Prior smoking status, clinical outcomes, and the

comparison of ticagrelor with clopidogrel in acute

coronary syndromesInsights from the PLATelet
inhibition and patient Outcomes (PLATO) trial
Jan H. Cornel, MD, PhD, a ,n Richard C. Becker, MD, b,n Shaun G. Goodman, MD, MSc, c,n
Steen Husted, MD, DSc, d,n Hugo Katus, MD, e,n Anwar Santoso, MD, PhD, f,n Gabriel Steg, MD, g,h,i,n
Robert F. Storey, MD, DM, j,n Marius Vintila, MD, PhD, k,n Jie L. Sun, MS, b,n Jay Horrow, MD, MS, l,n
Lars Wallentin, MD, PhD, m,n Robert Harrington, MD, b,n and Stefan James, MD, PhD m,n Alkmaar, Netherlands;
Durham, NC; Ontario, Canada; rhus, Denmark; Heidelberg, Germany; Jakarta, Indonesia; Paris, France;
Sheffield, UK; Bucharest, Romania; Wilmington, DE; and Uppsala, Sweden

Background Habitual smoking has been associated with increased platelet reactivity, increased risk of thrombotic
complications and greater efficacy of clopidogrel therapy over placebo. In the PLATO trial, ticagrelor compared to clopidogrel
in patients with acute coronary syndromes (ACS) reduced the primary composite end point of vascular death, myocardial
infarction and stroke, without increasing overall rates of major bleeding. We evaluated the results in relation to smoking habits.
Methods Interactions between habitual smokers (n = 6678) and in ex/nonsmokers (n = 11,932) and the effects of
randomized treatments on ischemic and bleeding outcomes were evaluated by Cox regression analyses.
Results Habitual smokers had an overall lower risk profile and more often ST-elevation ACS. After adjustment for baseline
imbalances, habitual smoking was associated with a higher incidence of definite stent thrombosis (adjusted HR, 1.44 [95% CI,
1.07-1.94]); there were no significant associations with other ischemic or bleeding end points. The effects of ticagrelor
compared to clopidogrel were consistent for all outcomes regardless of smoking status. Thus, there was a similar reduction in
the primary composite end point for habitual smokers (adjusted HR, 0.83 [95% CI, 0.68-1.00]) and ex/nonsmokers (adjusted
HR, 0.89 [95% CI, 0.79-1.00]) (interaction P = .50), and in definite stent thrombosis for habitual smokers (adjusted HR, 0.59
[0.39-0.91]) and ex/nonsmokers (adjusted HR, 0.69 [95% CI, 0.45-1.07]) (interaction P = .61).
Conclusions In patients hospitalized with ACS, habitual smoking is associated with a greater risk of subsequent stent
thrombosis. The reduction of vascular death, myocardial infarction, stroke, and stent thrombosis by ticagrelor compared to
clopidogrel is consistent regardless of smoking habits. (Am Heart J 2012;0:1-9.e1.)

From the aMedisch Centrum Alkmaar, Alkmaar, Netherlands, bDuke Clinical Research Smoking has several adverse effects, which inuence
Institute, Durham, NC, cDivision of Cardiology, University of Toronto, Ontario, Canada, cardiovascular physiology and increase the risk of cardio-
d
Department of Cardiology, rhus University Hospital, rhus, Denmark, eUniversittskli-
f
vascular disease including increased platelet reactivity,
nikum Heidelberg, Heidelberg, Germany, Department of Cardiology, Faculty of Medicine,
University of Indonesia, National Cardiovascular Centre, Harapan Kita Hospital, Jakarta,
proinammatory responses, 1-3 endothelial dysfunction,
Indonesia, gINSERM U-698, Paris, France, hUniversit Paris-Diderot 7, Paris, France, dyslipidemia, and insulin resistance. 4,5 Habitual smokers
i
Hpital Bichat, AP-HP, Paris, France, jUniversity of Sheffield, Sheffield, UK, kCarol Davila develop coronary artery disease at a younger age and
University, Bucharest, Romania, lAstra Zeneca R&D, Wilmington, DE, and mDepartment
of Medical Sciences and Uppsala Clinical Research Center, Uppsala University,
appear particularly sensitive to anti-platelet therapy. 6 In
Uppsala, Sweden. response to clopidogrel, habitual smokers have greater
n
for the PLATO study group. inhibition of platelet aggregation compared to non-
Clinical Trial Registration Information: www.clinicatrials.gov: NCT00391872.
smokers. 7 The higher platelet reactivity in smokers may
Steven R. Steinhubl, MD, served as guest editor for this article.
Submitted February 17, 2012; accepted June 7, 2012.
lead to greater benet from more intensive platelet
Reprint requests: Jan H Cornel, MD, PhD, Medisch Centrum Alkmaar, Wilhelminalaan 12, inhibitory therapy. Habitual smoking may enhance the
1815 JD Alkmaar, Netherlands. inhibitory effects of clopidogrel leading to less hyporespon-
E-mail: j.h.cornel@mca.nl
siveness, 8 possibly by increasing active metabolite forma-
0002-8703/$ - see front matter
2012, Mosby, Inc. All rights reserved. tion though induction of the CYP1A2 enzyme. 9 Conicting
doi:10.1016/j.ahj.2012.06.005 data exist whether clinical outcome of patients on

2 Cornel et al
clopidogrel is inuenced by smoking status, especially in
patients with an acute coronary syndrome (ACS). 6,10,11
Ticagrelor is an oral non-thienopyridine P2Y12 inhibit-
ing agent with a reversible and direct action on the
receptor that provides faster, greater and more consistent
platelet inhibition than clopidogrel. 12,13 The PLATelet
inhibition and patient Outcomes (PLATO) trial showed
that ticagrelor was superior to clopidogrel for the
prevention of vascular death, myocardial infarction (MI)
or stroke without a signicant increase in overall major
bleeding rates in a broad population of patients with
ACS. 14 We report the relationship between smoking
habits and efcacy and safety outcomes and any
interactions between smoking habits and the effects of
ticagrelor versus clopidogrel.
Methods
The PLATO trial enrolled 18,624 patients between October
2006 and July 2008 (Clinical Trial Registration Information: www.
clinicatrials.gov: NCT00391872). Details regarding the study
design, patients, outcome denitions, and results have previously
been published. 14,15 Patients were included if they were
hospitalized for potential ST-segment elevation or nonST-
segment elevation ACS, with symptom onset during the previous
24 hours. For nonST-segment elevation ACS, at least 2 of the
following 3 criteria were required: (1) ST-segment depression or
transient elevation 1 mm in 2 or more contiguous leads; (2)
positive biomarker indicating myocardial necrosis; or (3) one
additional risk indicator: age over 60 years, previous MI or
coronary artery bypass graft (CABG), carotid artery disease,
previous ischemic stroke, transient ischemic attack, carotid
stenosis or cerebral revascularization, diabetes mellitus, peripheral
artery disease or chronic kidney disease (CrCl b 60 mL/min
estimated by Cockcroft Gault equation). For ST-segment elevation
ACS, inclusion required a planned primary percutaneous coronary
intervention (PCI). The most important exclusion criteria were a
contraindication to clopidogrel, brinolytic therapy within
24 hours, need for oral anticoagulation therapy, need for dialysis,
and clinically important anemia or thrombocytopenia.
After written informed consent, patients were randomly
assigned to ticagrelor or clopidogrel before any PCI procedure
was performed. Ticagrelor was given in a loading dose of
180 mg followed by 90 mg twice daily. Patients randomized to
clopidogrel who had not received an open-label loading dose
and had not been taking clopidogrel for at least 5 days before
randomization, received a 300 mg loading dose of clopidogrel
study drug followed by 75 mg daily. Others continued a
maintenance dose of 75 mg daily as clopidogrel study drug.
All patients received aspirin unless intolerant. A daily dose of
75 to 100 mg was recommended but doses up to 325 mg daily
was allowed for 6 months after stent placement. The median
duration of study treatment was 9.1 months.
The primary efcacy variable was time to rst occurrence of
any event from the composite of death from vascular causes, MI,
and stroke. Secondary efcacy variables included the individual
end points of MI, vascular death, stroke, and all-cause mortality.
The primary safety variable was the time to rst occurrence of
any PLATO dened major bleeding.
American Heart Journal
Month Year
Smoking status
At randomization smoking behavior was assessed. Patients
were classied as habitual smokers if they reported smoking 1
or more cigarettes, cigars, or equivalent tobacco per day;
ex-smokers if they had smoked previously (stopped N 1 month
ago) or if they reported to smoke less than 1 cigarette, cigar, or
equivalent tobacco per day; and nonsmokers if they reported
not to smoke, either currently or previously. Owing to very
similar baseline characteristics and outcomes in ex-smokers
and nonsmokers, we merged these groups for the nal
analyses. Thus, we present smoking status in two groups for
the purpose of this analysis: habitual smokers versus ex/
nonsmokers. Information about the amount of tobacco
equivalents per day was not available. After an ACS event,
patients may decide to quit smoking. To examine the potential
for intervening events changing the relationship of baseline
factor with outcomes, we additionally evaluated the associa-
tion between baseline smoking status and events within the
rst 10 days of follow-up.
Statistics
Patient characteristics were compared by smoking status
using Pearson chi-square tests for categorical variables and
Wilcoxon rank sum tests for continuous variables. Kaplan-
Meier estimates were plotted by treatment group and smoking
status. Log-Rank test was used to compare event rates
between groups.
Unadjusted and adjusted multivariable Cox proportional
hazards regression models were used to evaluate the relation-
ships between smoking groups, treatment groups, and clinical
outcomes. The factors considered in modeling each of the eight
clinical outcomes were identied using clinical judgment and a
literature search of previously developed models.
Model selection utilized a backward algorithm repeated in
200 bootstrap samples, with a 0.05 signicance level required to
retain the variable in the model. The nal model included
those variables selected in 75% of the 200 bootstrap runs.
Proportional hazards assumption and linearity assumption were
tested. Variable transformations were implemented to resolve
violations of the linearity assumption. Adjusted models included
the pre-specied baseline risk factors, smoking, and randomized
treatments (see the Appendix). The interaction of smoking and
randomized treatment was included into these models to
evaluate whether the treatment effect varied according to
smoking status. Subgroups based on intended treatment strategy
and the presence or absence of ST elevation at entry were
analyzed using Cox models with 3-way interaction terms for
smoking, treatment, and subgroup.
All analyses were performed according to the intention-
to-treat principle, using SAS version 9.1, with 2-sided .05
signicance levels.
The PLATO study was funded by AstraZeneca. Support for
the analysis and interpretation of results and preparation of
the current manuscript was provided through funds to the
Uppsala Clinical Research Center and Duke Clinical Research
Institute as part of the Clinical Study Agreement. The authors are
solely responsible for the design and conduct of this study, all
study analyses, the drafting and editing of the manuscript, and
its nal contents.
American Heart Journal
Volume 0, Number 0
Cornel et al 3

Table I. Baseline characteristics by smoking status

Habitual smoker n = 6678 Ex/nonsmoker n = 11,932 P n = 18,610

Age, median (quartiles) 56 (49-62) 66 (58-73) b.001

Male, Gender, n (%) 5457 (81.7) 7869 (65.9) b.001
Body weight, kg, median (quartiles) 80 (70-90) 80 (70-90) .001
Body weight b 60 kg, n (%) 430 (6.4) 881 (7.4)
BMI, median (quartiles) 26.9 (24.3-30.0) 27.6 (25.0-30.7) b.001
Race, (n) %
White 6072 (90.9) 10,992 (92.1) .035
Black 85 (1.3) 144 (1.2)
Oriental 435 (6.5) 661 (5.5)
Other 86 (1.3) 135 (1.1)
Region, n (%)
North America 622 (9.3) 1191 (10.0) .141
Cardiovascular risk factors and history, n (%)
Hypertension 3573 (53.5) 8609 (72.2) b.001
Dyslipidemia 2818 (42.2) 5871 (49.2) b.001
Diabetes mellitus 1156 (17.3) 3506 (29.4) b.001
Angina pectoris 2402 (36.0) 5956 (49.9) b.001
Myocardial infarction 1072 (16.1) 2751 (23.1) b.001
Congestive heart failure 206 (3.1) 843 (7.1) b.001
Percutaneous coronary intervention 676 (10.1) 1815 (15.2) b.001
Coronary artery bypass graft 204 (3.1) 902 (7.6) b.001
Transient ischemic attack 137 (2.1) 362 (3.0) b.001
Nonhemorrhagic stroke 171 (2.6) 551 (4.6) b.001
Peripheral arterial disease 434 (6.5) 710 (6.0) .135
Chronic kidney disease 130 (1.9) 655 (5.5) b.001
Medication at randomization, n (%)
Clopidogrel pretreatment 3193 (47.8) 5561 (46.6) .113
Aspirin 6254 (93.7) 11,170 (93.7) .955
-Blockade 4448 (66.7) 8548 (71.7) b.001
ACE-inhibition and/or ARB 3825 (57.3) 7956 (66.7) b.001
Cholesterol lowering (Statin) 6356 (95.2) 11,101 (93.1) b.001
Ca-inhibitor 671 (10.1) 2037 (17.1) b.001
Diuretic 1119 (16.8) 3203 (26.9) b.001
Proton pump inhibitors 2307 (34.6) 4065 (34.1) .515
Baseline labs, median (quartiles)
Creatinine, mol/L 80 (71-88) 88 (71-97.2) b.001
Creatinine clearance, ml/min 91.2 (75.7-106.6) 77.8 (63.3-90.9) b.001
Glucose, mmol/L 6.7 (5.7-8.4) 6.9 (5.8-9.0) b.001
Hemoglobin A1c, % 5.9 (5.6-6.4) 6.0 (5.7-6.8) b.001
Total Cholesterol, mmol/L 5.3 (4.5-6.1) 5.1 (4.2-5.9) b.001
Hemoglobin, mg/L 144 (134-153) 137 (127-147) b.001
Hematocrit, % 0.4 (0.4-0.5) 0.4 (0.4-0.4) b.001
NT-proBNP, ng/L 45.3 (14.8-137.0) 68.9 (22.2-216.4) b.001
First (central laboratory) troponin I, g/L 2.9 (0.4-16.2) 1.7 (0.2-10.0) b.001
Antithrombotic medication during index hospitalization, n (%)
Aspirin 6512 (97.6) 11,561 (97.0) .016
GP IIb/IIIa inhibitors 2133 (32.0) 2904 (24.4) b.001
Unfractionated heparin 4146 (62.1) 6630 (55.6) b.001
Low-molecular-weight heparin 3314 (49.7) 6334 (53.1) b.001
Fondaparinux 157 (2.4) 352 (3.0) .016
Bivalirudin 127 (1.9) 247 (2.1) .432

ARB, Angiotensin receptor blockers; NT-proBNP, N-terminal pro-Brain natriuretic peptide.

Results less often had hypertension, diabetes mellitus, chronic

Patients
Among the 18,610 patients with known smoking status
at randomization in the PLATO study, 6678 (36%) were
reported as being habitual smokers (Table I). Habitual
smokers had an overall lower risk prole. They were
younger (56 vs 66 years) and more likely to be men, and
kidney disease, or a history of cardiovascular disease.
Most habitual smokers were already on cardiovascular
medication although less frequently than in non/ex
smokers. Habitual smokers more often presented with
an ST-elevation MI (Table II) and an invasive strategy
was intended and used more commonly (PCI 73.8%
 American Heart Journal
4 Cornel et al Month Year

Table II. Type of index event and treatment

Habitual smoker Ex/non smoker
n = 6678 n = 11,932 P

Type of ACS, n (%) b.001

ST-elevation MI 3304 (49.5) 3722 (31.3)
Non ST-elevation ACS 3374 (50.5) 8210 (68.7)
b.001
Intention for invasive treatment 5349 (80.1) 8047 (67.4)
ST-elevation MI 93.9 89.1
Non ST-elevation ACS 65.8 55.9
Intention for non invasive treatment 1329 (19.9) 3885 (32.6)
Risk indicators STEMI, n (%)
Killip class N2 35 (1.0) 52 (1.3) .358
TIMI STEMI risk score N2 896 (30.7) 1808 (56.2) b.001
Risk indicators nonST-elevation ACS, n (%)
Troponin positive 2772 (87.7) 5894 (78.3) b.001
ST-segment depression N0.1 mV 1849 (56.6) 4509 (58.0) .162
TIMI nonSTE-ACS risk score N4 1194 (42.5) 3086 (44.1) .146
PCI during study 4926 (73.8) 7051 (59.1) b.001

Table III. Association of habitual smoking with end points

Unadjusted HR Adjusted HR
Characteristic Event Unadjusted 2 (95% CI) P Adjusted 2 (95% CI) P

Vascular death, MI, or stroke 1878 33.40 0.75 (0.68-0.83) b.0001 1.34 1.08 (0.95-1.22) .25
Vascular death, MI, or stroke (first 10 days) 652 3.94 0.85(0.72-0.99) .047 0.42 1.07 (0.87-1.33) .52
All cause death 905 31.68 0.66 (0.57-0.76) b.0001 0.64 1.08 (0.90-1.30) .42
Vascular death 795 30.89 0.64(0.55-0.75) b.0001 0.004 1.01(0.83-1.23) .95
Cancer related mortality 27
Vascular death/MI 1723 29.56 0.75 (0.68-0.83) b.0001 1.90 1.10 (0.96-1.25) .17
Major bleeding 1890 29.53 0.76 (0.69-0.84) b.0001 0.46 0.96 (0.85-1.08) .50
Non-CABG related major bleeding 668 17.78 0.70 (0.59-0.83) b.0001 1.39 1.14 (0.92-1.41) .24
Definite stent thrombosis 177 7.55 1.51 (1.13-2.03) .006 5.65 1.44 (1.07-1.94) .02
Definite stent thrombosis (first 10 days) 107 3.42 1.43 (0.98-2.09) .06 1.88 1.31 (0.89-1.92) .17
Any stent thrombosis 276 0.36 1.08 (0.88-1.37) .547 2.629 1.24 (0.96-1.62) .10

Of the 11,283 patients who received a stent, 4,691 were habitual smokers.

vs 59.1%). Furthermore, habitual smokers had a lower
NT-proBNP.
These and other baseline characteristics, medications
and procedures were well matched between the
randomized treatment groups (ticagrelor vs clopidogrel)
within both smoking strata (data not shown).
Outcomes in relation to smoking status
In univariable analysis habitual smoking, at the time
of randomization, was associated with all evaluated
efcacy and bleeding end points. Habitual smokers had
a lower incidence of most end points, but the
incidence of denite stent thrombosis and any stent
thrombosis was higher. However, after adjustment for
the differences in baseline characteristics, habitual
smoking was associated with a 44% higher incidence
of denite stent thrombosis (P = .018) while no other
associations persisted (Table III). In addition, the
associations between baseline smoking status and
either primary composite end point or stent thrombo-
sis at 10 days of follow-up showed similar results with
a 31% higher incidence of denite stent thrombosis
(P = .17).
Outcomes in relation to smoking status and
randomized treatment
Ticagrelor reduced the primary composite end point,
vascular death/MI, all cause mortality, and stent
thrombosis consistently in both groups. If anything,
the reductions in event rates were numerically
greater in habitual smokers than in ex or nonsmokers
(Figure 1, Figure 2). In habitual smokers, there was a
2.1% absolute (17% relative; 95% CI, 0.68-1.00)
reduction of the primary composite end point, a
2.7% absolute (24% relative; 95% CI, 0.63-0.93)
reduction of vascular death/MI, and a 1% absolute
American Heart Journal
Volume 0, Number 0
Figure 1
Forest plot of efficacy and bleeding outcomes. Adjusted primary and secondary efficacy and bleeding outcomes in relation to smoking status and
treatment allocation.
(41% relative; 95% CI, 0.39-0.91) reduction of denite
stent thrombosis. In ex/nonsmokers the corresponding
results were a 1.4% absolute (11% relative; 95% CI,
0.79-1.00) reduction of the primary composite end
point, a 1.4% absolute (13% relative; 95% CI, 0.77-0.98)
reduction of vascular death/MI, and a 0.4% absolute
(31% relative; 95% CI, 0.45-1.07) reduction of denite
stent thrombosis. No smoking status-by-treatment
interaction was found (Figure 1). Moreover the
landmark analysis of the treatment effects at 10 days
of follow-up were similar with early separation of the
event curves, especially in smokers (Figure 3).
Cornel et al 5
Habitual smoking was not associated with an increase
in the hazard for major bleeding and there was no
signicant interaction between smoking habits and the
effects of the treatment on major bleeding (Table III).
The adjusted hazard ratio for the rate of all PLATO-
dened major bleeding for ticagrelor versus clopidogrel
in habitual smokers was 1.18 (95% CI, 0.98-1.43) and
1.04 (95% CI, 0.92-1.18) in ex/nonsmokers. For non-
CABG related major bleeding events the corresponding
adjusted hazard ratios for ticagrelor versus clopidogrel
were 1.22 (95% CI, 0.88-1.69) in habitual smokers and
1.35 (95% CI, 1.10-1.67) in ex/nonsmokers.
 American Heart Journal
6 Cornel et al Month Year

Figure 2

Adjusted survival plots. Adjusted cumulative incidence of: A, the primary composite of vascular death, myocardial infarction and stroke; B, definite
stent thrombosis and in the clopidogrel (dotted lines) and ticagrelor (solid lines) groups stratified for smoking status at the time of randomization.
Ticag, tigacrelor; Clop, clopidogrel.
American Heart Journal
Volume 0, Number 0
For all outcomes, the comparisons of treatment effects
in relation to smoking habits were consistent across
various subgroups with no interactions, including for
type of ACS (ST-elevation MI or non-ST elevation ACS) or
initial intended treatment strategy (non-invasive or
invasive treatment) (data not shown).
Discussion
The present subgroup analysis of the PLATO trial shows
that in patients hospitalized with ACS, habitual smoking is
associated with a greater risk of subsequent stent
thrombosis. The reduction of vascular death, myocardial
infarction, stroke and stent thrombosis by ticagrelor
compared to clopidogrel is consistent regardless of
smoking habits. We did not nd any attenuation of the
relative benet of ticagrelor vs. clopidogrel treatment on
efcacy and safety end points in smokers versus ex/non-
smokers supporting a potentiation of the clopidogrel
treatment effect.
This analysis conrms that habitual smokers presenting
with ACS are younger and have a lower cardiovascular
risk prole than ex-/nonsmokers. These differences in
baseline risk likely account for the signicantly lower
event rate of the primary composite efcacy end point,
previously described as the smokers paradox. 16 After
correction for the differences in baseline characteristics,
the risk for recurrence of cardiovascular events did not
differ signicantly between habitual and ex/nonsmokers,
except for denite stent thrombosis which was more
common in smokers.
Smokers with coronary artery disease have a higher risk
for events than nonsmokers with similar levels of
coronary artery disease. 17 In patients with stable known
coronary artery disease, several studies have demonstrat-
ed that smokers have a higher cardiovascular event rate
after revascularization compared to ex/nonsmokers. 18,19
Smoking is associated with increased platelet reactivity
and greater therapeutic response to anti-platelet inter-
ventions. 1,2,6 The enhanced thrombotic tendency may
lead to more pronounced thrombus formation in
response to plaque rupture and a greater likelihood of
total coronary artery occlusions. Our nding that smok-
ing is more frequently associated with ST-elevation MI as
the index event and stent thrombosis as a complication
event is consistent with previous reports. 20,21
Many factors including genetic, metabolic, cellular,
and clinical factors explain a highly variable inter-patient
response to clopidogrel. 22,23 Smokers on clopidogrel
treatment have been reported to have greater platelet
inhibition than nonsmokers and it has been suggested
that this is due to induction of CYP1A2 leading to more
efcient active metabolite production. 7 Recently, this
mechanism was conrmed only in cytochrome CYP1A2
( 163CNA) allele carriers, suggesting a genotype-depen-
dent effect of smoking on clopidogrel responsiveness. 24
Cornel et al 7
Since ticagrelor is not dependent on metabolism for its
activity, smoking should not inuence its inhibitory
effects. Thus, it might be expected that habitual
smoking might reduce the treatment effect of ticagrelor
compared to clopidogrel. However, we did not observe
any signicant interaction between smoking status and
the relative efcacy of ticagrelor compared to clopido-
grel. Therefore, these results do not support that
smoking might increase the efcacy of clopidogrel.
A recent report from the TRITON-TIMI 38 cohort
likewise found no signicant interaction of smoking
status at baseline with the clinical efcacy of prasugrel
versus clopidogrel. 25
Our ndings contrast with those from a post hoc
analysis of the CLARITY-TIMI 28 trial comparing clopido-
grel to placebo in ST-elevation MI (STEMI) patients
receiving brinolysis, in which smoking was associated
with an apparent enhanced protective effect of clopido-
grel on angiographic and clinical outcomes 2 to 30 days
after MI, particularly among smokers of 10 cigarettes
per day. However, a GRACE report did not nd that early
clopidogrel use among smokers presenting with ACS was
independently associated with a greater reduction in
cardiovascular events in-hospital and at 6 months
compared to nonsmokers. 11
A diminished response to clopidogrel is associated with
an increased risk of ischemic events. 8,26 However, the
present subgroup analysis of the PLATO trial supports the
concept that a more potent and consistent antiplatelet
therapy may have the same benet both in smokers and
ex/nonsmokers equally without any interaction between
smoking status and treatment effects. The hypothetical
improvement in clopidogrel metabolism due to smoking
does not translate into reduced benet of ticagrelor over
clopidogrel among smokers. Therefore a potential
enhanced efcacy of clopidogrel in smokers appears to
be a less favorable management approach when com-
pared with the superior effects of ticagrelor both in
smokers and nonsmokers.
In this PLATO subgroup analysis, smoking was not
associated with a change in the risk for major bleeding.
Furthermore, there was no interaction between treat-
ment with the more potent P2Y12 inhibitor ticagrelor or
clopidogrel and the risk of bleeding. These ndings are in
accordance with previous studies showing that the
addition of clopidogrel to aspirin was associated with
an increase in major bleeding but without any interaction
with respect to smoking status. 6,11
Limitations
There are several limitations to this report. First, this
was a post hoc analysis. Although we prespecied
smoking status and multivariate modeling was used to
adjust for imbalances in baseline characteristics, residual
confounding may inuence the outcome of this analysis

8 Cornel et al
since one cannot correct for unknown confounders.
Second, recurrent risk analyses are prone to index event
bias. 27 Selection of patients because of the occurrence of
an index event may induce paradoxical results since risk
factors among smokers are less frequent at the time of an
ACS. Third, these analyses depend on a single baseline
recording of smoking status, with no data after enroll-
ment and no information regarding the quantity of
smoking. These omissions prevent evaluation of the
impact of changes in smoking status after hospital
discharge and the potential effects of heavy smoking
compared to nonsmoking. To examine the potential for
intervening events changing the relationship of baseline
factor with outcomes, we additionally evaluated the
association between baseline smoking status and events
within the rst 10 days of follow-up, during which time
CYP enzyme induction from smoking up to the time of
enrolment might be expected to have an inuence on the
pharmacodynamic response to clopidogrel.
Conclusions
Active smoking prior to an ACS is associated with an
increased risk for stent thrombosis. The lack of interac-
tion between smoking status and treatment allocation
suggests that the benets of more potent and consistent
platelet P2Y12 receptor blockade with ticagrelor com-
pared to clopidogrel is not affected by smoking status in a
broad spectrum of ACS patients.
Acknowledgements
The complete list of PLATO investigators and main
study committees has been published previously. We
acknowledge Ebba Bergman, PhD, and Elin Lindhagen,
PhD, Uppsala Clinical Research center, for excellent
editorial support.
Disclosures
Dr Cornel: advisory board fees from AstraZeneca, Eli
Lilly/Daiichi Sankyo; consultancy fees from Merck. Dr
James reports receiving institutional research grant and
honoraria from AstraZeneca, Eli Lilly, Merck, and Bristol-
Myers Squibb and being an advisory board member for
AstraZeneca, Eli Lilly, and Merck. Dr Becker reports
receiving research grant from Johnson & Johnson, Bayer,
and Regado Biosciences; honoraria from Johnson&John-
son, Regado Biosciences, and Daiichi-Sankyo; being a
consultant for AstraZeneca, Regado Biosciences, John-
son&Johnson, and Daiichi-Sankyo. Dr Goodman reports
receiving Research grant support and speaker/consulting
honoraria from: AstraZeneca, Bristol-Myers Squibb, Daii-
chi-Sankyo, Eisai, Lilly, Merck, Sano-Aventis. Dr Husted
reports being an advisory board member for AstraZeneca,
Bristol-Myers Squibb, Pzer, and Bayer; research support
American Heart Journal
Month Year
from GlaxoSmithKline, Pzer, and Sano-Aventis. Dr
Katus declares consulting and lecture fees from AstraZe-
neca. Dr Steg has received research grant support from
Servier; and speaking/consulting honoraria from Astellas,
Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, Daii-
chi Sankyo, Eli Lilly, GlaxoSmithKline, Eisai, Medtronic,
MSD, Pzer, Roche, Sano-Aventis, The Medicines
Company, AstraZeneca, Servier, Merck, Otsuka, and he
is a Stockholder Aterovax. KW Mahaffey has received
research grant support from Eli Lilly, Johnson&Johnson,
Portola, Sano-Aventis, The Medicines Company, Scher-
ing-Plough, Bayer, Bristol-Myers Squibb, Daiichi Sankyo;
and speaking/consulting honoraria from Bristol Myers
Squibb, Daiichi Sankyo, Johnson&Johnson, Sano-Aven-
tis, Schering-Plough, AstraZeneca, Eli Lilly. Dr Storey
reports receiving research grant from AstraZeneca,
EliLilly/Daiichi-Sankyo and Merck; research support
from Accumetrics; honoraria from AstraZeneca, Eli
Lilly/Daiichi-Sankyo, Merck, Novartis, The Medicines
Company, Iroko, Sano-Aventis/Bristol-Myers Squibb,
GlaxoSmithKline, Accumetrics, Medscape, and Eisai;
and being a consultant for AstraZeneca, Merck, Novartis,
Accumeterics, and Eisai. Dr Vintila reports receiving
research grants from Sano-Aventis and Servier; consul-
tant fees from Pzer and Sano-Aventis; speaker fees from
Abbott, AstraZeneca, Bayer, Boehringer Ingelheim,
Menamni, Merck, Novartis, Pzer, Sano-Aventis, and
Servier; and has advisory board membership with
AstraZeneca, Novartis, Pzer, Sano-Aventis, and Servier.
Dr Santoso and Mrs Sun report having no relationship
with the industry. Dr Horrow: employee of AstraZeneca
and having equity ownership in AstraZeneca. Dr Wallen-
tin reports receiving research grants from AstraZeneca,
Merck/Schering-Plough, Boehringer-Ingelheim, Bristol-
Myers Squibb/Pzer, GlaxoSmithKline, and Schering-
Plough; being a consultant for Merck/Schering-Plough,
Regado Biosciences, Protola, CSL Behring, Athera Bio-
technologies, Boehringer-Ingelheim, AstraZeneca, Glax-
oSmithKline, and Bristol-Myers Squibb/Pzer; lecture fees
from AstraZeneca, Boehringer-Ingelheim, Bristol-Myers
Squibb/Pzer, GlaxoSmithKline, Schering-Plough. Dr
Harrington reports advisory board fees from Novartis,
Portola Pharmaceutical, and Merck; consulting fees from
AstraZeneca, Bristol-Myers Squibb, Merck, Novartis,
Portola, and Sano-Aventis; honoraria/lecture fees from
Eli Lilly, Merck, and AstraZeneca; grant support from
AstraZeneca, Bristol-Myers Squibb, Portola Pharmaceuti-
cal, Merck, Novartis and The Medicines Company; travel
support from AstraZeneca, Novartis, and Merck.
References
1. Nowak J, Murray JJ, Oates JA, et al. Biochemical evidence of
a chronic abnormality in platelet and vascular function in
healthy individuals who smoke cigarettes. Circulation 1987;76:
6-14.
American Heart Journal
Volume 0, Number 0
2. Hung J, Lam JYT, Lacoste L, et al. Cigarette smoking acutely increases
platelet thrombus formation in patients with coronary artery disease
taking aspirin. Circulation 1995;92:2432-6.
3. Harding SA, Sarma J, Josephs DH, et al. Upregulation of the
CD40/CD40 ligand dyad and platelet-monocyte aggregation in
cigarette smokers. Circulation 2004;109:1926-9.
4. Ambrose JA, Barua RS. The pathophysiology of cigarette smoking
and cardiovascular disease: an update. J Am Coll Cardiol 2004:
1731-7.
5. Willi C, Bodenmann P, Ghali WA, et al. Active smoking and the
risk of type 2 diabetes: a systematic review and meta-analysis. JAMA
2007;298:2654-64.
6. Desai NR, Mega JL, Jiang S, et al. Interaction between cigarette
smoking and clinical benefit of clopidogrel. J Am Coll Cardiol 2009;
53:1273-8.
7. Bliden KP, Dichiara J, Lawal L, et al. The association of cigarette
smoking with enhanced platelet inhibition by clopidogrel. J Am Coll
Cardiol 2008;52:531-3.
8. Matetzky S, Shenkman B, Guetta V, et al. Clopidogrel resistance is
associated with increased risk of recurrent atherothrombotic events
in patients with acute myocardial infarction. Circulation 2004;109:
3171-5.
9. Kroon LA. Drug interactions with smoking. Am J Health Syst Pharm
2007;64:1917-21.
10. Berger JS, Bhatt DL, Steinhubl SR, et al. Smoking, clopidogrel, and
mortality in patients with established cardiovascular disease.
Circulation 2009;120:2337-44.
11. Sibbald M, Yan AT, Huang W, et al. Association between smoking,
outcomes, and early clopidogrel use in patients with acute coronary
syndrome: insights from the global registry of acute coronary events.
Am Heart J 2010;160:855-61.
12. Gurbel PA, Bliden KP, Butler K, et al. Response to ticagrelor in
clopidogrel nonresponders and responders and effect of switching
therapies: The RESPOND study. Circulation 2010;121:1188-99.
13. Storey RF, Angiolillo D, Patil S, et al. Inhibitory Effects of Ticagrelor
Compared to Clopidogrel on Platelet Function in Patients with Acute
Coronary Syndromes: the PLATO PLATELET Substudy. J Am Coll
Cardiol 2010;56:1456-62.
14. Wallentin L, Becker RC, Budaj A, et al. Ticagrelor versus clopidogrel
in patients with acute coronary syndromes. N Engl J Med 2009;361:
1045-57.
15. James S, Akerblom A, Cannon CP, et al. Comparison of ticagrelor,
the first reversible oral P2Y12 receptor antagonist, with clopidogrel
in patients with acute coronary syndromes: Rationale, design, and
Cornel et al 9
baseline characteristics of the PLATelet inhibition and patient
Outcomes (PLATO) trial. Am Heart J 2009;157:599-605.
16. Barbash GI, Reiner J, White HD, et al. Evaluation of paradoxic
beneficial effects of smoking in patients receiving thrombolytic
therapy for acute myocardial infarction: mechanism of the smoker's
paradox from the GUSTO-I trial, with angiographic insights. Global
Utilization of Streptokinase and Tissue-Plasminogen Activator for
Occluded Coronary Arteries. J Am Coll Cardiol 1995;26:1222-9.
17. van Werkhoven JM, Schuijf JD, Pazhenkottil AP, et al. Influence of
smoking on the prognostic value of cardiovascular computed
tomography coronary angiography. Eur Heart J 2011;32:365-70.
18. Hasdai D, Garrat KN, Grill DE, et al. Effect of smoking status on the
long-term outcome after successful percutaneous coronary revascu-
larization. N Engl J Med 1997;336:755-61.
19. van Domburg RT, op Reimer WS, Hoeks SE, et al. Three life-years
gained from smoking cessation after coronary artery bypass surgery:
a 30-year follow-up study. Am Heart J 2008;156:473-6.
20. Baran KW, Lasala JM, Cox DA, et al. A clinical risk score for the
prediction of very late stent thrombosis in drug eluting stent patients.
EuroIntervention 2011;6:949-54.
21. Goto K, Nikolsky E, Lansky AL, et al. Impact of smoking on outcomes
of patients with ST-segment elevation myocardial infarction (from the
HORIZONS-AMI trial). Am J Cardiol 2011;108:1387-94.
22. Gurbel PA, Bliden KP, Butler K, et al. Randomized double-blind
assessment of the onset and offset of the antiplatelet effects of
ticagrelor versus clopidogrel in patients with stable coronary artery
disease: the ONSET/OFFSET study. Circulation 2009;120:
2577-85.
23. 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:
1505-16.
24. Park KW, Park JJ, Jeon KH, et al. Enhanced clopidogrel
responsiveness in smokers: smokers' paradox is dependent on
cytochrome P450 CYP1A2 status. Arterioscler Thromb Vasc Biol
2011;31:665-71.
25. Hochholzer W, Trenk D, Mega JL, et al. Impact of smoking on
antiplatelet effect of clopidogrel and prasugrel after loading dose and
on maintenance therapy. Am Heart J 2011;162:518-26.
26. Gurbel PA, Bliden KP, Samara W, et al. Clopidogrel effect on platelet
reactivity in patients with stent thrombosis: results of the CREST study.
J Am Coll Cardiol 2005;46:1820-6.
27. Dahabreh IJ, Kent DM. Index event bias as an explanation for the
paradoxes of recurrence risk research. JAMA 2011;305:822-3.
American Heart Journal
Volume 0, Number 0
Cornel et al 9.e1

Appendix

Figure 3

A 4

3.5

3
CV Death, MI or Stroke

2.5

1.5

0.5

0
0 10
Days after randomization
Non-smoker w/ Clopidogrel 4670 4469
Non-smoker w/ Ticagrelor 4641 4450
Current-smoker w/ Clopidogrel 2589 2493
Current-smoker w/ Ticagrelor 2607 2517

B 1.4

1.2
Definite stent thrombosis

0.8

0.6

0.4

0.2

0
0 10
Days after randomization
Non-smoker w/ Clopidogrel 3329 3252
Non-smoker w/ Ticagrelor 3244 3171
Current-smoker w/ Clopidogrel 2299 2251
Current-smoker w/ Ticagrelor 2382 2332

Adjusted survival plots for the first 10 days. Adjusted cumulative incidence of: A, the primary composite of vascular death, myocardial infarction
and stroke; and B, definite stent thrombosis and in the Clopidogrel (dotted lines) and Ticagrelor (solid lines) groups stratified for smoking status at
the time of randomization. Ticag, tigacrelor; Clop, clopidogrel.