Elevated Plasma Fibrinogen and Diabetes Mellitus Are Associated With Lower Inhibition of Platelet Reactivity With Clopidogrel

Transcription

1 Journal of the American College of Cardiology Vol. 52, No. 13, by the American College of Cardiology Foundation ISSN /08/$34.00 Published by Elsevier Inc. doi: /j.jacc Antiplatelet Therapy Elevated Plasma Fibrinogen and Diabetes Mellitus Are Associated With Lower Inhibition of Platelet Reactivity With Clopidogrel Lawrence Ang, BS, Vachaspathi Palakodeti, MD, FACC, Ahmer Khalid, BS, Sotirios Tsimikas, MD, FACC, Zaheib Idrees, BS, Phillip Tran, MD, Paul Clopton, MS, Nayab Zafar, MD, Guilherme Bromberg-Marin, MD, Shahin Keramati, MD, FACC, Ehtisham Mahmud, MD, FACC San Diego, California Objectives Background Methods The goal of this study was to identify factors associated with lower platelet inhibition (PI) with clopidogrel in subjects with cardiovascular disease (CVD). A heterogeneous platelet reactivity response to clopidogrel exists, and the clinical or biochemical predictors of suboptimal PI with clopidogrel remain unclear. This study prospectively enrolled subjects with CVD requiring treatment with clopidogrel (75 mg daily for 7 days or 600-mg bolus 24 h before recruitment). A bedside rapid platelet function assay (VerifyNow, Acccumetrics, San Diego, California) to measure maximal and clopidogrel-mediated platelet reactivity was utilized, and factors associated with lower PI were identified. Results A heterogeneous, normally distributed PI (mean %) response to clopidogrel was observed in 157 subjects (age years; 59.9% men). Multiple variable analysis of clinical and biochemical factors known to affect platelet reactivity revealed lower PI in patients with an elevated plasma fibrinogen level ( 375 mg/dl), diabetes mellitus, and increased body mass index (BMI) ( 25 kg/m 2 ). On testing for interaction, elevated fibrinogen level was associated with diabetic status, resulting in lower PI in diabetic patients ( % vs %, p 0.001), but not nondiabetic patients ( % vs %, p 0.244). Increased BMI remained independently associated with lower PI after clopidogrel therapy regardless of diabetic status or fibrinogen level ( % vs %, p 0.001). Conclusions Elevated plasma fibrinogen ( 375 mg/dl) in the presence of diabetes mellitus and increased BMI ( 25 kg/m 2 ) are associated with lower PI with clopidogrel in patients with CVD. (J Am Coll Cardiol 2008;52:1052 9) 2008 by the American College of Cardiology Foundation From the Division of Cardiovascular Medicine, University of California, San Diego School of Medicine, San Diego, California. Support for Mr. Ang was provided, in part, by the University of California SMART (SIRA Medical-Student Aging Research Training) Grant. Steven E. Nissen, MD, MACC, served as Guest Editor for this article. Manuscript received January 6, 2008; revised manuscript received May 27, 2008, accepted May 28, Dual antiplatelet therapy with aspirin and clopidogrel reduces adverse cardiovascular events, including death, stroke, myocardial infarction, recurrent ischemia, and urgent target vessel revascularization after percutaneous coronary intervention (PCI) (1) with enhanced platelet reactivity associated with an increased incidence of adverse ischemic events (2). Aspirin and clopidogrel are associated with a heterogeneous platelet aggregation response as shown using various platelet function assays (3 5). Light transmission aggregometry (LTA) is considered the gold standard for platelet function testing (6) and has been used to demonstrate significant heterogeneity in platelet aggregation after antiplatelet therapy in populations with and without cardiovas- See page 1060 cular disease (CVD) (4). However, due to labor-intensive assays, expense, confinement to a clinical laboratory, potential for inconsistency, and delayed availability of results, LTA is difficult to use clinically at the bedside (6). Additionally, an increasing clinical interest in evaluating immediate patient response to antiplatelet therapy has led to the development of various rapid platelet function assays (7). The VerifyNow System (Accumetrics, San Diego, California) is closely modeled after LTA and uses similar

2 JACC Vol. 52, No. 13, 2008 September 23, 2008: Ang et al. Predictors of Platelet Inhibition With Clopidogrel 1053 reagents and photospectometry to measure light transmittance in whole samples of blood (8). A previous study using the VerifyNow System to measure platelet reactivity reported a heterogeneous response to clopidogrel in subjects with multiple risk factors for CVD (3). The etiology behind this heterogeneous response to clopidogrel is incompletely understood. Multiple studies attempting to characterize this phenomenon have identified interpatient pharamacokinetic and pharamacodynamic variation, including differences in drug absorption, bioactivation by hepatic cytochrome P450 enzymes, platelet cell surface P2Y12 and glycoprotein (GP) receptor polymorphisms, drug drug interactions, baseline platelet reactivity, and metabolic profile as potential factors (9 11). However, clinically useful indicators and biomarkers to identify patients at risk for suboptimal platelet inhibition (PI) with clopidogrel remain unclear. Increased tendency for platelet aggregation is associated with diabetes mellitus (DM) (12), smoking (13), hyperlipidemia (14,15), higher fibrinogen (16), and a history of coronary atherosclerosis (17). Further, plasma fibrinogen 375 mg/dl is associated with suboptimal PI with the GP IIb/IIIa inhibitor epitifibatide (18). We undertook this study to determine the level of PI achieved with clopidogrel as measured with the rapid platelet function assay VerifyNow in patients with CVD, and to define the role of clinical and biochemical factors known to affect platelet reactivity on the platelet response to clopidogrel. Methods Patient selection. The study protocol was approved by the Institutional Review Board of the University of California, San Diego, and informed written consent was obtained from study participants. Patients with established CVD requiring treatment with clopidogrel were prospectively recruited and enrolled. Subjects included, but were not limited to those with prior coronary or peripheral vascular interventions, those undergoing elective PCI, elective peripheral vascular intervention, and being medically treated for an acute coronary syndrome. All subjects were considered for inclusion if they were treated with clopidogrel 75 mg daily for a minimum of 7 days or were administered a 600-mg bolus at least 24 h before enrollment. Exclusion criteria included age younger than 18 years, anemia (hemoglobin 10 g/dl), thrombocytopenia (platelet count 100,000 cells/ l), diagnosis of inherited platelet disorders, serum triglycerides 400 mg/dl, chronic liver disease, GP IIb/IIIa inhibitor administration (within previous 30 days), and acute myocardial infarction within 72 h. Measurements. Demographic information, baseline laboratories including complete blood count, lipid panel, and serum markers of inflammation (C-reactive protein [CRP], fibrinogen) were obtained in all subjects. The complete blood count was performed using the Abbott Cell-Dyne 3500 Analyzer and commercial reagents (Abbott, Abbott Park, Illinois). Lipid panel and CRP were measured using a Beckman Coulter Synchron LX 20 analyzer and commercial reagents (Beckman Coulter, Fullerton, California). Fibrinogen was measured using the Beckman Coulter ACL-TOP analyzer with Instrumentation Laboratory reagents (Instrumentation Laboratory, Lexington, Massachusetts). Platelet function testing was performed using the bedside rapid platelet function VerifyNow P2Y12 and aspirin assays. Blood sample collection. Whole samples of blood were obtained from the femoral artery Abbreviations and Acronyms BMI body mass index CRP C-reactive protein CVD cardiovascular disease DM diabetes mellitus GP glycoprotein LTA light transmittance aggregometry PCI percutaneous coronary intervention PI platelet inhibition PRU P2Y12 response units catheter introducer sheath inserted before PCI or from the antecubital fossa vein using standard phlebotomy techniques. The first 4 ml of blood taken from either location were discarded before being drawn into a syringe and transferred into standard tubes for respective serology assays and 2-ml Bio-One Vacuette tubes (Greiner Bio-One North America Inc., Monroe, North Carolina) containing 3.2% sodium citrate specifically for platelet function testing. VerifyNow System. The cartridge-based VerifyNow System utilizes optical turbidometry to rapidly assess platelet function and strongly correlates with traditional LTA (5). This method of platelet function testing depends on platelet reactivity in vitro and allows for evaluation of platelet function with thienopyridine, aspirin, or GP IIb/IIIa inhibitor therapy. In the P2Y12 assay, adenosine diphosphate agonist activates platelets at P2Y12 and P2Y1 cell surface receptors while prostaglandin E1 increases assay specificity by blocking signaling through the P2Y1 receptor. Activation at the P2Y12 receptor causes flattening of platelet shape, liberation of intracellular GP IIb/IIIa receptor reserves, and conformational change of the GP IIb/IIIa beta integrin subunit that increases receptor affinity for fibrinogen (19). These effects lead to increased platelet aggregation around fibrinogen-coated beads and light transmittance through the sample reported as P2Y12 response units (PRU). Unique to the P2Y12 assay, a high concentration of thrombin receptor activating protein (iso-trap) is added to the blood sample to circumvent the P2Y12 receptor signaling pathway and induce maximal platelet activation and aggregation. This is designed to allow platelet function before clopidogrel treatment ( baseline ) to be determined without discontinuing treatment, and for PI to be measured in clopidogrel users. Platelet inhibition is calculated as the quotient of PRU base minus post-treatment PRU result and PRU base (20). In the aspirin assay, a lyophilized arachidonic acid reagent is converted to thromboxane A2 platelet agonist by the cyclooxygenase-1 enzyme (21). Thromboxane A2-induced platelet activation in vitro is reported as aspirin response units,

3 1054 Ang et al. JACC Vol. 52, No. 13, 2008 Predictors of Platelet Inhibition With Clopidogrel September 23, 2008: which negatively relates to the efficacy of cyclooxygenase-1 inhibition by aspirin. This assay has high sensitivity (85%) and specificity (100%) (22) with aspirin response units 550 indicating the failure of aspirin to adequately inhibit platelets. Statistical analysis. Data were collected and analyzed with the use of SPSS software (version 15.0, SPSS Inc., Chicago, Illinois). Categorical variables are reported as percentages while continuous variables are reported as mean SD. Distribution of PI achieved was confirmed for normality using the Kolmogorov-Smirnov test. Correlations between continuous variables and PI achieved were evaluated using Pearson s correlation or point-biserial correlation as appropriate. Categorical variables were analyzed by comparing the PI achieved within each subgroup using independentsamples 2-tailed t tests. The PI by fibrinogen tertiles was determined and compared using an analysis of variance. The p values 0.05 indicated significant relationships. Age, gender, and all factors known to affect platelet reactivity that were associated with PI in univariate analysis (p 0.05) were entered into a multiple variable analysis model. The effects of DM, elevated fibrinogen, and their interaction were evaluated in multiple regression after controlling for the aforementioned covariates including age, gender, venous blood sampling, history of hyperlipidemia, and body mass index (BMI) 25 kg/m 2. Adjusted PI means for groups based on DM and fibrinogen were computed using the significant control variables from the regression analysis. Baseline Demographics of the Study Population Table 1 Baseline Demographics of the Study Population Variable (n 157) Age (yrs), mean SD Male gender, % 59.9 BMI (kg/m 2 ), mean SD Cardiovascular history, % Myocardial infarction 18.8 Coronary artery bypass graft surgery 13.4 Percutaneous coronary intervention 52.9 Stable angina 26.3 Unstable angina 14.1 CVD risk factors, % Family history of CVD 44.5 Hypertension 86.6 Diabetes mellitus 51.0 Hyperlipidemia 80.3 Current smoker 11.5 BMI 25 kg/m Baseline medications, % Chronic clopidogrel (75 mg daily 7 days) 87.3 Aspirin 89.1 Statins 72.4 Beta-blockers 60.9 ACE inhibitors/arb 57.7 Nitrates 19.2 Calcium-channel blockers 25.6 Diabetes treatment (n 80), % Diet control only 20.3 Oral hypoglycemic agent 62.2 Insulin requiring 23.0 Indication for clopidogrel, % Drug-eluting or bare-metal stent 72.0 Coronary stent 58.6 Peripheral vascular stent 13.4 Medical therapy for CVD 23.6 Coronary balloon angioplasty 3.2 Post-coronary artery bypass graft surgery 1.3 ACE angiotensin-converting enzyme; ARB angiotensin receptor blocker; BMI body mass index; CVD cardiovascular disease. Results Figure 1 Patient Flow Diagram Patients with established cardiovascular disease requiring treatment with clopidogrel were prospectively recruited from catheterization laboratories and outpatient clinics. The majority of subjects presented for evaluation of coronary artery disease (CAD). A subset of patients evaluated for CAD presented with symptoms of an acute coronary syndrome (ACS), including unstable angina or recent myocardial infarction (MI) 72 h before recruitment. PAD peripheral arterial disease. A total of 157 subjects ( years; 59.9% men) treated with clopidogrel were enrolled in the study (Fig. 1, Table 1). There was a high prevalence of multiple risk factors for CVD, and the majority of subjects had previously undergone coronary revascularization. Clopidogrel was administered to the majority of patients for coronary or peripheral vascular stent placement. Concomitant aspirin and statin use was observed in 89.1% and 72.4% of the study population, respectively. The majority of subjects were being treated with clopidogrel 7 days (n 137) while the rest received a 600-mg bolus dose of clopidogrel 24 h before recruitment (n 20). The measured PI followed a normal, bell-shaped distribution (mean %) (Fig. 2) with a strong negative correlation with the final PRU values (r 0.879; p 0.001) (Table 2). There was no correlation

4 JACC Vol. 52, No. 13, 2008 September 23, 2008: PI Presence Achieved or With Absence Clopidogrel of Categorical in the Variables Table 3 Ang et al. Predictors of Platelet Inhibition With Clopidogrel PI Achieved With Clopidogrel in the Presence or Absence of Categorical Variables Mean %PI Achieved SD 1055 Figure 2 Normal Distribution of Platelet Inhibition Achieved With Clopidogrel The platelet response of 157 subjects with cardiovascular disease to clopidogrel therapy was measured with the rapid platelet function assay VerifyNow (Accumetrics, San Diego, California) and shows a normal, bell-shaped distribution ( %). between PI with clopidogrel and the baseline PRU (r 0.058; p 0.455). The presence of elevated fibrinogen ( 375 mg/dl), DM, increased BMI ( 25 kg/m 2 ), venous blood sampling, nitrate therapy, history of hyperlipidemia, and elevated serum triglyceride level each demonstrated a significant relation with PI achieved (Tables 3 and 4). A significantly lower PI was observed in the highest tertile of fibrinogen level ( % first tertile vs % second tertile vs % third tertile; p 0.032) (Fig. 3). Patient Baseline Platelet Function Laboratory Assay and Results Table 2 Baseline Laboratory and Platelet Function Assay Results Variable (n 157) Lipid panel (mg/dl), mean SD High-density lipoprotein Low-density lipoprotein Triglycerides CBC, mean SD White blood cell count (cells/ l) Hemoglobin (g/dl) Platelet count (cells/ l) Fibrinogen (mg/dl) CRP 0.5 mg/dl, % 29.3 VerifyNow results, mean SD Base PRU Final PRU Platelet inhibition, % ARU ARU 550, % 11.3 ARU aspirin response units; CBC complete blood count; CRP C-reactive protein; PRU P2Y12 response units. Variable Factor Present Factor Absent p Value Male gender Clinical details Acute coronary syndrome Blood drawn from venous access Cardiovascular history Myocardial infarction CABG PCI Stable angina Unstable angina Cardiovascular risk factors Family history of CVD Hypertension Hyperlipidemia Current smoker Diabetes mellitus Elevated fibrinogen ( 375 mg/dl) BMI 25 kg/m CRP 0.5 mg/dl Baseline medications Aspirin Statins Atorvastatin Beta-blocker ACE inhibitor/arb Nitrates Calcium-channel blocker Chronic clopidogrel ( 7 days) CABG coronary artery bypass graft surgery; CRP C-reactive protein; PCI percutaneous coronary intervention; PI platelet inhibition; other abbreviations as in Table 1. response to aspirin coadministration revealed no significant relation with level of PI with clopidogrel (r 0.097; p 0.239). Indicators of inflammation including CRP 0.5 mg/dl, white blood cell count, platelet count, and past cardiovascular history including prior history of stable angina or unstable angina, myocardial infarction, coronary artery bypass graft surgery, and PCI did not demonstrate a significant relation with PI achieved. Similarly, smoking was not found to be associated with the PI achieved with clopidogrel. Notably, significantly higher final PRU values were observed in the presence of an elevated fibrinogen ( vs , p 0.015), DM ( vs , p 0.022), and increased BMI ( vs , p 0.022), respectively. Subject age, gender, and measured variables known to affect platelet reactivity that showed significant univariate relationships with PI were included in a multiple variable linear regression model. Before testing for interactions, elevated fibrinogen ( 375 mg/dl) (p 0.047), DM (p 0.021), and increased BMI ( 25 kg/m 2 )(p 0.031) were

5 1056 Ang et al. JACC Vol. 52, No. 13, 2008 Predictors of Platelet Inhibition With Clopidogrel September 23, 2008: Correlation With Clopidogrel Between andplatelet Continuous Inhibition Variables Table 4 Correlation Between Platelet Inhibition With Clopidogrel and Continuous Variables Variable r Value p Value Demographics Age (yrs) Body mass index (kg/m 2 ) Laboratory values HDL cholesterol (mg/dl) LDL cholesterol (mg/dl) Triglycerides (mg/dl) White blood cell count (cells 10 3 / l) Hemoglobin (g/dl) Platelet count (cells 10 3 / l) Fibrinogen level (mg/dl) Base PRU Final PRU ARU HDL high-density lipoprotein; LDL low-density lipoprotein; other abbreviations as in Table 2. the only factors associated with lower PI with clopidogrel (Table 5). Evaluation of these variables for interactive effects revealed a significant interaction between elevated fibrinogen and DM while no other interaction was detected. After including this interaction in the multiple variable model, elevated fibrinogen and DM lost statistical association with PI (p and 0.457, respectively) independent of each other. Elevated fibrinogen level in the presence of DM was associated with lower PI (p 0.024) while higher BMI ( 25 kg/m 2 ) regardless of fibrinogen level or DM remained significantly associated with lower PI (p 0.026) (Table 5). The adjusted PI (mean SD) was significantly lower for diabetic subjects with elevated fibrinogen ( %) compared with diabetic subjects without elevated fibrinogen ( %) (p 0.001) (Fig. 4). No significant difference existed in PI for nondiabetic subjects with elevated fibrinogen ( %) compared with that in nondiabetic subjects without elevated fibrinogen ( %) (p 0.244) (Fig. 4). Adjusted PI was also significantly lower for subjects with increased BMI ( %) compared to those with normal BMI ( %) (Fig. 5). an important step in explaining the prothrombotic effects observed in diabetic patients. A persistently elevated fibrinogen level is associated with the development of CVD (23), and an elevated plasma fibrinogen level is also associated with achieving suboptimal PI with the GP IIb/IIIa inhibitor eptifibatide (18). The current study demonstrates a similar finding with the thienopyridine clopidogrel. The relation between an elevated fibrinogen level and lower PI in response to clopidogrel was strongest in diabetic subjects who are also known to have higher fibrinogen levels (24). Fibrinogen mediates platelet plug formation and stabilizes clots though the polymerization of fibrin. Platelet plugs are formed as fibrinogen binds and crosslinks exposed GP IIb/IIIa receptors on platelet cell surfaces while fibrin is produced by the proteolytic cleavage of fibrinogen by thrombin (19). Antiplatelet therapy interferes with this process by inhibiting platelet activation, limiting the availability of GP IIb/IIIa receptors at the platelet surface, and lowering GP IIb/IIIa binding affinity to fibrinogen (9). Fibrinogen concentration in serum may also decrease by 10% to 25% with thienopyridine administration, further contributing to the antiplatelet effects of these agents (25). The etiology for clopidogrel resistance in diabetic patients is multifactorial, and insulin resistance and systemic inflammation play a significant role (26). Insulin resistance characteristic of type II DM exhibits both direct and indirect effects on platelets. Directly, decreased platelet sensitivity to insulin is associated with reduced P2Y12 receptor inhibition and increased platelet reactivity (27). Indirectly, hyperglycemic states have also been implicated in nonenzymatic glycation of platelet GPs resulting in altered platelet struc- Discussion This study documents a heterogeneous PI response to clopidogrel for patients with CVD as measured with a rapid platelet function assay. The presence of elevated fibrinogen ( 375 mg/dl), DM, and high BMI ( 25 kg/m 2 ) are all associated with lower PI after treatment with the oral thienopyridine clopidogrel in subjects with CVD. Most importantly, elevated plasma fibrinogen is identified as a unique factor associated with the lower PI in diabetic subjects who are generally at higher risk for cardiovascular events. This study uniquely addresses the phenomenon of suboptimal PI achieved in the diabetic patient by revealing a significant interaction between elevated plasma fibrinogen 375 mg/dl and the presence of DM. This finding may be Figure 3 Platelet Inhibition With Clopidogrel as Stratified by Serum Fibrinogen Tertiles Subjects with serum fibrinogen in the highest tertile range ( 368 mg/dl) are observed to have significantly lower platelet inhibition with clopidogrel compared with subjects in the first and second tertiles with lower serum fibrinogen levels. Data are shown as mean standard error of mean. p

6 JACC Vol. 52, No. 13, 2008 September 23, 2008: Results of Multiple Variable Analysis Table 5 Results of Multiple Variable Analysis Ang et al. Predictors of Platelet Inhibition With Clopidogrel 1057 B* Standard Error Beta t p Value Multiple variable analysis excluding interactions Fibrinogen 375 mg/dl Diabetes mellitus BMI 25 kg/m Multiple variable analysis including interactions Fibrinogen 375 mg/dl Diabetes mellitus BMI 25 kg/m Fibrinogen 375 mg/dl and diabetes mellitus *B value indicates the relative effect that each significant variable exerts on the linear regression model. BMI body mass index. ture, conformation, and membrane lipid dynamics (28). Further, strict control of serum glucose levels have been shown to diminish platelet aggregation (26). Clinically, Buonamici et al. (2) have demonstrated that DM is significantly more prevalent in individuals who obtain 30% PI with clopidogrel, who are also at an increased rate of late stent thrombosis and death compared with those achieving 30% PI. In this study, we also identify the presence of DM as a factor associated with lower PI. However, this finding was only significant in the presence of an elevated serum fibrinogen level. Diabetic subjects with fibrinogen 375 mg/dl only achieve 23.9% PI with clopidogrel, and an elevation in this biomarker might identify a group of diabetic patients who require a higher dose of this thienopyridine or an alternative agent for optimal PI. In addition to being a common end product of both intrinsic and extrinsic clotting cascades, fibrinogen is also an acute-phase reactant of inflammation. It, along with other acute-phase reactants including CRP, is synthesized and released by hepatocytes in increased amounts when triggered by specific inflammatory cytokine signaling produced in the presence of DM (29). However, in this study no difference in PI achieved was observed between individuals with lower circulating CRP 0.5 mg/dl and those with higher circulating CRP 0.5 mg/dl. This is consistent with the results of another study that failed to demonstrate a role of CRP with platelet aggregation (22). In addition, no relation between other markers of inflammation including white blood cell and platelet counts, and PI achieved with clopidogrel was observed. These findings are consistent with the hypothesis that fibrinogen affects PI by its direct interaction Figure 4 Platelet Inhibition With Clopidogrel in Diabetic Patients With Elevated Fibrinogen Subjects with both diabetes mellitus and elevated plasma fibrinogen ( 375 mg/dl) have significantly lower platelet inhibition with clopidogrel compared with nondiabetic patients or diabetic patients with lower fibrinogen levels. The adjusted platelet inhibition (mean SD) with the number of subjects in each group is shown. Figure 5 Platelet Inhibition With Clopidogrel as Stratified by BMI Increased body mass index (BMI) ( 25 kg/m 2 ) remains independently associated with lower platelet inhibition with clopidogrel regardless of diabetic status or fibrinogen level. The adjusted platelet inhibition (mean SD) in subjects with high and normal BMI is shown. p

7 1058 Ang et al. JACC Vol. 52, No. 13, 2008 Predictors of Platelet Inhibition With Clopidogrel September 23, 2008: with the GP IIb/IIIa receptor and not due to a systemic state of inflammation. This study also demonstrates that BMI 25 kg/m 2 is independently associated with lower PI with clopidogrel regardless of diabetic status or plasma fibrinogen level. The relationship between BMI and the efficacy of antiplatelet therapy remains a topic of debate since past studies have provided conflicting results. Though increased BMI is associated with higher platelet aggregation and decreased treatment efficacy (30), larger body size is also associated with superior 1-year clinical outcomes for death, myocardial infarction, and stroke after PCI (31). Potential confounders for these findings include the effects of age and gender differences in the 2 studies. The finding that higher BMI is associated with lower PI is intuitive, confirms previous findings (31), and raises into question the administration of the same maintenance dose of clopidogrel to all patients regardless of BMI. Several other factors known to affect platelet aggregation, including low-density and high-density lipoprotein levels, smoking status, presenting with an acute coronary syndrome, history of coronary artery disease, and inflammatory state (CRP, white blood cell, and platelets) were not found to have a significant relation with the PI observed with clopidogrel. Although a history of hyperlipidemia was associated with lower PI after univariate analysis, this relation did not retain significance after multivariate analysis. Further, the use of statins and specific use of atorvastatin was not associated with PI, a finding supported by a post-hoc analysis of the CHARISMA (Clopidogrel for High Atherothrombic Risk and Ischemic Stabilization, Management and Avoidance) trial, which revealed no adverse clinical interaction during coadministration of clopidogrel and statins (32). Additionally, patient response to concurrent aspirin administration did not show a significant relation to the degree of PI achieved demonstrating that the PI response observed with aspirin and clopidogrel therapy are independent of each other. Finally, in patients with established CVD, the mean PI of 40.8% as measured by the VerifyNow assay is similar to the 41.9% PI achieved in a similar population using LTA (4). A recent head-to-head comparison of LTA and the VerifyNow P2Y12 assay showed a strong correlation between final PRU values and percent peak aggregation according to LTA after clopidogrel administration (r 0.73; p 0.01) (5). Our findings also show that PRU result values, rather than base values, highly correlate with calculated PI (r 0.879; p 0.001), implying that baseline platelet reactivity might have a less significant role in the final PI achieved. Recent studies linking level of PI achieved with clinical outcomes have all utilized LTA, and studies with the VerifyNow P2Y12 assay linking clinical outcomes are required to further validate this assay for routine clinical practice. Study limitations. Polymorphisms of the hepatic cytochrome P450 enzymes were not evaluated, possibly allowing confounding effects from altered drug metabolism of clopidogrel. The diagnosis of DM in study subjects was determined from direct patient interviews and review of medical records rather than from evaluation of serum glucose and glycosylated hemoglobin levels. Nevertheless, in a clinical environment, fibrinogen is an easy marker to measure, and may serve as an indicator of a high-risk patient for thrombotic risk due to poor PI in response to therapy with the thienopyridine clopiodgrel. Conclusions A heterogeneous, normally distributed platelet reactivity response to clopidogrel in subjects with CVD is demonstrated using a rapid platelet function assay. Elevated plasma fibrinogen 375 mg/dl, the presence of DM, and BMI 25 kg/m 2 are associated with lower PI with clopidogrel therapy in subjects with CVD. A significant interaction between elevated plasma fibrinogen and DM exists, identifying elevated plasma fibrinogen as a unique factor associated with lower PI in diabetic patients. Increased BMI remains independently associated with lower PI after clopidogrel therapy. Reprint requests and correspondence: Dr. Ehtisham Mahmud, Cardiovascular Catheterization Laboratories, UCSD Medical Center, 200 West Arbor Drive, San Diego, California emahmud@ucsd.edu. REFERENCES 1. Braunwald E, Antman EM, Beasley JW, et al. American College of Cardiology; American Heart Association. ACC/AHA 2002 guideline update for the management of patients with unstable angina and non ST-segment elevation myocardial infarction summary article: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on the Management of Patients With Unstable Angina). J Am Coll Cardiol 2002;40: Buonamici P, Marcucci R, Migliorini A, et al. Impact of platelet reactivity after clopidogrel administration on drug-eluting stent thrombosis. J Am Coll Cardiol 2007;49: Malinin A, Pokov A, Spergling M, et al. Monitoring platelet inhibition after clopidogrel with the VerifyNow-P2Y12 rapid analyzer: the VERIfy Thrombosis risk Assessment (VERITAS) study. Thromb Res 2007;119: Serebruany VL, Steinhubl SR, Berger PB, Malinin AI, Bhatt DL, Topol EJ. Variability in platelet responsiveness to clopidogrel among 544 individuals. J Am Coll Cardiol 2005;45: Van Werkum JW, van der Stelt CAK, Seesing TH, Hackeng CM, ten Berg JM. A head-to-head comparison between the VerifyNow P2Y12 assay and light transmittance aggregometry for monitoring the individual platelet response to clopidogrel in patients undergoing elective percutaneous coronary intervention. J Thromb Haemost 2006; 4: Michelson AD, Frelinger AL 3rd, Furman MI. Current options in platelet function testing. Am J Cardiol 2006;98:S Mahmud E, Ang L. Monitoring antiplatelet therapy during peripheral vascular and coronary interventions. Tech Vasc Interv Radiol 2006;9:

8 JACC Vol. 52, No. 13, 2008 September 23, 2008: Ang et al. Predictors of Platelet Inhibition With Clopidogrel Steinhubl SR. The VerifyNow System. In: Michelson AD, editor. Platelets. 2nd edition. San Diego, CA: Elsevier/Academic Press, 2006; Angiolillo DJ, Fernandez-Ortiz A, Bernardo E, et al. Variability in individual responsiveness to clopidogrel: clinical implications, management, and future perspectives. J Am Coll Cardiol 2007;49: Gurbel PA, Tantry US. Clopidogrel resistance? Thromb Res 2007; 120: Nguyen TA, Diodati JG, Chantal P. Resistance to clopidogrel: a review of the evidence. J Am Coll Cardiol 2005;45: Carr ME. Diabetes mellitus: a hypercoagulable state. J Diabetes Complications 2001;15: Murray JJ, Nowak J, Oates JA, et al. Platelet-vessel wall interactions in individuals who smoke cigarettes. Adv Exp Med Biol 1990;273: Badimon JJ, Badimon L, Turitto VT, et al. Platelet deposition at high shear rates is enhanced by high plasma cholesterol levels. In vivo study in the rabbit model. Arterioscler Thromb Vasc Biol 1991;11: Surya II, Akkerman J. The influence of lipoproteins on blood platelets. Am Heart J 1993;125: Feng D, Lindpaintner K, Larson MG, et al. Platelet glycoprotein IIIa Pl A polymorphism, fibrinogen, and platelet aggregability. The Framingham Heart study. Circulation 2001;104: Furman MI, Benoit SE, Barnard MR, et al. Increased platelet reactivity and circulating monocyte-platelet aggregates in patients with stable coronary artery disease. J Am Coll Cardiol 1998;31: Mahmud E, Cavendish JJ, Tsimikas S, et al. Elevated fibrinogen level predicts suboptimal response to therapy with both single- and doublebolus eptifibatide during percutaneous coronary intervention. J Am Coll Cardiol 2007;49: Han J, Lim CW, Watanabe N, et al. Reconstructing and deconstructing agonist-induced activation of integrin IIb 3. Curr Biol 2006;16: Malinin A, Pokov A, Swain L, Kotob M, Serebruany V. Validation of a VerifyNow P2Y12 cartridge for monitoring platelet inhibition with clopidogrel. Methods Find Exp Clin Pharmacol 2006;28: Schror K. Aspirin and platelets: the antiplatelet action of aspirin and its role in thrombosis treatment and prophylaxis. Semin Thromb Hemost 1997;23: Malanin AI, Spergling M, Muhlestein B, Steinhubl S, Serebruany V. Assessing aspirin responsiveness in subjects with multiple risk factors for vascular disease with a rapid platelet function analyzer. Blood Coagul Fibrinolysis 2004;15: Lowe GDO. Circulating inflammatory markers and risks of cardiovascular and non-cardiovascular disease. J Thromb Haemost 2005;3: Kristiansen OP, Mandrup-Poulsen T. Interleukin-6 and diabetes: the good, the bad, or the indifferent? Diabetes 2005;54:S Woodward M, Lowe GDO, Francis LMA, Rumley A, Cobbe SM, CADET Study Investigators. A randomized comparison of the effects of aspirin and clopidogrel on thrombotic risk factors and C-reactive protein following myocardial infarction: the CADET trial. J Thromb Haemost 2004;2: Angiolillo DJ, Bernardo E, Ramirez C, et al. Insulin therapy is associated with platelet dysfunction in patients with type 2 diabetes mellitus on dual oral antiplatelet treatment. J Am Coll Cardiol 2006;48: Ferriera IA, Mocking AI, Feijge MA, et al. Platelet inhibition by insulin is absent in type 2 diabetes mellitus. Arterioscler Thromb Vasc Biol 2006;26: Winocour PD, Watala C, Perry DW, Kinlough-Rathbone RL. Decreased platelet membrane fluidity due to glycation or acetylation of membrane proteins. Thromb Haemost 1992;68: Kishimoto T, Akira S, Narazaki M, Taga T. Interleukin-6 family of cytokines and gp130. Blood 1995;86: Angiolillo DJ, Fernandez-Ortiz A, Bernardo E, et al. Platelet aggregation according to body mass index in patients undergoing coronary stenting: should clopidogrel loading-dose be weight adjusted? J Invasive Cardiol 2004;16: Kelly RV, Hsu A, Topol E, Steinhubl S. The influence of body mass index on outcomes and the benefit of antiplatelet therapy following percutaneous coronary intervention. J Invasive Cardiol 2006;18: Saw J, Brennan DM, Steinhubl SR, et al., CHARISMA Investigators. Lack of evidence of a clopidogrel-statin interaction in the CHARISMA trial. J Am Coll Cardiol 2007;50: Key Words: fibrinogen y platelet aggregation y clopidogrel y diabetes mellitus y body mass index.