Clinical Outcomes Using a Platelet Function-Guided Approach for Secondary Prevention in Patients With Ischemic Stroke or Transient Ischemic Attack
Background and Purpose—Antiplatelet therapy nonresponse is associated with worse clinical outcomes. We studied the clinical outcomes associated with platelet function-guided modifications in antiplatelet therapy in patients with ischemic stroke or transient ischemic attack.
Methods—From January 2005 to August 2007, 324 patients with ischemic stroke underwent platelet function testing using platelet aggregometry. Aspirin nonresponse was defined as a mean platelet aggregation ≥20% with 0.5 mg/mL arachidonic acid and/or ≥70% with 5 μmol/L adenosine diphosphate. Clopidogrel nonresponse was defined as a mean platelet aggregation ≥40% with 5 μmol/L adenosine diphosphate. A modification was any increase in antiplatelet therapy occurring after testing. Clinical outcomes were compared between patients with and without platelet function-guided antiplatelet therapy modifications using univariate and propensity score-adjusted analyses.
Results—In patients with ischemic stroke or transient ischemic attack, 43% (n=128) and 35% (n=54) were nonresponders to aspirin and clopidogrel, respectively. After platelet function testing, antiplatelet therapy was increased in 23% of patients (n=73). After propensity score matching (n=61 in each group), antiplatelet therapy modification was associated with significantly increased rates of death, ischemic events, or bleeding (hazard ratio, 2.24; 95% CI, 1.12–4.47; P=0.02) compared with no modification in antiplatelet therapy and a trend toward increased bleeding (hazard ratio, 3.56; 95% CI, 0.98–12.95; P=0.05). No differences in ischemic events were observed.
Conclusions—Platelet function-guided modification in antiplatelet therapy after an ischemic stroke or transient ischemic attack was associated with significantly higher rates of adverse clinical outcomes.
- ischemic stroke
- platelet function testing
- transient ischemic attack
Recurrent stroke after an ischemic stroke or transient ischemic attack (TIA) represents nearly 25% of the approximately 795 000 strokes occurring in the United States annually.1 After an ischemic stroke or TIA, antiplatelet therapy is currently recommended to reduce the risk of recurrent ischemic events.2 Antiplatelet agents reduce ischemic events by 22% in patients with prior ischemic stroke or TIA.3 However, the response to antiplatelet therapy with aspirin or clopidogrel is variable.4 The prevalence of aspirin or clopidogrel nonresponse ranges from 5% to 60% and 8% to 28%, respectively.5,6 Nonresponse to aspirin in patients with ischemic stroke is associated with an increased risk of death, adverse cardiac events, and worse functional status.7–9 Adverse cardiac events associated with clopidogrel nonresponse are well described, yet this association has not been well studied in patients with ischemic stroke or TIA.10 Despite antiplatelet nonresponse signifying a risk factor for adverse events, improvement in clinical outcomes by intensifying antiplatelet therapy has not been demonstrated in patients with ischemic stroke or TIA.11,12 The aim of our study was to determine the clinical efficacy and safety associated with platelet function-guided modifications in antiplatelet therapy in patients with ischemic stroke or TIA.
Between January 2005 and August 2007, a retrospective analysis identified 345 patients with ischemic stroke or TIA who underwent platelet function testing in the absence of any endovascular or surgical treatment for stroke. Patients whose antiplatelet therapy was decreased (n=15) or who had warfarin added (n=6) were excluded from the study. The final analysis comprised 324 patients with ischemic stroke (n=250) or TIA (n=74). All patient data were obtained through the electronic medical record system and/or paper charts and were independently verified by the authors. Institutional Review Board approval was obtained on October 18, 2007.
Platelet function was measured by optical platelet aggregation and the definitions used for aspirin and clopidogrel nonresponse are found in Appendix I in the online-only Data Supplement. Patients were separated into groups based on antiplatelet therapy modification. The definition used for antiplatelet therapy modification was any increase in the patient's antiplatelet regimen within 24 hours after the platelet function testing result was made available. Increased antiplatelet therapy was defined as any increase in the dosage of the patient's current antiplatelet agent, adding an additional antiplatelet agent, or switching to a more potent antiplatelet agent (eg, aspirin to clopidogrel).
The primary outcome of the study was a composite of death from any cause, bleeding, or ischemic event. Secondary outcomes included the individual end points of death, bleeding, or ischemic events. Patients were followed from the date of platelet function testing through August 1, 2011. Clinical outcomes are defined in Appendix I in the online-only Data Supplement.
The statistical methods are detailed in Appendix I in the online-only Data Supplement. Differences between the antiplatelet therapy modification and no modification groups were analyzed by univariate methods. Survival function estimates for clinical outcomes were evaluated through Kaplan-Meier analyses and log-rank tests. Propensity scores were created for antiplatelet therapy modification and no modification groups based on patient characteristics. Additional analyses on adjusted and matched propensity scores were conducted. A Cox proportional hazards model for each outcome was created with and without propensity score adjustment and by matching. All statistical analyses were performed using SAS 9.3 (SAS Institute, Cary, NC). A probability value <0.05 was considered statistically significant.
The study analyzed 324 patients with ischemic stroke (n=250) or TIA (n=74) who underwent platelet function testing on aspirin and/or clopidogrel. At the time of platelet function testing, 296 patients (91%) were receiving aspirin and 156 patients (48%) were currently being treated with clopidogrel. The prevalence of aspirin nonresponse (n=128) was 43%, whereas the prevalence for clopidogrel nonresponse (n=54) was 35%. Patients with nonresponse to both aspirin and clopidogrel comprised 19% (n=25) of those on both agents at the time of testing (n=129). Baseline clinical characteristics for the 324 patients in the study with (n=73) and without (n=251) antiplatelet therapy modification after platelet function testing are shown in Table 1. The majority of our study population (79%) underwent platelet function testing in the setting of a recurrent stroke or TIA (n=257). Patients who underwent antiplatelet therapy modification were older, more likely to undergo platelet function testing as an inpatient, and treated more often with β-blockers, calcium channel blockers, angiotensin-converting enzyme inhibitors, or angiotensin receptor blockers compared with patients without antiplatelet therapy modification. Aspirin nonresponse (64% versus 37%, P< 0.001), clopidogrel nonresponse (74% versus 25%, P<0.001), and nonresponse to both aspirin and clopidogrel (54% versus 11%, P<0.001) were significantly higher in patients with antiplatelet therapy modification compared with patients without any modification.
The antiplatelet therapy modifications after platelet function testing are shown in Table 2. Antiplatelet therapy was increased in 73 patients. Increasing the aspirin dosage (n=42) followed by adding or increasing the clopidogrel dosage (n=39) were the most common modifications in antiplatelet therapy. Clopidogrel was added in 30% of patients with an antiplatelet therapy increase and the dosage was increased to 150 mg/day or higher in 23% of patients (n=17). In aspirin nonresponders, the antiplatelet therapy was increased after platelet function testing in 33% of patients (n=42) by increasing the aspirin dosage (n=27), increasing the clopidogrel dosage (n=9), or adding clopidogrel (n=9). An increase in antiplatelet therapy after platelet function testing occurred in 43% of clopidogrel nonresponders (n=23) by either increasing the clopidogrel (n=17) and/or aspirin dosage (n=8). Compared with Table 1, no changes were observed in the distribution of baseline patient characteristics when compared in patients with and without antiplatelet therapy modification for the aspirin or clopidogrel nonresponse subgroups.
Clinical follow-up was available for all patients with a mean follow-up period of 4.6±1.1 years. The clinical outcomes for the study are shown in Table 3. The composite rate of death, bleeding, or ischemic event was significantly higher in patients who underwent antiplatelet therapy modification compared with patients without modification (40% versus 21%, P=0.001). Combined bleeding or ischemic events were also significantly higher in the antiplatelet therapy modified group (34% versus 19%, P=0.006). Kaplan-Meier estimates of cumulative freedom from death, bleeding, or ischemic event (log-rank P< 0.001), bleeding or ischemic event (log-rank P=0.003), and ischemic event (log-rank P=0.04) were significantly higher in patients without antiplatelet therapy modification compared with patients who underwent modification (Figure). A trend toward higher freedom from bleeding was also observed in the patients without antiplatelet therapy modification (log-rank P=0.09; Figure). The rate of death was not significantly different between the 2 groups. Antiplatelet therapy modification was associated with a significant increase in bleeding events (19% versus 10%, P=0.04) compared with patients without modification, which was predominantly driven by increased Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries (GUSTO) moderate bleeding (14% versus 7%, P=0.06) and a significant increase in gastrointestinal bleeding (14% versus 5%, P=0.02). The highest rates of bleeding were observed in patients in whom clopidogrel was added or increased (23% or 9 of 39 patients) or in whom both aspirin and clopidogrel dosages were increased (50% or 4 of 8 patients). A slight trend toward increased ischemic events was associated with antiplatelet therapy modification compared with no modification (15% versus 9%, P=0.12), primarily due to a significantly higher rate of ischemic stroke or TIA (14% versus 6%, P=0.04).
Retesting platelet function after antiplatelet therapy modification was performed in 24 patients (33%). In patients with aspirin nonresponse, 43% were responsive by increasing aspirin and 71% were responsive by increasing or adding clopidogrel. In clopidogrel nonresponsive patients, 88% were responsive by increasing the clopidogrel dose. After retesting, 37.5% of patients remained nonresponsive to aspirin and/or clopidogrel. No significant differences were seen in any clinical outcome between patients who became responsive compared with those who remained nonresponsive.
In patients who were nonresponsive to aspirin, increasing antiplatelet therapy (n=42) compared with no modification or a decrease in antiplatelet therapy (n=86) was associated with increased death, bleeding, or ischemic event (38% versus 19%, P=0.02), which was primarily due to an increase in ischemic events (19% versus 5%, P=0.02), mainly ischemic stroke or TIA (17% versus 2%, P=0.006; Table I in the online-only Data Supplement). Increasing the antiplatelet therapy (n=23) in clopidogrel nonresponsive patients was associated with an increased rate of death, bleeding, or ischemic event (48% versus 19%, P=0.03) compared with no modification or a decrease in antiplatelet therapy (n=31; Table II in the online-only Data Supplement).
The unadjusted and propensity score-adjusted hazard ratios for clinical outcomes with and without modification of antiplatelet therapy are shown in Table 4. With propensity score adjustment, antiplatelet therapy modification was associated with higher rates of death, bleeding, or ischemic event (hazard ratio, 1.84; 95% CI, 1.10–3.09; P=0.02) and a trend toward increased bleeding or ischemic event (hazard ratio, 1.65; 95% CI, 0.95–2.87; P=0.07) compared with no modification. No significant differences were seen in the propensity score-adjusted individual rates of death, bleeding, or ischemic events between the 2 groups. In additional analyses performed after propensity score matching of patients in the antiplatelet therapy modification (n=61) and no modification (n=61) groups, rates of combined death, bleeding, or ischemic event and combined bleeding or ischemic event remained significantly higher in the antiplatelet therapy modification group, whereas a trend toward higher rates of bleeding was observed in the modification group (hazard ratio, 3.56; 95% CI, 0.98–12.95; P=0.05). No significant difference in ischemic events was seen between the 2 matched groups.
The results of our study suggest that modifying antiplatelet therapy after platelet function testing may be associated with increased death, bleeding, or ischemic events compared with patients without any antiplatelet therapy modification. Increasing antiplatelet therapy in patients with aspirin and/or clopidogrel nonresponse was not associated with better, or even similar, clinical outcomes as in those without modification of their antiplatelet regimen but rather was associated with higher event rates. Our findings may reflect the fact that patients who underwent antiplatelet therapy modification were at higher risk, as evidenced by a higher rate of antiplatelet modification in patients who were hospitalized. The results do not provide support for testing, as the strategy of antiplatelet therapy modification was not associated with any evidence of better outcomes.
The prevalence of aspirin or clopidogrel nonresponse in our study was similar to the prevalence reported in prior studies.5,6 Nonresponse to antiplatelet therapy is associated with poorer clinical outcomes.7–9 Beyond noncompliance, the mechanisms associated with aspirin and clopidogrel nonresponse are complex and mutilfactorial.13 Clinical factors associated with antiplatelet therapy nonresponse (eg, diabetes mellitus) may also portend worse clinical outcomes.6 The increased risk associated with nonresponse to antiplatelet agents may reflect increased morbidity of the patient as opposed to a modifiable risk factor. Moreover, clinical factors that predict antiplatelet nonresponse are not consistent between different platelet function tests.14 Platelet function testing is not recommended in the current guidelines for management of ischemic stroke or TIA.2
Preventing recurrent stroke after ischemic stroke/TIA with antiplatelet therapy remains a challenge. Despite the reduction in recurrent events with antiplatelet therapy, a significant number of patients will have a recurrent ischemic event.3 Aspirin plus dipyridamole or clopidogrel may be more effective at preventing recurrent ischemic events compared with aspirin alone in patients with prior stroke/TIA,15–17 yet the current guidelines for ischemic stroke/TIA do not favor one antiplatelet agent over another for prevention of recurrent ischemic events and recommend that selection of an antiplatelet regimen should be individualized to each patient based on patient characteristics, patient preference, and cost.2
The majority of our study population consisted of patients who underwent platelet function testing after a recurrent ischemic stroke or TIA and were on aspirin (91%) and/or clopidogrel (48%). The use of platelet function testing to select patients for optimization of their antiplatelet regimen resulted in heterogeneous regimens used to prevent further ischemic events. Increasing antiplatelet therapy dosages or potency was the predominant modification in patients with nonresponse, yet a number of patients who were nonresponsive did not undergo any antiplatelet therapy modification. Increasing the aspirin dose was used in a number of the patients with aspirin and/or clopidogrel nonresponse. Higher dosages of aspirin do not reduce ischemic events in unselected patients with prior ischemic stroke and are associated with a higher risk of bleeding, mainly gastrointestinal.18–20 In our study, a minority of patients (7%) were changed from aspirin to clopidogrel. The Clopidogrel versus Aspirin in Patients at Risk of Ischemic Events (CAPRIE) trial demonstrated that clopidogrel compared with aspirin reduces ischemic events in patients with recent atherosclerotic vascular disease (ie, ischemic stroke, myocardial infarction, or symptomatic peripheral arterial disease).17 Clopidogrel does not reduce ischemic events in patients with ischemic stroke compared with aspirin plus dipyridamole and was associated with lower rates of intracranial hemorrhage compared with aspirin plus dipyridamole.21 Adding clopidogrel to aspirin was also used to optimize antiplatelet therapy. Dual antiplatelet therapy with aspirin and clopidogrel can increase platelet inhibition in patients with ischemic stroke compared with aspirin or clopidogrel alone.22 Despite enhanced platelet inhibition with aspirin and clopidogrel, 2 randomized trials did not demonstrate any clear clinical benefit with aspirin and clopidogrel compared with aspirin alone in patients with ischemic stroke or TIA but did show a significant increase in the risk for bleeding.23,24 The clopidogrel dosage was increased in 23% of patients whose antiplatelet therapy was modified. Currently, a higher dosage of clopidogrel (>75 mg daily) has not been studied in a randomized trial of patients with ischemic stroke or TIA. In an acute coronary syndrome population, using a higher dosage of clopidogrel (150 mg daily for 7 days) for a short duration after acute coronary syndrome resulted in increased major bleeding compared with standard clopidogrel dosages (75 mg daily).25 Aspirin plus dipyridamole does augment platelet inhibition compared with aspirin alone and, as mentioned previously, may provide some clinical benefit in patients with ischemic stroke compared with aspirin alone.26 In our study, only one patient was changed to aspirin plus dipyridamole. To date, higher dosages of aspirin, adding clopidogrel, or increasing the dosage of clopidogrel have not yielded any clinical benefit in preventing recurrent stroke and, moreover, are associated with an increased risk of bleeding. The current guidelines are consistent with this evidence and do not recommend increasing the dosage of antiplatelet therapy or switching therapy in patients with recurrent ischemic events.2
Several important limitations of our study should be considered. Our study is retrospective and observational. The choice to use platelet function tests in our patient population was determined by the patient's primary physician and may limit the generalizability of the results. Additionally, the diverse modifications in antiplatelet regimens used after platelet function testing were at the physician's discretion. It is unknown what clinical factors led each physician to consider testing for platelet function as well as decide which therapeutic regimen to use after platelet function testing, thus making it very difficult to control for selection bias. Although careful analysis was performed to account for any differences between patients with and without antiplatelet therapy modification, unknown confounders may have contributed to the differences in clinical outcomes between both groups. Furthermore, the infrequency of retesting limited our ability to determine if responsiveness after antiplatelet therapy modification resulted in any clinical benefit.
Platelet function testing may be useful as a marker of increased risk for recurrent events after ischemic stroke or TIA. However, the results of our study demonstrate that testing is not currently useful to optimize clinical outcomes, because the therapeutic responses to platelet function testing in this patient population did not improve clinical outcomes and may be associated with increased adverse events. The results of our study should be interpreted with caution given the possible confounding role of selection bias. Randomized trials are needed to determine if a platelet function-guided approach is beneficial and safe to prevent recurrent events after ischemic stroke or TIA.
Dr Bhatt is on the Advisory Board of Medscape Cardiology; the Board of Directors of the Boston VA Research Institute, Society of Chest Pain Centers; Chair of the American Heart Association Get With The Guidelines Science Subcommittee; received honoraria from the American College of Cardiology (Editor, Clinical Trials, Cardiosource), Duke Clinical Research Institute (clinical trial steering committees), Slack publications (Chief Medical Editor, Cardiology Today Intervention), and WebMD (CME steering committees); has received research grants from Amarin, AstraZeneca, Bristol-Myers Squibb, Eisai, Ethicon, Medtronic, Sanofi Aventis, and The Medicines Company; and received unfunded research from FlowCo, PLx Pharma and Takeda.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.112.655084/-/DC1.
- Received March 3, 2012.
- Revision received May 2, 2012.
- Accepted May 8, 2012.
- © 2012 American Heart Association, Inc.
- Roger VL,
- Go AS,
- Lloyd-Jones DM,
- Benjamin EJ,
- Berry JD,
- Borden WB,
- et al
- Furie KL,
- Kasner SE,
- Adams RJ,
- Albers GW,
- Bush RL,
- Fagan SC,
- et al
Antithrombotic Trialists' Collaboration. Collaborative meta-analysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk patients. BMJ. 2002;324:71–86.
- Selim MH,
- Molina CA
- Eikelboom JW,
- Emery J,
- Hankey GJ
- Farrell B,
- Godwin J,
- Richards S,
- Warlow C
- Grau AJ,
- Reiners S,
- Lichy C,
- Buggle F,
- Ruf A
- Diener HC,
- Bogousslavsky J,
- Brass LM,
- Cimminiello C,
- Csiba L,
- Kaste M,
- et al
- Mehta SR,
- Tanguay JF,
- Eikelboom JW,
- Jolly SS,
- Joyner CD,
- Granger CB,
- et al
- Serebruany VL,
- Malinin AI,
- Sane DC,
- Jilma B,
- Takserman A,
- Atar D,
- et al