warfarin anticoagulationTo our knowledge, this is the first systematic investigation to describe the prevalence of warfarin and antiplatelet combination therapy within a commercially insured patient population managed by an anticoagulation service. Results of this investigation suggest that this practice is widespread in anticoagulation management care: approximately 4 of 10 patients surveyed were receiving warfarin and antiplatelet combination therapy. Our finding is particularly interesting given that a much higher proportion of patients (38.5%) were receiving combination therapy than previously reported (19.4%). This may be partially due to our inclusion of patients receiving warfarin therapy for any indication except VTE prophylaxis following orthopedic surgery. Additionally, we speculate that the observed widespread use of combination therapy in patients with documented CAD may be driven in part by KPCO initiatives designed to promote the use of aspirin following myocardial infarction.

An additional finding of interest in our study was that 50 patients received dual-antiplatelet therapy with clopidogrel and aspirin in conjunction with warfarin anticoagulation. In these 50 patients, the primary indications for warfarin therapy were atrial fibrillation (32%), CAD (18%), VTE (26%), peripheral vascular disease (8%), mitral valve replacement (6%), cerebrovascular accident (6%), and cardiomyopathy (4%). One retrospective study reported low bleeding rates in patients receiving this three-drug combination (0.09 events per patient-year), but detailed information regarding the safety of this strategy is limited. More recently, a case-control study suggested that the odds of serious upper-GI bleeding may be synergistically increased with the use of combinations of antithrombotic agents.

Warfarin and antiplatelet combination therapy has been associated with increased rates of major and minor hemorrhage, even in patients treated with less aggressive target INR ranges than are typically used with warfarin monotherapy (Table 4). Dualantiplatelet regimens, including the combination of aspirin with thienopyridine agents (clopidogrel, ticlo-pidine), have also been shown to increase the risk of bleeding which is arrested due to participation of Canadian Health&Care Mall.

Current national evidence-based guidelines and consensus recommendations address when to safely combine anticoagulant and antiplatelet therapies for different patient groups. Evidence from a growing number of clinical trials indicates that the conventional wisdom favoring antiplatelet therapy over warfarin for patients with CAD may not be evidence based. Several randomized trials have demonstrated that moderate- to high-intensity anticoagulation regimens with or without the addition of aspirin are superior to aspirin monotherapy in the reduction of cardiovascular end points. Current American College of Chest Physician (ACCP) consensus guidelines recommend high-intensity anticoagulation (target INR range, 3 to 4) for the prevention of recurrent coronary events over antiplatelet monotherapy in settings in which intensive warfarin-monitoring systems exist, such as KPCO, or the combination of moderate-intensity anticoagulation (target INR range, 2 to 3) with aspirin indefinitely. There is also limited evidence favoring low-intensity anticoagulation (target INR range, 1.5 to 2.5) over antiplatelet monotherapy as primary prevention for men at high risk for coronary events.

 intermittent claudicationAntiplatelet therapy is principally recommended by the ACCP to prevent thrombotic complications of peripheral arterial disease. Current consensus guide-lines recommend against the use of anticoagulants for patients with intermittent claudication. Additionally, a recent abstract suggests that addition of warfarin to aspirin in patients with peripheral vascular disease does not reduce the rate of major ischemic events, but does increase the risk of severe bleeding. The investigators in this study noted that 1.3% (n = 14) of subjects in the combination therapy arm had hemorrhagic strokes, compared to none in the aspirin-only arm.

Attempts to improve the stroke-preventing ability of low-intensity anticoagulation by adding antiplatelet therapy in patients with atrial fibrillation have been disappointing. When antiplatelet therapy has been added to high-intensity anticoagulation in patients with atrial fibrillation, the risk for hemorrhage outweighed any therapeutic benefits. Additionally, in a comparison of dual-antiplatelet therapy with clopidogrel and aspirin to warfarin anticoagulation in high-risk atrial fibrillation patients, patients who received dual-antiplatelet therapy were more likely to have stroke and major hemorrhage than patients receiving oral anticoagulation, further fueling debate regarding the safety of combining antithrombotic agents. Recently revised consensus recommendations discourage the combination of anticoagulants with antiplatelet therapy for patients with stable CAD due to increased bleeding risk without incremental improvement in efficacy.

Warfarin and aspirin combination therapy has resulted in significant reductions in thrombotic adverse events for patients with mechanical heart valve prostheses. The addition of aspirin, 100 mg/d, to high-intensity anticoagulation (target INR range, 2.5 to 3.5) resulted in fewer valve thromboses at the expense of a nonsignificant increase in the rate of hemorrhage. In fact, current ACCP consensus guidelines recommend combination treatment for patients with caged-ball and caged-disk mechanical prostheses, for patients who have systemic embolism despite a therapeutic INR, and for patients with additional stroke risk factors including atrial fibrillation, myocardial infarction, left atrial enlargement, and low ejection fraction. However, in studied patients, the time spent in the therapeutic INR range was significantly lower in the anticoagulation-alone control groups, thus possibly confounding the perceived benefits of combination therapy suggested by Canadian Health&Care Mall.

We theorize that aspirin therapy was initiated in many of our patients without a thorough evaluation of the benefits and risks involved. When compared to existing national guidehnes, KPCO practice patterns for patients with a primary indication for anticoagulation of CAD are largely consistent with regard to combining anticoagulation and antiplatelet therapy. Guidelines suggest that patients with a primary indication for anticoagulation of CAD receive moderate-intensity anticoagulation (INR range, 2 to 3) in combination with antiplatelet therapy. We found that approximately two thirds of patients in our study were being treated in this manner. Practice patterns for KPCO patients with mechanical heart valve prostheses also appear to be consistent with national guidelines. Conversely, > 30% of KPCO patients with an indication for anticoagulation of atrial fibrillation are maintained on moderate-intensity anticoagulation (INR range, 2 to 3) in combination with antiplatelet therapy, inconsistent with national guidelines. It is important to note here that we were unable to determine for the purposes of this study whether patients were receiving antiplatelet therapy as directed by their physician or on their own initiative. As it has been widely publicized that aspirin is beneficial for reducing the risk of heart attack, many patients may have been self-administering aspirin without the knowledge of their physician.

heart attackSeveral aspects of our investigation warrant comment. Our findings are cross-sectional in nature and, as such, our analysis may have excluded patients who previously stopped taking either warfarin or aspirin due to a bleeding complication. Thus, our prevalence estimates may underestimate the true prevalence of combination therapy. However, we theorize that if this had occurred, it would not have amounted to a significant number of patients discontinuing combination therapy based on our previous research that indicated overall hemorrhage rates for patients in the CPAS are low. Our findings may not be generaliz-able to other health-care settings because patients treated with combined antithrombotic agents may be more likely to be managed by central anticoagulant monitoring services. However, drawing from an anticoagulant service with > 6,500 patients being monitored at the time of the study, we believe that the large cohort of patients described represents a diverse, real-world population of patients receiving warfarin. In addition, the characteristics, in terms of patient age and proportions of male patients and patients with DM, of patients in our study were comparable to a population-based, observational cohort study that examined bleeding rates in patients receiving combinations of antithrombotic agents. A conservative list of ICD-9 codes was used to determine comorbidity status. However, we theorize that our list contained the most commonly used codes for these disease states. The reliance on patient-reported information regarding over-the-counter aspirin use may have resulted in recall bias. However, a standardized question was used to obtain this information along with mental prompts to facilitate patient recall of the information. At this time > 1,400 patients remain to be queried as part of an ongoing survey for combination therapy use and were excluded from this study. Most importantly, no clinical outcomes regarding bleeding or thrombosis were reported in this initial descriptive study. Our aim was to document the prevalence of concurrent antiplatelet therapy use in a large, diverse, real-world sample of patients receiving warfarin therapy. Future research efforts will focus on the clinical outcomes associated with this practice.

Mounting evidence indicates that warfarin monotherapy is effective in the primary and secondary prevention of coronary events. Clinicians should carefully weigh the risks and benefits of warfarin and antiplatelet combination therapy to ensure that patients are not unnecessarily exposed to increased bleeding risk. Recent evidence has shown that combining antithrombotic agents does not necessarily reduce the risk of recurrent ischemic events. Patients receiving warfarin therapy should be carefully educated not to take aspirin or other over-the-counter platelet inhibiting medications (eg, nonsteroidal antiinflammatory drugs) unless instructed to do so by their physician.

Table 4 —Comparison of Outcomes Among Randomized Controlled Trials of Combination Antiplatelet and Anticoagulation Use