscopy was performed for evidence of GIB and ulcer complications, dyspepsia not responsive to antacids, a decrease in hemoglobin of more than 2 g/dL, or stool positive for occult blood. Events that occurred during treatment or within 14 days of treatment discontinuation were included in the analysis. Two interim analyses were planned and the study was stopped following the second analysis.12
Three patients did not attend follow-up evaluations. Nine GI events were adjudicated as recurrence of ulcer complications. All were in the clopidogrel group, resulting in an annual recurrence rate of 13.6%. H. pylori relapse was reported in one patient and concomitant NSAID use in another. The cumulative incidence of recurrent ulcers was 0% with aspirin plus esomeprazole and 13.6% in the clopidogrel group. This difference of 13.6% (95% CI 6.3 to 20.9; p = 0.0019) exceeded the 5% upper boundary established, thereby disproving equivalence.12 Again, from a clinical perspective, aspirin plus esomeprazole would be considered superior to clopidogrel when considering the risk of GIB.
This study is limited by the 100-mg dose of aspirin used after randomization, as most patients were using doses greater than 100 mg at baseline. This may bias the results in favor of aspirin. The generalizability of this study may be limited by its predominantly male and Asian population.12
Both of these studies are limited by their lack of generalizability to the current management of coronary artery disease, more specifically ACS. The use of long-term dual antiplatelet therapy has increased substantially with the increasing use of drug-eluting stents. A larger proportion of patients being medically managed for ACS now also receive dual antiplatelet therapy. An entry criterion for both of the studies reviewed here was anticipated regular use of antiplatelet therapy, and only slightly more than half of the patients had coronary artery disease as their ischemic condition. Only Lai et al.12 reported the duration of aspirin use prior to the study; less than 5% of that study's participants had been using aspirin for less than 4 weeks. Three-quarters of the subjects had been on aspirin for more than one year prior to study entry. The trials excluded patients requiring concomitant therapy with other antiplatelet medications, although less than 3% of all patients screened were excluded due to concomitant therapy with warfarin, NSAIDs, or other antiplatelet agents. Therefore, the relevant portion of the study population represents patients with stable coronary artery disease but without a current requirement for dual antiplatelet therapy.
In patients with a previous GIB, the studies reviewed here support use of aspirin with a PPI rather than clopidogrel alone to reduce GIB risk in patients requiring single antiplatelet therapy. However, theoretically, the efficacy of aspirin could be compromised by an acid-suppressing agent. Aspirin is a weak acid that crosses mucosa in its nonionized state. When acid production is suppressed, aspirin becomes ionized, and thus its absorption and bioavailability may be reduced. The findings addressing this area are controversial due to conflicting results.14,15
Two trials of healthy volunteers used a crossover design to examine the impact of omeprazole on salicylic acid concentrations.14,15 Both studies examined levels following administration of single doses of aspirin with and without a preceding course of omeprazole. Washout periods of 15 and 21 days were applied. The first study (n = 11) administered aspirin 650 mg and found significantly lower concentrations at 30 (p = 0.02), 60 (p = 0.01), and 90 (p = 0.003) minutes following a 2-day course of omeprazole.14 These results contrast with those of the second study (n = 14) that used aspirin 125 mg and a 4-day course of omeprazole.15 This trial found no significant differences in plasma concentrations of aspirin or salicylic acid at time points up to 5 hours or any difference in the area under the curve. This study also examined the biological effects of aspirin and reported that omeprazole pretreatment was not found to impact skin bleeding time or platelet aggregation assays. Although it is difficult to correlate these results with clinical outcomes, they provide some reassurance regarding the potential impact of omeprazole. Currently, it is unclear whether there is a clinically relevant impact of the addition of a PPI to aspirin therapy.
|
Summary |
|---|
|
TopAbstractRequestResponseSummaryReferences |
|---|
For patients with a history of GIB who require single antiplatelet therapy, aspirin 80–100 mg/day in combination with esomeprazole 20 mg once daily should be recommended. Given the common use of dual antiplatelet therapy, further research regarding the best strategies for prevention of GIBs in patients requiring dual antiplatelet therapy is important. For patients with a previous GIB, although the addition of a PPI to dual antiplatelet therapy is likely to reduce the risk of recurrent hemorrhage, the risks and benefits of therapy must be assessed for each patient.
|
References |
|---|
|
TopAbstractRequestResponseSummaryReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
 |
 |
 |
 |
