HomeCirculationVol. 118, No. 18ACCF/ACG/AHA 2008 Expert Consensus Document on Reducing the Gastrointestinal Risks of Antiplatelet Therapy and NSAID Use Free AccessReview ArticlePDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessReview ArticlePDF/EPUBACCF/ACG/AHA 2008 Expert Consensus Document on Reducing the Gastrointestinal Risks of Antiplatelet Therapy and NSAID UseA Report of the American College of Cardiology Foundation Task Force on Clinical Expert Consensus Documents Writing Committee Members Deepak L. Bhatt, James Scheiman, Neena S. Abraham, Elliott M. Antman, Francis K.L. Chan, Curt D. Furberg, David A. Johnson, Kenneth W. Mahaffey and Eamonn M. Quigley Writing Committee Members Search for more papers by this author , Deepak L. BhattDeepak L. Bhatt Search for more papers by this author , James ScheimanJames Scheiman Search for more papers by this author , Neena S. AbrahamNeena S. Abraham Search for more papers by this author , Elliott M. AntmanElliott M. Antman Search for more papers by this author , Francis K.L. ChanFrancis K.L. Chan Search for more papers by this author , Curt D. FurbergCurt D. Furberg Search for more papers by this author , David A. JohnsonDavid A. Johnson Search for more papers by this author , Kenneth W. MahaffeyKenneth W. Mahaffey Search for more papers by this author and Eamonn M. QuigleyEamonn M. Quigley Search for more papers by this author Originally published3 Oct 2008https://doi.org/10.1161/CIRCULATIONAHA.108.191087Circulation. 2008;118:1894–1909is corrected byCorrectionOther version(s) of this articleYou are viewing the most recent version of this article. Previous versions: October 3, 2008: Previous Version 1 Preamble…1895Introduction…1895Prevalence of Use—NSAIDs/Aspirin (ASA)…1895Mechanisms of GI Injury—NSAIDs…1895Mechanisms of GastroduodenalInjury—Clopidogrel…1896 GI Complications of ASA and Non-ASA NSAIDs…1896 GI Effects of ASA…1897 GI Effects of Combined ASA and Anticoagulant Therapy…1898 GI Effects of Clopidogrel…1898 GI Effects of Combined Clopidogrel and Anticoagulant Therapy…1899 Treatment and Prevention of ASA- and NSAID-Related Gastroduodenal Injury…1899 Role of H pylori…1901 A. Diagnosis of H pylori…1901 B. Tests for Active H pylori…1901 C. Treatment of H pylori…1902 Discontinuation of Antiplatelet Therapy Because of Bleeding…1902 Endoscopy in Patients on Mono- or Dual Antiplatelet Therapy…1902Summary…1903References…1903Appendix 1…1906Appendix 2…1908PreambleThis document has been developed by the American College of Cardiology Foundation (ACCF) Task Force on Clinical Expert Consensus Documents, the American College of Gastroenterology (ACG), and the American Heart Association (AHA). Expert consensus documents (ECDs) are intended to inform practitioners, payers, and other interested parties of the opinion of the ACCF and document cosponsors concerning evolving areas of clinical practice and/or technologies that are widely available or new to the practice community. Topics chosen for coverage by ECDs are so designed because the evidence base, the experience with technology, and/or the clinical practice are not considered sufficiently well developed to be evaluated by the formal American College of Cardiology/American Heart Association (ACC/AHA) practice guidelines process. Often the topic is the subject of ongoing investigation. Thus, the reader should view ECDs as the best attempt of the ACCF and other cosponsors to inform and guide clinical practice in areas where rigorous evidence may not be available or the evidence to date is not widely accepted. When feasible, ECDs include indications or contraindications. Topics covered by ECDs may be addressed subsequently by the ACC/AHA Practice Guidelines Committee as new evidence evolves and is evaluated.The Task Force on ECDs makes every effort to avoid any actual or potential conflicts of interest that might arise as a result of an outside relationship or personal interest of a member of the writing panel. Specifically, all members of the writing panel are asked to provide disclosure statements of all such relationships that might be perceived as real or potential conflicts of interest to inform the writing effort. These statements are reviewed by the parent task force, reported orally to all members of the writing panel at the first meeting, and updated as changes occur. The relationships with industry information for writing committee members and peer reviewers are listed in Appendixes 1 and 2, respectively.Robert A. Harrington, MD, FACC Chair, ACCF Task Force on Clinical Expert Consensus DocumentsIntroductionThe use of antiplatelet therapies continues to increase as a result of accumulation of evidence of benefits in both primary and secondary treatment strategies for cardiovascular disease.1,2 These antiplatelet agents, however, have recognizable risks—in particular, gastrointestinal (GI) complications such as ulceration and related bleeding. These risks may be further compounded by the ancillary use of other adjunctive medications, such as nonsteroidal anti-inflammatory drugs (NSAIDs), corticosteroids, and anticoagulants. Given the high prevalence of antiplatelet therapy in clinical practice, coupled with an increased emphasis on their extended use, especially after implantation of a drug-eluting stent,3,4 it is imperative that physicians know the potential benefits and the associated risks of antiplatelet therapy for primary or secondary prevention of cardiac ischemic events when combined with NSAID agents. Only with this understanding can physicians appropriately and fully evaluate the risk profile for each patient and either change medications or initiate prophylactic therapy in an attempt to reduce GI complications. This document provides consensus recommendations from the ACCF, the AHA, and the ACG on the combined use of antiplatelets and NSAID agents.Many NSAIDs, both selective and nonselective, increase the risk of cardiovascular and cerebrovascular events. This issue was addressed in ascientific statement from the AHA.5 In terms of cardiovascular, GI, renal, and hypertension-inducing risks, there are important differences among the NSAIDs (especially the cyclo-oxygenase-2 [COX-2] inhibitors), which should also be understood and considered in managing patients in need of these agents.6 The AHA statement introduces a stepped-care approach for selection of drugs to manage musculoskeletal discomfort in patients with known cardiovascular disease or risk factors for ischemic heart disease, based on the risk/benefit balance from a cardiovascular perspective. A further discussion of the cardiovascular and cerebrovascular risks of NSAIDs is beyond the scope of this report but may be found in several reviews.5,7Prevalence of Use—NSAIDs/Aspirin (ASA)The use of NSAIDs, including ASA, is common in the treatment of pain, inflammation, and fever. Additionally, low-dose ASA is used routinely in primary and secondary prophylaxis of cardiovascular and cerebrovascular events. These agents, both through prescription and over-the-counter (OTC) use, are the most widely used class of medications in the United States.8 Not surprisingly, NSAID use increases among the elderly. In a survey of people 65 years of age and older, 70% used NSAIDs at least once weekly, and 34% used them at least daily. The prevalence of at least weekly ASA usage was 60%.9 More than 111 million NSAID prescriptions were written in 2004.10Recognizably, much of this usage comes from noncardiac indications, such as arthritis and related musculoskeletal complaints, in particular. In 1990, the estimated prevalence of self-reported arthritis in the United States was 37.9 million cases, or 15% of the population. By 2020, it is projected that 59.4 million will be affected—a 57% increase from 1990.11 As the incidence of arthritis complaints increases, the use of prescription and OTC NSAIDs is also expected to increase.Mechanisms of GI Injury—NSAIDsA complete discussion of the pathogenesis of ASA- and NSAID-associated injury is beyond the scope of this article; however, ASA, like all NSAIDs, injures the gut by causing topical injury to the mucosa and systemic effects induced by prostaglandin depletion. Tissue prostaglandins are produced via 2 pathways: a COX-1 and a COX-2 pathway. The COX-1 pathway is the predominant constitutive pathway; prostaglandins derived from this enzyme mediate many effects, most notably facilitating gastroduodenal cytoprotection, renal perfusion, and platelet activity. The COX-2 pathway, in contrast, is inducible by inflammatory stimuli and mediates effects through prostaglandins, which result in inflammation, pain, and fever.Inhibition of the COX-1 pathway blocks production of prostaglandins that play an important protective role in the stomach by increasing mucosal blood flow and stimulating the synthesis and secretion of mucus and bicarbonate, as well as promoting epithelial proliferation. Accordingly, the inhibition of these prostaglandins impairs these protective factors, resulting in a gastric environment that is more susceptible to topical attack by endogenous factors, such as acid, pepsin, and bile salts.12 A major consequence of prostaglandin depletion is to create an environment that is conducive to peptic ulcer formation and serious GI complications. Since prostaglandins are essential to both the maintenance of intact GI defenses and normal platelet function, nonselective NSAIDs such as ASA promote ulcer formation as well as bleeding.13Because COX-2 is the primary intended target for anti-inflammatory drug therapy, agents that selectively block COX-2, while having little to no effect on COX-1, should result in effective pain relief with reduced GI toxicity. This concept, called the “COX-2 hypothesis,” has been challenged by data from animal studies, which indicated that both COX-1 and COX-2 must be inhibited for gastric ulceration to occur. Interestingly, while the selective inhibition of either COX-1 or COX-2 alone failed to cause gastric damage, inhibition of both COX isoforms produced gastric ulceration.14 Thus, the explanation for reduced GI toxicity for COX-2–specific inhibitors may be their lack of dual COX inhibition rather than their COX-1–sparing effects.In this framework, taking both a cardioprotective dose of ASA (primarily a COX-1 inhibitor at low dose [ie, 325 mg or less]) and a COX-2 inhibitor creates the ulcer risk of a traditional NSAID. A high percentage of individuals requiring cardioprotective doses of ASA have chronic pain and receive a traditional NSAID or a COX-2–selective NSAID (coxib). A survey that queried chronic coxib users found that 50% or more users were also taking ASA.15 Moreover, because coxibs were heralded as having an improved safety profile, related primarily to a lower rate of GI toxicity than traditional NSAIDs, the potential loss of this safety advantage when a COX-2 inhibitor is combined with ASA or an OTC NSAID remains underappreciated by clinicians. Heightened attention to the cardiovascular risks of NSAIDs has likely further increased the rate of addition of ASA to anti-inflammatory therapy.16Mechanisms of Gastroduodenal Injury—ClopidogrelPlatelet aggregation plays a critical role in healing through the release of various platelet-derived growth factors that promote angiogenesis. Angiogenesis, in turn, is critical for the repair of GI mucosal disruptions. Experimental animals with thrombocytopenia have been shown to have reduced ulcer angiogenesis and impaired ulcer healing.17 Additionally, adenosine diphosphate-receptor antagonists impair the healing of gastric ulcers by inhibiting platelet release of pro-angiogenic growth factors, such as vascular endothelial growth factor, which promotes endothelial proliferation and accelerates the healing of ulcers. GI bleeding is also a major toxic effect of chemotherapeutic agents that use monoclonal antibodies directed at circulating vascular endothelial growth factor.18 Although clopidogrel and other agents that impair angiogenesis may not be a primary cause of gastroduodenal ulcers, their anti-angiogenic effects may impair healing of gastric erosions or small ulcerations that develop because of other medications or Helicobacter pylori infection. This may then, in the presence of acid, lead to clinically significant ulceration and related complications.1. GI Complications of ASA and Non-ASA NSAIDsRecommendation: As the use of any NSAID, including COX-2–selective agents and OTC doses of traditional NSAIDs, in conjunction with cardiac-dose ASA, substantially increases the risk of ulcer complications, a gastroprotective therapy should be prescribed for at-risk patients.Upper gastrointestinal events (UGIE), symptomatic or complicated ulcers, occur in 1 of every 20 NSAID users and in 1 of 7 older adults using NSAIDs,19 accounting for 30% of UGIE-related hospitalizations and deaths.20–22 Dyspepsia, defined as upper abdominal pain or discomfort, may occur in individuals taking NSAIDs, including ASA. Dyspepsia is not clearly predictive of the presence of an ulcer, as it is far more prevalent. Some patients may also experience an increase in symptoms of gastroesophageal reflux disease on NSAIDs as well.23 Endoscopic ulcers are used as a surrogate marker in clinical trials for risk of medications and in treatment trials; this document focuses on patients with dyspepsia and an ulcer (symptomatic ulcer) or those with serious (life threatening) ulcer complications such as bleeding or perforation. The annual incidence of NSAID-related UGIE is 2.0% to 4.5%,19 and the risk of bleeding, perforation, or obstruction is 0.2% to 1.9%.19,24 NSAIDs contribute to 10 to 20/1000 hospitalizations per year and are associated with a 4-fold increase in mortality.20 In the United States alone, NSAID use has been extrapolated to account for approximately 107000 hospitalizations and 16 500 deaths per year among patients with arthritis.25 More recent information regarding these estimates related to NSAIDs suggests that these numbers may be too high, but increasing use of antiplatelet medications may contribute to an increased burden of GI bleeding.26–28 According to these reports, GI hospitalization rates markedly declined (from 1.5% to 0.5%) between 1992 and 2000. Four potential explanations were given: use of lower doses of NSAIDs, less use of “more toxic” NSAIDs, increased use of “safer” NSAIDs, and increased use of proton pump inhibitors (PPIs).Among elderly veterans, NSAID exposure has been shown to increase risk of UGIE-related mortality3-fold, even after adjustment for advancing age, comorbidity, and proportion of time spent on a traditional or COX-2–selective NSAID.26 In fact, if deaths resulting from NSAID-associated upper GI complications were tabulated separately, it would represent the 15th most common cause of death in the United States.29 National data from the Department of Veterans Affairs reveal that 43.0% of the veterans prescribed NSAIDs are considered to be at high risk for UGIE and that patients 65 years or older constitute the largest high-risk subset (87.1%).8 Among elderly veterans, the risk of NSAID-related UGIE has been estimated as 2753 UGIE in 220662 person-years of follow-up.30Those who combine an NSAID with ASA represent another high-risk group. When patients combine an NSAID with ASA, the annual risk of UGIE is 5.6%, with coxibs providing no additional gastroprotection (7.5% UGIE/year). A number of observational studies have noted a 2- to 4-fold increased risk of UGIE associated with the concomitant prescription of NSAIDs with low-dose ASA. Data from Scandinavia indicated an annual incidence of hospital admission for UGIE of 1.4% related to use of NSAIDs plus low-dose ASA versus 0.6% for low-dose ASA. Estimates of the relative risk (RR) of UGIE for NSAID plus ASA range from 3.8 (95% confidence interval [CI]: 1.8 to 7.8)14 to 5.6 (95% CI: 4.4 to 7.0) when compared with ASA alone.30Endoscopic trials suggest that the GI toxicity of a coxib plus ASA is additive, resulting in an overall risk of endoscopic ulcer formation that parallels that seen with a nonselective NSAID.25,31 Additionally, evidence from observational studies and randomized controlled trials (RCTs) reveals that the risk of an NSAID plus ASA exceeds that of a coxib plus ASA, although both were markedly increased by ASA.9,27,29 In this context, whether one chooses a nonselective NSAID or a selective COX-2 inhibitor has a minimal, and perhaps clinically insignificant, impact on the likelihood of serious adverse GI outcomes. Thus, the selection of anti-inflammatory drug therapy in such patients must involve consideration of overall GI and cardiovascular risk of NSAIDs.32 The ongoing PRECISION (Prospective Randomized Evaluation of Celecoxib Integrated Safety vs Ibuprofen or Naproxen; NCT00346216) study, which is randomizing arthritis patients with or at risk of cardiovascular disease to ibuprofen, naproxen, or celecoxib, should provide more data to help clarify these issues.2. GI Effects of ASARecommendation: The use of low-dose ASA for cardioprophylaxis is associated with a 2- to 4-fold increase in UGIE risk. Enteric-coated or buffered preparations do not reduce the risk of bleeding. For patients at risk of adverse events, gastroprotection should be prescribed. The risk of UGIE increases with ASA dose escalation; thus, for the chronic phase of therapy, doses greater than 81 mg should not be routinely prescribed.The AHA recommends low-dose ASA use among patients with a 10-year cardiovascular risk that is greater than or equal to 10%,33,34 and the US Preventive Services Task Force recommends ASA cardioprophylaxis for patients with a 5-year risk of greater than or equal to 3%.35 It has been estimated that 50 million Americans use low-dose ASA (ie, 325 mg/day or less) regularly for cardioprophylaxis.36 The use of low-dose ASA is associated with a 2- to 4-fold increased risk of UGIE,37,38 which is not reduced by the use of buffered or enteric-coated preparations.39,40 Fourteen randomized placebo-controlled trials have presented data on UGIE with cardiac-dose ASA (75 to 325 mg per day) in adults. When these data are pooled, the absolute increased risk per year of UGIE with ASA is 0.12% when compared with placebo (number needed to harm=833), with conflicting evidence of risk reduction with lower doses (75 to 162.5 mg) versus higher doses (greater than 162.5 to 325 mg).41The estimated average excess risk of UGIE related to cardioprophylactic doses of ASA is 5 cases per 1000 ASA users per year.42 Among elderly patients, the odds ratios (ORs) of bleeding with daily doses of ASA of 75, 150, and 300 mg are 2.3, 3.2, and 3.9, respectively.37 Dose reduction does not appear to reduce antithrombotic benefits; however, dose escalation does seem to increase bleeding complications.43 Additionally, case series implicate OTC use of low-dose ASA in over one third of the patients admitted for GI hemorrhage,44 suggesting that patients who self-medicate may be unaware of the significant increase in their risk of UGIE.The complexities of confirming a significant difference across the range of the low doses of ASA used for cardioprotection are discussed below. Meta-analyses have been contradictory in demonstrating a significant difference in the risk of GI bleeding.45,46 Observational studies are somewhat contradictory, supporting evidence of a trend for an association between higher ASA dose and risk of upper GI complications.37,47 The ACC and AHA recommend lowering the dose from 325 to 81 mg among those with a high risk of UGIE.2 However, some experts feel it may be prudent to use up to 325 mg a day of ASA for 1 month after a stent procedure, although it is not clear from the data whether this dose is really necessary.2 While this low-dose ASA approach makes sense intuitively because of the lack of demonstrated additional cardiovascular benefits at the higher dose (with certain limited exceptions, such as acute coronary syndrome [ACS]), coupled with a likelihood of increased risk of GI harm at the higher dose, the key point is that the benefit, in terms of GI bleeding risk reduction with the lower dose, remains insufficient to protect high-risk patients and mandates the addition of other GI bleeding risk-reduction approaches. However, it is unknown what the optimal dose of ASA really is. The Antithrombotic Trialists' Collaboration meta-analysis provides indirect evidence that higher doses of ASA are not more effective, at least at a population level.48 There are observational data from the CURE (Clopidogrel in unstable angina to prevent recurrent events) trial that suggest no benefit from higher doses of ASA but a greater risk of bleeding.49 The CURRENT/OASIS-7 (Clopidogrel Optimal Loading Dose Usage to Reduce Recurrent EveNTs/Optimal Antiplatelet Strategy for InterventionS-7; NCT00335452) trial is randomizing ACS patients to higher (300 to 325 mg) or lower (75 to 100 mg) ASA doses in the range used for cardiovascular disease and may help to clarify this issue once the results are known.The use of enteric-coated or buffered formulations does not appear to reduce the risk of GI bleeding complications,39,40,50 a finding that suggests that the upper GI side effects of ASA are a result of a systemic effect, in addition to its potent topical action to induce chemical injury. Anecdotal reports of reduced dyspepsia with these products likely contribute to their uptake in practice.51While the risk factors for NSAID-related UGIEs have been well characterized, there are much less data on the risk of antiplatelet therapy. The synergism between ASA and NSAIDs was reviewed in detail in the previous section. A history of peptic ulcer, particularly with associated bleeding, appears to be the most important risk factor. Age is an important risk factor as well, with the relative increase beginning at age 60 years and rising in a nonlinear fashion with age. Gender is a less important concern, although the risk of men is slightly higher than that of women.42 The risk associated with combination antiplatelet and anticoagulant therapies is substantial as well, and each is discussed below given their importance in cardiology clinical practice.3. GI Effects of Combined ASA and Anticoagulant TherapyRecommendation: The combination of ASA and anticoagulant therapy (including unfractionated heparin, low-molecular-weight heparin, and warfarin) is associated with a clinically meaningful and significantly increased risk of major extracranial bleeding events, a large proportion from the upper GI tract. This combination should be used with established vascular, arrhythmic, or valvular indication; patients should receive concomitant PPIs as well. When warfarin is added to ASA plus clopidogrel, an international normalized ratio (INR) of 2.0 to 2.5 is recommended.52The use of antiplatelet drugs for the initial management of ACS is common and known to be effective.1,2 In some clinical settings, such as the initial and long-term management of ACS, the combination of anticoagulant and antiplatelet therapy is superior to antiplatelet therapy alone53 but is associated with a substantial increase in UGIE, as shown in observational studies54–56 and multiple RCTs.A meta-analysis of 4 RCTs of unfractionated heparin plus ASA versus ASA alone for ACS demonstrated a 50% increase in major bleeds,57 representing an excess of 3 major bleeds per 1000 patients. Low-molecular-weight heparin given in conjunction with ASA also increases major bleeding, as demonstrated in the FRISC-1 (Fragmin during Instability in Coronary Artery Disease-1) study58 and CREATE (Clinical Trial of Reviparin and Metabolic Modulation in Acute Myocardial Infarction Treatment Evaluation).59 A comprehensive meta-analysis of over 25307 patients demonstrated that the benefits of adding warfarin to ASA in the treatment of ACS must be weighed against a 2-fold increased risk in major extracranial bleeding (OR 2.4; 95% CI: 1.4 to 4.1), suggesting that as few as 67 additional patients would need to be treated with ASA plus warfarin to result in 1 additional major extracranial bleeding event.60Conditions such as venous thromboembolism or mechanical heart valves may necessitate long-term anticoagulation. With certain mechanical heart valves, an INR target of 2.0 to 2.5 may not be appropriate, and a higher INR may be required. Depending on the patient's specific bleeding and thrombotic risks, consideration may be given to stopping the antiplatelet agent, as warfarin also has cardioprotective effects.614. GI Effects of ClopidogrelRecommendation: Substitution of clopidogrel for ASA is not a recommended strategy to reduce the risk of recurrent ulcer bleeding in high-risk patients and is inferior to the combination of ASA plus PPI.Because of their alternative molecular targets and inhibition of platelet activation, thienopyridines (ie, clopidogrel, ticlopidine) taken on their own, or in combination with ASA, have beencompared with ASA. The ACC/AHA practice guidelines recommend the use of clopidogrel for hospitalized patients with ACS who are unable to take ASA because of major GI intolerance (Class I, Level of Evidence: A recommendation).2 This recommendation was largely based on the safety data of the CAPRIE (Clopidogrel Versus Aspirin in Patients at Risk of Ischaemic Events) study.62 This study compared clopidogrel 75 mg daily with a relatively high cardioprotective dose of ASA (325 mg daily) for the prevention of ischemic events, including myocardial infarction, stroke, and peripheral arterial disease. After a median follow-up of 1.91 years, the incidence rate of major GI bleeding was lower in the clopidogrel group (0.52%) when compared with the ASA group (0.72%; P less than 0.05). The rate of hospitalization for GI bleeding was 0.7% with clopidogrel versus 1.1% with ASA (P=0.012).63 Although the risk of GI bleeding with clopidogrel was lower than that with ASA, the difference was small (0.2%). Clopidogrel with ASA for at least 1 month is also recommended for patients with a recent non–ST-segment elevation-ACS, with a preference of 12 months if the bleeding risk is not high.2,64 In patients who have received drug-eluting stents, at least 12 months of uninterrupted dual antiplatelet therapy is recommended.65 Data from the CURE,66 MATCH (Management of Atherothrombosis with Clopidogrel in High-Risk Patients),67 and CHARISMA (Clopidogrel for High Atherothrombotic Risk and Ischemic Stabilization, Management, and Avoidance) studies68 provide confirmatory evidence that combined ASA and clopidogrel therapy is associated with significantly increased risk of UGIE complications when compared with either agent alone.69 In patients at high risk of bleeding who require a stent, a bare-metal stent, with its shorter requisite duration of dual antiplatelet therapy, may be preferable.4,70Concomitant use of clopidogrel and an NSAID (including low-dose ASA) has been associated with impaired healing of asymptomatic ulcers17 and disruption of platelet aggregation,71 with a consequent increase in serious UGIE (OR 7.4; 95% CI: 3.5 to 15).28 Few human studies document clopidogrel's potential for independent injury to the GI mucosa. A single endoscopic study with limited follow-up failed to demonstrate mucosal injury in humans.72 In a hospital-based, case-control study of 2777 consecutive patients with major upper GI bleeding and 5532 controls, it was found that non-ASA antiplatelet drugs (clopidogrel, ticlopidine) had a similar risk of upper GI bleeding (adjusted RR 2.8; 95% CI 1.9 to 4.2) to ASA, at a dose of 100 mg/day (adjusted RR 2.7; 95% CI: 2.0 to 3.6), or anticoagulants (adjusted RR 2.8; 95% CI: 2.1 to 3.7).73A prospective, double-blind RCT comparing ASA plus esomeprazole against clopidogrel among H pylori–negative patients with recent UGIE secondary to low-dose ASA demonstrated a significantly higher proportion of recurrent UGIE in the clopidogrel arm versus the ASA plus esomeprazole (20 mg twice daily) arm during the 12 months of study (8.6% versus 0.7%; 95% CI on the difference: 3.4% to 12.4%).74 A subsequent randomized trial with very similar design has shown virtually identical results (13.6% UGIE in the clopidogrel group versus 0% in the ASA plus esomeprazole group [20 mg daily]; 95% CI on the difference: 6.3% to 20.9%).75 These data suggest that use of clopidogrel alone to reduce GI bleeding as an alternative to ASA is not a safe strategy and support ASA cotherapy with once-daily PPI. It remains unclear whether clopidogrel exerts an independent injurious effect on the GI mucosa, or whether it merely induces bleeding in already damaged mucosa via its antiplatelet effects. Observational studies have suggested that PPI cotherapy is beneficial to reduce the risk of clopidogrel monotherapy as well.765. GI Effects of Combined Clopidogrel and Anticoagulant TherapyRecommendation: The combination of clopidogrel and warfarin therapy is associated with an increased incidence of major bleeding when compared with monotherapy alone. Use of combination antiplatelet and anticoagulant therapy should be considered only in cases in which the benefits are likely to outweigh the risks. When warfarin is added to ASA plus clopidogrel, an INR of 2.0 to 2.5 is recommended.52A paucity of evidence informs the clinical risk of combination therapy with clopidogrel or ticlopidine. Anticoagulant agents are not by themselves ulcerogenic; however, they are associated with an increased risk of UGIE because of an exacerbation of pre-existing lesions in the GI tract associated with NSAIDs, ASA, or H pylori infection.76 Clinically, this combination of ASA plus clopidogrel or ticlopidine together with anticoagulation, while not rou