Author + information
- Received January 3, 2008
- Revision received April 18, 2008
- Accepted April 30, 2008
- Published online August 12, 2008.
- Kevin P. Bliden, BS⁎,
- Joseph DiChiara, BS⁎,
- Lookman Lawal, MD⁎,
- Anand Singla, MD⁎,
- Mark J. Antonino, BS⁎,
- Brian A. Baker, PharmD†,
- William L. Bailey, PharmD†,
- Udaya S. Tantry, PhD⁎ and
- Paul A. Gurbel, MD, FACC⁎,⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. Paul A. Gurbel, Sinai Center for Thrombosis Research, Shapiro Building, Room 209, 2401 West Belvedere Avenue, Baltimore, Maryland 21215.
Objectives The purpose of this study was to examine the effect of cigarette smoking on the platelet response to clopidogrel.
Background Response variability to clopidogrel therapy has been demonstrated. Clopidogrel is metabolically activated by several hepatic cytochrome P450 (CYP) isoenzymes, including CYP1A2. Cigarette smoking induces CYP1A2 and may, therefore, enhance the conversion of clopidogrel to its active metabolite.
Methods Among 259 consecutive patients undergoing elective coronary stenting; 120 were on chronic clopidogrel therapy and were not loaded; and 139 were clopidogrel naïve and were loaded with 600 mg. There were 104 current smokers (CS) and 155 nonsmokers (NS). The adenosine diphosphate (ADP)-stimulated platelet aggregation (PA) was assessed by conventional aggregometry. The ADP-stimulated total and active glycoprotein (GP) IIb/IIIa expression were assessed with flow cytometry. Low PA was defined as the lowest quartile of 5 μmol/l ADP-induced post-treatment PA.
Results Current smokers on chronic clopidogrel therapy displayed significantly lower PA and ADP-stimulated active GP IIb/IIIa expression compared with NS (p ≤ 0.0008 for both). Similarly, CS treated with 600 mg of clopidogrel displayed greater platelet inhibition and lower active GP IIb/IIIa expression compared with NS (p ≤ 0.05). In a multivariate Cox regression analysis, current smoking was an independent predictor of low PA (p = 0.0001).
Conclusion Clopidogrel therapy in CS is associated with increased platelet inhibition and lower aggregation as compared with NS. The mechanism of the smoking effect deserves further study and may be an important cause of response variability to clopidogrel therapy.
Clopidogrel, a prodrug, is metabolized by 2 consecutive cytochrome P450 (CYP)-dependent steps to its active metabolite, which binds irreversibly to the platelet P2Y12 receptor. The hepatic enzymes involved in the metabolism of clopidogrel include CYP1A2, 2B6, 2C9, 2C19, and 3A4/5 (1). High residual P2Y12 receptor reactivity is present in selected patients who are treated with clopidogrel, and high on-treatment platelet reactivity to adenosine diphosphate (ADP), as measured by conventional aggregometry, has been associated with post-percutaneous coronary intervention (PCI) ischemic event occurrence (2). Recently, patients on chronic clopidogrel therapy with high pre-PCI ADP-induced platelet aggregation were shown to be at increased risk for long-term post-PCI ischemic events (3).
Cigarette smoking is an inducer of CYP1A2 (4). Recently, it has been suggested that smoking influences the cardiovascular benefit of long-term therapy with clopidogrel and aspirin in PCI patients (5). However, there are no laboratory data to support a relation between increased platelet inhibition by clopidogrel in current smokers (CS) as compared with nonsmokers (NS). Moreover, there are no current data to demonstrate that the aforementioned reported clinical effect of clopidogrel treatment is related to better platelet inhibition in smokers versus nonsmokers.
The objective of the study was to investigate the effect of cigarette smoking on the platelet response to clopidogrel as measured by ADP-induced aggregation with conventional turbidometric aggregometry in platelet-rich plasma and by flow cytometry measurements of ADP-induced expression of activation-dependent receptors in patients treated with elective PCI. On the basis of previous observations, we hypothesized that 5 μmol/l ADP-induced aggregation would be lower in CS after loading and maintenance doses of clopidogrel compared with NS (6).
We analyzed platelet function data in 259 CS or NS undergoing elective PCI treated with clopidogrel consecutively enrolled in clinical trials at our center (7,8). There were 104 CS (smoking within 2 weeks of PCI) and 155 NS (no smoking within 1 year of PCI). Previous smokers were excluded because the influence of smoking on CYP1A2 activity is highly variable upon smoking cessation (9) and the interval between the last smoking event and enrollment in the study was inconsistent.
Patients who were on chronic clopidogrel therapy before PCI (n = 120, 75 mg/day) were not loaded; and clopidogrel-naïve patients (n = 139) were loaded with 600 mg immediately after PCI. Exclusion criteria were described previously (7,8). Blood was drawn before PCI from all patients. In patients loaded with clopidogrel, platelet studies were also conducted either at 18 to 24 h after PCI in patients not treated with eptifibatide (n = 63) or at ≥5 days after PCI in patients treated with eptifibatide (n = 76). Platelet aggregation was assessed by conventional aggregometry; ADP-stimulated total and active glycoprotein (GP) IIb/IIIa expression were assessed with flow cytometry (7,8).
Patient demographics and procedural characteristics were similar between the groups. However, CS had a greater prevalence of previous myocardial infarction (46% vs. 23%, p = 0.005) and treatment of bifurcation lesions (12% vs. 3%, p = 0.005), whereas NS more often had diabetes (33% vs. 48%, p = 0.01). No patients had been treated with immunosuppressants, antibiotics, or anti–human immunodeficiency virus agents that may influence CYP activity within 2 weeks of study.
Current smokers on chronic clopidogrel therapy displayed significantly lower 5 and 20 μmol/l ADP-induced platelet aggregation (32 ± 12% vs. 44 ± 13%, p < 0.0001; 43 ± 14% vs. 52 ± 17%, p = 0.008, respectively) and ADP-stimulated active GP IIb/IIIa expression compared with NS (7 ± 4 mean fluorescence intensity vs. 19 ± 18 mean fluorescence intensity, p = 0.0007). Relative inhibition of platelet aggregation (IPA) was calculated in patients receiving a clopidogrel loading dose from the measured maximum platelet aggregation (MPA) where IPA = [(MPApredose − MPApostdose)/MPApredose] × 100%. Current smokers treated with 600 mg of clopidogrel displayed greater IPA (5 μmol/l ADP, 46 ± 23% vs. 34 ± 21% p = 0.002 and 20 μmol/l ADP, 38 ± 6% vs. 31 ± 21%, p = 0.04) and greater relative inhibition of active GP IIb/IIIa expression compared with NS (58 ± 24% vs. 40 ± 36%, p = 0.05). An analysis of variance in patients on chronic clopidogrel therapy demonstrated significantly lower 5 μmol/l and 20 μmol/l ADP-induced platelet aggregation in patients currently smoking ≥1/2 pack/day compared with NS and patients currently smoking <1/2 pack/day (p < 0.05) (Figs. 1A and 1B). However, no significant differences in platelet aggregation were observed between NS and CS smoking <1/2 pack/day.
A logistic regression analysis evaluated the impact of age, body mass index, diabetes, history of myocardial infarction, hypertension, smoking status, and use of calcium antagonists, angiotensin-converting enzyme inhibitors, beta blockers, and statins on low post-treatment platelet aggregation (≤29%), defined as the lowest quartile of 5 μmol/l ADP-induced platelet aggregation after clopidogrel therapy (MedCalc Software, Mariakerke, Belgium). Current smoking was the only variable significantly associated with low post-treatment platelet aggregation (odds ratio 2.63, p < 0.007).
The current analysis demonstrates a positive relation between current cigarette smoking and the enhancement of the antiplatelet effect of clopidogrel administered to patients undergoing nonemergent coronary stenting. The relation between clopidogrel effect and current smoking was robust in patients who were treated with a chronic maintenance dose who were not loaded and in clopidogrel-naïve patients administered a 600-mg loading dose. The measurements of platelet function used in our study are the most widely described in the literature to assess the platelet response to clopidogrel.
In the current study, in a multivariate analysis, we demonstrated the significant effect of smoking on platelet function during clopidogrel therapy. These data are consistent with the first observations of clopidogrel response variability in a study of 96 patients treated with 300 mg of clopidogrel for elective coronary artery stenting (10). In the latter investigation, we observed that 13% of the nonresponders to clopidogrel were smokers as compared with 28% of the responders. In addition, Matetzky et al. (6) reported that smokers were less often clopidogrel resistant. Finally, we observed that the amount of cigarettes smoked appeared to affect the response to clopidogrel.
In patients who smoked <1/2 pack/day treated with chronic clopidogrel therapy, we did not observe an effect of smoking on ADP-induced platelet aggregation. It has been suggested that cigarette smoking induces CYP1A2 activity in a “dose-related” manner. In normal volunteers, CYP1A2 activity was increased relative to the number of cigarettes smoked per day, with a 1.22-, 1.47-, 1.66-, and 1.72-fold increase in activity observed with smoking 1 to 5, 6 to 10, 11 to 20, and >20 cigarettes/day, respectively (11).
The data from our study are consistent with these previous observations and suggest that the number of cigarettes smoked is an important consideration when evaluating potential smoking effects on PA in clopidogrel-treated patients. Our hypothesis is that significant cigarette smoking (i.e., ≥10 cigarettes/day) affects platelet inhibition in clopidogrel-treated patients through the induction of CYP1A2. The aforementioned results may provide an explanation for the cardiovascular event reduction observed in smokers compared with NS treated with clopidogrel plus aspirin in a subanalysis of the CREDO (Clopidogrel for the Reduction of Events During Observation) trial (5).
Because our study was a post-hoc analysis, we cannot rule out the influence of selection bias or other variables that may affect the hepatic metabolism of clopidogrel and were not recorded in the case report form such as use of caffeine, proton pump inhibitors, or herbal agents. To address these concerns, a well-controlled prospective study is needed.
Cigarette smoking influences platelet responsiveness to clopidogrel. The mechanism of the effect deserves further study and may further elucidate the causes of response variability to clopidogrel and the clinical efficacy of clopidogrel in specific patient groups.
This study was supported by a grant from Daiichi Sankyo, Inc. and Sinai Hospital of Baltimore, Maryland.
- Abbreviations and Acronyms
- adenosine diphosphate
- current smokers
- consecutive cytochrome P450
- inhibition of platelet aggregation
- maximum platelet aggregation
- percutaneous coronary intervention
- Received January 3, 2008.
- Revision received April 18, 2008.
- Accepted April 30, 2008.
- American College of Cardiology Foundation
- Gurbel P.A.,
- Becker R.C.,
- Mann K.G.,
- Steinhubl S.R.,
- Michelson A.D.
- Bliden K.P.,
- DiChiara J.,
- Tantry U.S.,
- Bassi A.K.,
- Chaganti S.K.,
- Gurbel P.A.
- Saraff K.Y.,
- Steinhubl S.R.,
- Hsu A.P.,
- Topol E.J.
- Matetzky S.,
- Shenkman B.,
- Guetta V.,
- et al.
- Gurbel P.A.,
- Bliden K.P.,
- Guyer K.,
- et al.
- Gurbel P.A.,
- Bliden K.P.,
- Zaman K.A.,
- Yoho J.A.,
- Hayes K.M.,
- Tantry U.S.
- Gurbel P.A.,
- Bliden K.P.,
- Hiatt B.L.,
- O'Connor C.M.