The impact of smoking on the antiplatelet action of clopidogrel in non-ST-elevation myocardial infarction patients: results from the ISAR-REACT 4 platelet substudy

Authors


Dirk Sibbing, Deutsches Herzzentrum München and 1. Medizinische Klinik rechts der Isar, Technische Universität München, Lazarettstrasse 36, 80636 München, Germany.
Tel.: +49 89 1218 0; fax: +49 89 1218 4013.
E-mail: dirk@sibbing.net

Cigarette smoking constitutes a major health hazard that contributes significantly to cardiovascular disease progression [1]. Among coronary artery disease patients undergoing percutaneous coronary intervention (PCI), some are active smokers at the time of the procedure. Along with the PCI procedure, these patients are recommended to stop smoking (class IB recommendation) [2] and to strictly adhere to the intake of a dual antiplatelet treatment consisting of aspirin and a P2Y12 receptor inhibitor such as clopidogrel [2]. The antiplatelet response to the thienopyridine derivative clopidogrel is highly variable, and an insufficient response – also termed high on-treatment platelet reactivity (HPR) – has been linked to a higher risk of thrombotic events in observational studies [3]. Numerous clinical variables, including diabetes mellitus and body mass index, were discovered to be relevant factors causing HPR [3,4]. The first study that investigated the issue of smoking and response to clopidogrel treatment reported amplified platelet inhibition in clopidogrel-treated smokers [5]. These results were explained by a possible influence of smoking on the activity of cytochrome P450 1A2, an enzyme that – among others – may impact on in vivo clopidogrel bioactivation. Meanwhile, numerous studies have been published in support of [5–7] but also against [4,8,9] an association between smoking and amplified platelet inhibition in clopidogrel-treated patients. The largest collaborative analysis so far [8] on this subject, in ∼ 4500 clopidogrel-treated patients with ex vivo assessment of clopidogrel response by different methods for platelet function testing, failed to confirm the association between smoking and an enhanced response to clopidogrel. Thus, the available data with regard to this issue are highly conflicting. Specifically for patients with acute coronary syndrome (ACS), a condition in which blood platelets are known to be highly activated and the antiplatelet drug response is attenuated [10,11], data on the influence of smoking and the response to clopidogrel treatment are limited. To close this gap in our knowledge, the primary objective of this study was to assess the impact of smoking on the antiplatelet action of clopidogrel in ACS patients with a non-ST-elevation myocardial infarction (NSTEMI).

All patients included in this study were participants in the Intracoronary Stenting and Antithrombotic Regimen: Rapid Early Action for Coronary Treatment 4 (ISAR-REACT 4) trial [12] (ClinicalTrials.gov number NCT00373451). The cohort of patients investigated here is identical to the patient cohort of the ISAR-REACT 4 platelet substudy [13]. Patients were recruited at the Deutsches Herzzentrum München and the Klinikum rechts der Isar (Technische Universität München, Munich, Germany). Detailed information on the study design, modalities for platelet function testing, inclusion and exclusion criteria and baseline characteristics of the patients have been reported previously [12,13]. In brief, all patients in this study received a single high intravenous loading dose (LD) of 600 mg of clopidogrel and 500 mg of aspirin in preparation for the PCI procedure. For the subsequent assessment of post-clopidogrel LD platelet aggregation values, whole blood for platelet function testing on the Multiplate analyzer (Verum Diagnostika, Munich, Germany) was obtained in the catheterization laboratory from the arterial sheath of all patients immediately before PCI and prior to the administration of any adjunct antithrombotic treatment (abciximab with unfractionated heparin or bivalirudin). ADP-induced platelet aggregation (quantified as aggregation units [AU] × min) was measured and compared between the groups of active smokers and non-smokers. In addition, the presence or absence of HPR was determined for both groups. The definition of HPR (≥ 468 AU × min) was based on the consensus document [3] of the Working Group on High On-Treatment Platelet Reactivity. Platelet aggregation values and the proportions of patients with and without HPR were compared between groups with a two-sided unpaired Wilcoxon test and the chi-square test, respectively. All analyses were performed with the software package s-plus (TIBCO Software., Palo Alto, CA, USA). For all statistical analyses, a P-value of < 0.05 was considered to be statistically significant.

This focused post hoc analysis of the ISAR-REACT 4 platelet substudy (n = 564 NSTEMI patients) included patients with a mean age of 67.7 years, and a proportion of women of 22% (n = 126). Among the 564 patients, 124 (22%) were active smokers and 440 (78%) were non-smokers at the time of the index PCI with platelet function testing. The baseline characteristics of smokers and non-smokers are shown in Table S1. ADP-induced platelet aggregation (median [interquartile range]) did not differ significantly between active smokers (365 [156–623] AU × min) and non-smokers (299 [178–578] AU × min; P = 0.20). Figure 1A shows individual on-treatment measurements in both study groups. In the entire cohort, a total of 205 patients (36%) showed HPR at the time of the PCI procedure. Figure 1B shows the proportions of patients with and without HPR among active smokers and non-smokers; these did not differ significantly between the two study groups (P = 0.09).

Figure 1.

 Platelet aggregation and high on-treatment platelet reactivity (HPR) in smokers and non-smokers. (A) Platelet aggregation measurements following a 600-mg loading dose of clopidogrel for active smokers and non-smokers. The dots denote individual measurements. The lines represent the median values per group. (B) The rates of HPR in active smokers and non-smokers. AU, aggregation units.

To the best of our knowledge, this is the first study to specifically assess the impact of smoking on the antiplatelet action of clopidogrel in NSTEMI patients undergoing PCI. The key result of this study is that smoking showed no significant influence on the antiplatelet action of clopidogrel in this high-risk cohort of NSTEMI patients. In contrast to prior observations [5], platelet aggregation values and the rates of HPR were numerically higher in smokers than in non-smokers. It is well recognized that ACS patients are very special with regard to their platelet physiology [11], and thus deserve particular attention. Our results suggest that the concept of amplified platelet inhibition in clopidogrel-treated patients [5,14] cannot be extrapolated to ACS patients presenting with an NSTEMI.

Besides conflicting pharmacodynamic data on smoking and clopidogrel, a recently published viewpoint article [14] on this important topic summarized clinical data from randomized trials in support of a greater benefit of clopidogrel in active smokers. However, like ACS patients, active smokers must be understood as a cohort of patients with specific properties in terms of their platelet physiology and risk for thrombotic events. In the absence of any antiplatelet drugs, smokers are known to exhibit high levels of platelet reactivity [15], and are thus more prone to suffer from thrombotic events than non-smokers [16]. In consequence, smokers may derive the greatest benefit from any antiplatelet drug that they receive. In addition, the ‘smoking paradox’ that is commonly seen in randomized trials irrespective of the investigated treatment (e.g. thrombolysis [17]) must be considered here. Explanations for this paradox include the facts that smokers are younger (see also Table S1) and have lower rates of other cardiovascular risk factors [17].

Some limitations of our study merit mention. First, this analysis was a post hoc analysis of a study population that stems from a randomized trial [12], and is therefore subject to the limitations inherent in all such analyses. In general, nearly all studies on this subject, including our study presented here, are limited by the fact that they only included pharmacodynamic measurements obtained with different methods for ex vivo platelet function testing. In contrast to this, Yousef et al. [9] investigated the pharmacokinetics of clopidogrel in 26 healthy volunteers. Measurements of the carboxyl metabolite of clopidogrel revealed lower levels in smokers than in non-smokers, suggesting decreased in vivo bioactivation of clopidogrel in the smoking cohort of subjects. In this regard, further pharmacokinetic studies investigating clopidogrel-treated CAD patients are urgently needed.

In conclusion, smoking did not amplify the antiplatelet action of clopidogrel in NSTEMI patients undergoing PCI. Thus, ACS patients who are active smokers do not derive any benefit from smoking with respect to their response to clopidogrel treatment. Further studies on this subject are urgently needed to resolve the existing contradictions with regard to smoking and clopidogrel responsiveness.

Disclosure of Conflict of Interests

This work was supported in part by Nycomed Pharma, Unterschleissheim, Germany (former distributor of bivalirudin in Europe), and a grant (KKF 04-06 [974404]) from Deutsches Herzzentrum, Munich, Germany. Material for platelet function testing was, in part, provided free of charge by Verum Diagnostika. The sponsors had no role in the design and conduct of the study, collection, management, analysis, and interpretation of the data, or preparation, review or approval of the manuscript. D. Sibbing reports receiving fees for advisory board activities from Verum Diagnostika and Eli Lilly, and speaker fees from Astra Zeneca. A. Kastrati reports receiving consulting fees from AstraZeneca, Bristol-Myers Squibb, and Daiichi Sankyo/Eli Lilly, and lecture fees from Abbott, Biotronik, Cordis, and Medtronic. J. Mehilli reports receiving lecture fees from Abbott, Daiichi Sankyo/Eli Lilly, and Terumo.

Ancillary