Calcium-channel blockers do not alter the clinical efficacy of clopidogrel after myocardial infarction: a nationwide cohort study.

Abstract

The purpose of this study was to determine the risk of adverse cardiovascular events associated with concomitant use of clopidogrel and calcium-channel blockers (CCBs) in patients with myocardial infarction (MI).
CCBs inhibit a variety of cytochrome P-450 enzymes, some of which contribute to clopidogrel metabolic activation. This interaction may diminish the efficacy of clopidogrel.
All patients surviving 30 days after a first-time MI in the period 2000 to 2006 in Denmark were identified by individual-level linkage of nationwide administrative registers. The cohort was divided into patients treated with and without clopidogrel and followed for 1 year after discharge. The risk of a composite of cardiovascular death, MI, or stroke and the risk of the individual components of the composite end point and all-cause death associated with CCBs were analyzed with multivariable Cox proportional hazard models and in univariate propensity score-matched models.
A total of 56,800 patients were included, of whom 24,923 were treated with clopidogrel and 13,380 with CCBs. In the Cox analyses, the risk of the composite end point associated with CCBs was increased in both patients treated and not treated with clopidogrel, with a hazard ratio of 1.15 (95% confidence interval [CI]: 1.07 to 1.24) and 1.05 (95% CI: 1.01 to 1.11), respectively. The increased risk was independent of clopidogrel use; the hazard rate ratio was 1.08 (95% CI: 0.99 to 1.18). Analyses of all additional adverse end points and propensity score-matched models provided similar results.
The clinical efficacy of clopidogrel in patients with a recent MI is not modified by concomitant CCB treatment. This potential drug interaction is unlikely to have clinical significance.

Full text

Page 1

Cardiac Pharmacology
Calcium-Channel Blockers Do Not Alter the Clinical
Efficacy of Clopidogrel After Myocardial Infarction
A Nationwide Cohort Study
Jonas B. Olesen, MB, Gunnar H. Gislason, MD, PHD, Mette G. Charlot, MD,
Emil L. Fosbøl, MD, PHD, Charlotte Andersson, MB, Peter Weeke, MD, Ole Ahlehoff, MD,
Christian Selmer, MD, Christian Torp-Pedersen, MD, DMSC, Peter R. Hansen, MD, PHD, DMSC
Hellerup, Denmark
Objectives
The purpose of this study was to determine the risk of adverse cardiovascular events associated with concomi-
tant use of clopidogrel and calcium-channel blockers (CCBs) in patients with myocardial infarction (MI).
Background
CCBs inhibit a variety of cytochrome P-450 enzymes, some of which contribute to clopidogrel metabolic activa-
tion. This interaction may diminish the efficacy of clopidogrel.
Methods
All patients surviving 30 days after a first-time MI in the period 2000 to 2006 in Denmark were identified by
individual-level linkage of nationwide administrative registers. The cohort was divided into patients treated with
and without clopidogrel and followed for 1 year after discharge. The risk of a composite of cardiovascular death,
MI, or stroke and the risk of the individual components of the composite end point and all-cause death associ-
ated with CCBs were analyzed with multivariable Cox proportional hazard models and in univariate propensity
score-matched models.
Results
A total of 56,800 patients were included, of whom 24,923 were treated with clopidogrel and 13,380 with CCBs.
In the Cox analyses, the risk of the composite end point associated with CCBs was increased in both patients
treated and not treated with clopidogrel, with a hazard ratio of 1.15 (95% confidence interval [CI]: 1.07 to 1.24)
and 1.05 (95% CI: 1.01 to 1.11), respectively. The increased risk was independent of clopidogrel use; the hazard
rate ratio was 1.08 (95% CI: 0.99 to 1.18). Analyses of all additional adverse end points and propensity score–
matched models provided similar results.
Conclusions
The clinical efficacy of clopidogrel in patients with a recent MI is not modified by concomitant CCB treatment.
This potential drug interaction is unlikely to have clinical significance.
© 2011 by the American College of Cardiology Foundation
(J Am Coll Cardiol 2011;57:409–17)
Clopidogrel is a thienopyridine that by inhibiting the
platelet P2Y12adenosine diphosphate receptor reduces the
risk of new atherothrombotic events in patients after myo-
cardial infarction (MI) and after percutaneous coronary
intervention (1,2). Clopidogrel is a prodrug that is metab-
olized to its active thiol metabolite by a variety of hepatic
cytochrome P-450 (CYP) enzymes, including the 3A4 and
2C19 isoforms (3–6). Calcium-channel blockers (CCBs),
which are widely used for the treatment of hypertension and
angina pectoris, are metabolized by the same CYP3A4
enzymes as clopidogrel (7–9). Therefore, it has been pro-
posed that CCBs can interfere with the metabolism of
clopidogrel to its active form and thereby reduce the clinical
efficacy of clopidogrel. Recent studies have demonstrated a
diminished effect of clopidogrel by reduced ex vivo platelet
inhibition in patients treated concomitantly with CCBs
(10–12). However, although patients with MI are often
treated with both a CCB and clopidogrel, the clinical
importance of this hypothetical drug interaction has never
been examined on a population level. We therefore per-
formed a nationwide study including all patients discharged
after a first-time MI in Denmark. By using nationwide
registries of hospitalizations and drug dispensing from
pharmacies, we analyzed the effect of CCBs on clinical end
points in patients treated and not treated with clopidogrel.
Methods
Registry data sources. In Denmark, a unique and perma-
nent civil registration number for every citizen enables
From the Department of Cardiology, Copenhagen University Hospital Gentofte,
Hellerup, Denmark. The authors have reported that they have no relationships to
disclose.
Manuscript received March 28, 2010; revised manuscript received August 12, 2010,
accepted August 18, 2010.
Journal of the American College of Cardiology
© 2011 by the American College of Cardiology Foundation
Published by Elsevier Inc.
Vol. 57, No. 4, 2011
ISSN 0735-1097/$36.00
doi:10.1016/j.jacc.2010.08.640

Page 2

individual-level linkage of infor-
mation from nationwide admin-
istrative registries. Since 1978, all
admissions to Danish hospitals
are registered in the Danish Na-
tional Patient Registry with 1
primary and, if appropriate, ?1
secondary discharge diagnoses
according to the International
Classification of Diseases (ICD)
(13). Since 1995, all prescrip-
tions dispensed from Danish
pharmacies are accurately registered in the Danish Registry
of Medicinal Product Statistics (prescription registry) ac-
cording to the international Anatomical Therapeutic
Chemical (ATC) classification system (14). The civil reg-
istration system holds information on vital statistics for all
citizens and the National Causes of Death Registry holds
information on primary and contributing causes of death.
Study population and outcomes. From the National Pa-
tient Registry, we identified all patients hospitalized with a
first-time MI in the period 2000 to 2006 in Denmark (Fig. 1).
Treatment with clopidogrel (ATC: B01AC04) was defined
by claimed prescriptions within 30 days of discharge. The
use of clopidogrel has been found to be stable over time in
this cohort with 1-year continuation of use in 89% of the
patients after 2004 (15). To avoid immortal time bias (i.e.,
avoid predicting on the future), we only included patients
surviving 30 days after discharge. Exposure to CCBs (ATC:
C08) was also defined according to prescription claims, but
because the continuation of CCB treatment has never been
validated in this population, exposure to CCBs was defined
as a time-dependent covariate in the follow-up period.
CCBs were divided into their main pharmacological sub-
classes (i.e., verapamil [ATC: C08DA01], diltiazem [ATC:
C08DB01], and dihydropyridines [ATC: C08CA01]) and
into high- and low-dose administration with the following
cutoff values for high daily doses: verapamil ?240 mg,
diltiazem ?240 mg, amlodipine ?5 mg, felodipine ?5 mg,
isradipine ?2.5 mg, lacidipine ?2 mg, lercanidipine ?10
mg, nifedipine ?40 mg, nimodipine ?2 mg, and nitren-
dipine ?10 mg. Patients emigrating during the study period
were censored at the time of emigration. The primary study
outcome was a combined end point of cardiovascular death
(ICD-10: I00 to I99), rehospitalization for nonfatal MI
(ICD-10: I21 to I22), or nonfatal stroke (ICD-10: I63 to
I66). Secondary outcomes were the individual components
of the combined end point and all-cause death. The MI and
stroke diagnoses were previously validated in the National
Patient Registry (16,17). The population was followed for 1
year after discharge.
Comorbidity, socioeconomic status, and concomitant
medication. Comorbidity was defined by diagnoses at dis-
charge after the index MI as specified in the Ontario acute
MI mortality prediction rule (18). The comorbidity index
was further enhanced by adding diagnoses from the year
before the index event, as done by Rasmussen et al. (19).
We used treatment with loop diuretics (ATC: C03C) and
glucose-lowering drugs (ATC: A10) up to 90 days after
discharge as proxies for heart failure and diabetes mellitus,
respectively, as done previously (20). Socioeconomic status
Abbreviations
and Acronyms
ATC ? Anatomical
Therapeutic Chemical
CCB ? calcium-channel
blocker
CYP ? cytochrome P-450
ICD ? International
Classification of Diseases
MI ? myocardial infarction
Figure 1Selection of Study Population
CCB ? calcium-channel blocker.
410Olesen et al.
Calcium-Channel Blockers and Clopidogrel
JACC Vol. 57, No. 4, 2011
January 25, 2011:409–17

Page 3

was defined by the individual average annual gross income
during the 5-year period before inclusion, and patients were
divided in quintiles according to their income. From the
prescription registry, we also obtained information on
claimed prescriptions up to 90 days after discharge for the
other concomitant medications listed in Table 1.
Statistical analyses. For the baseline characteristics, com-
parisons of categorical and continuous covariates were
Baseline Characteristics of Patients With First-Time Myocardial Infarction Surviving 30 Days From Discharge
Table 1Baseline Characteristics of Patients With First-Time Myocardial Infarction Surviving 30 Days From Discharge
Treatment With Clopidogrel
(n ? 24,923)
No Treatment With Clopidogrel
(n ? 31,877)
Without CCB
(n ? 19,201)
With CCB
(n ? 5,722)
p Value for
Difference
Without CCB
(n ? 24,219)
With CCB
(n ? 7,658)
p Value for
Difference
Overall age, yrs 64.6 ? 12.766.5 ? 12.2
?0.001 71.2 ? 13.4 71.0 ? 12.50.22
Men 13,494 (70.3)3,630 (63.4)
?0.001 14,339 (59.2)4,300 (56.2)
?0.001
Age, yrs 62.8 ? 12.264.8 ? 11.9
?0.001 68.2 ? 13.268.5 ? 12.3 0.30
Women 5,707 (29.7)2,092 (36.6)
?0.0019,880 (40.8) 3,358 (43.9)
?0.001
Age, yrs69.1 ? 12.769.3 ? 12.2 0.46 75.5 ? 12.574.2 ? 11.9
?0.001
Year of inclusion
?0.001
?0.001
2000–2002 4,074 (21.2) 1,446 (25.3)14,758 (60.9)4,885 (63.8)
2003–2004 7,463 (38.9)2,135 (37.3) 5,838 (24.1) 1,732 (22.6)
2005–20067,664 (39.9)2,141 (37.4) 3,623 (15.0)1,041 (13.6)
Income group
?0.001
?0.001
0 (lowest income quintile) 1,764 (9.2)724 (12.7)4,013 (16.6)1,465 (19.1)
1 2,423 (12.6) 914 (16.0)5,571 (23.0)1,782 (23.3)
2 3,913 (20.4)1,235 (21.6) 5,203 (21.5) 1,584 (20.7)
3 4,438 (23.1)1,256 (22.0)4,873 (20.1)1,510 (19.7)
4 (highest income quintile)6,663 (34.7)1,593 (27.8) 4,559 (18.8)1,317 (17.2)
PCI13,357 (69.6)3,329 (58.2)
?0.001 2,276 (9.4)631 (8.2) 0.002
Comorbidity
Shock131 (0.7) 28 (0.5) 0.11364 (1.5)88 (1.2)0.02
DM with complications 746 (3.9)325 (5.7)
?0.0011,341 (5.5) 505 (6.6)
?0.001
Peptic ulcer168 (0.9)68 (1.2)0.03548 (2.3)162 (2.1)0.45
Pulmonary edema 128 (0.7)33 (0.6) 0.46 361 (1.5)92 (1.2)0.06
Peripheral vascular disease193 (1.0)100 (1.8)
?0.001469 (1.9) 215 (2.8)
?0.001
Cerebrovascular disease 584 (3.0) 232 (4.1)
?0.0011,568 (6.5)467 (6.1)0.24
Cancer63 (0.3) 16 (0.3)0.57104 (0.4)25 (0.3)0.22
Cardiac dysrhythmias1,382 (7.2) 455 (8.0)0.06 3,301 (13.6)1,115 (14.6)0.04
Acute renal failure 77 (0.4) 41 (0.7)0.002285 (1.2) 127 (1.7) 0.001
Chronic renal failure 113 (0.6)115 (2.0)
?0.001374 (1.5) 215 (2.8)
?0.001
Concomitant medication
Loop diuretic 5,268 (27.4) 1,775 (31.0)
?0.00111,796 (48.7) 3,586 (46.8)0.004
Spironolactone1,557 (8.1) 405 (7.1) 0.013,217 (13.3)803 (10.5)
?0.001
Aspirin 13,960 (72.7)3,699 (64.7)
?0.00111,643 (48.1)3,615 (47.2)0.18
Vitamin K antagonist992 (5.2)275 (4.8)0.28 2,251 (9.3)774 (10.1)0.03
Statin16,905 (88.0)4,973 (86.9)0.0211,418 (47.1)4,120 (53.8)
?0.001
Beta-blocker17,147 (89.3)4,441 (77.6)
?0.00117,336 (71.6)4,735 (61.8)
?0.001
ACE inhibitor9,889 (51.5)3,066 (53.6)0.00610,349 (42.7)3,228 (42.2)0.37
Glucose-lowering drug2,034 (10.6) 844 (14.8)
?0.0013,051 (12.6)1,141 (14.9)
?0.001
Verapamil 572 (10.0)1,308 (17.1)
Diltiazem526 (9.2)1,062 (13.9)
Dihydropyridines4,624 (80.8)5,288 (69.0)
Amlodipine3,990 (69.7)4,522 (59.1)
Felodipine312 (5.5)344 (4.5)
Isradipine19 (0.3)35 (0.5)
Lacidipine13 (0.2)26 (0.3)
Lercanidipine46 (0.8)35 (0.5)
Nifedipine237 (4.1)311 (4.1)
Nimodipine0 (0.0)1 (0.0)
Nitrendipine 7 (0.1)14 (0.2)
Values are mean ? SD or n (%).
ACE ? angiotensin-converting enzyme; CCB ? calcium-channel blocker; DM ? diabetes mellitus; PCI ? percutaneous coronary intervention.
411
JACC Vol. 57, No. 4, 2011
January 25, 2011:409–17
Olesen et al.
Calcium-Channel Blockers and Clopidogrel

Page 4

performed by the chi-square test and Student t test, respec-
tively. For all analyses, a 2-sided p value ?0.05 was
considered statistically significant. To estimate the risk
associated with CCBs with or without concomitant treat-
ment with clopidogrel, we used 2 different methods. First,
Cox proportional hazard models were constructed for the
entire population. The models were adjusted for age, sex,
percutaneous coronary intervention, income, concomitant
medication, and comorbidity. Exposure to CCBs was in-
cluded in the models as a time-dependent covariate (i.e.,
patients were only considered at risk when they were taking
CCBs). For sensitivity analyses, we performed separate
analyses for the main pharmacological subclasses of CCBs,
high- and low-dose CCB exposure, and selected patient
subgroups. Additionally, we performed propensity score–
matched analyses. A propensity score for the likelihood of
receiving CCBs within the first year from discharge was
quantified by a multivariable logistic regression analysis
conditional on the baseline covariates. Using the Greedy
matching macro (21), each CCB-exposed patient was
matched by propensity score to 1 nonexposed control
subject from the study population. The risk associated with
CCB use in the propensity score-matched groups was
subsequently estimated in univariate Cox proportional haz-
ard models. In all Cox models, the model assumptions (i.e.,
proportional hazards, linearity of continuous covariates, and
lack of interactions) were tested systematically and found to
be valid unless otherwise indicated. All analyses were
performed with SAS version 9.1 (SAS Institute Inc., Cary,
North Carolina).
Ethics. No ethical approval is required for retrospective
registry studies in Denmark. The study was approved by
The Danish Data Protection Agency (no. 2003-54-1269).
Results
The Danish population age 10 years or older on January 1,
1997, comprised 4,614,807 individuals. We identified
71,987 patients hospitalized with MI during this 7-year
study period, of which 56,800 patients had a first-time MI
and met the inclusion criterion of 30 days survival from
discharge (Fig. 1). Clopidogrel treatment was initiated in
24,923 patients (43.9%) within 30 days after discharge, and
the use of CCBs during the 1-year follow-up period was
identified in 5,722 patients (23.0%) receiving clopidogrel
and 7,658 patients (24.0%) not receiving clopidogrel. Of
patients treated with a CCB during follow-up, 47.4%,
67.2%, 66.5%, and 70.1% were receiving CCB treatment at
days 30, 90, 180, and 360 after discharge, respectively. The
baseline characteristics of the study population are shown in
Table 1. Men composed 63.0% of the population, and their
mean age was 68.5 ? 13.3 years. Overall, patients were
more likely to begin clopidogrel treatment at the end of the
study period, and patients treated with clopidogrel were
younger, more often men, had less comorbidity, and had
more often undergone percutaneous coronary intervention.
Of patients treated with CCBs in the 1-year follow-up
period, 74.1% were treated with a dihydropyridine (pre-
dominantly amlodipine). Patients treated with a CCB more
often had diabetes mellitus and previous hospital admissions
for renal failure (Table 1).
Time-dependent Cox proportional hazard analyses. As
shown in Table 2, the Cox proportional hazard analyses
demonstrated an increased risk of the composite end point
of cardiovascular death, rehospitalization for MI, or stroke
associated with CCBs in patients treated with clopidogrel
(hazard ratio: 1.15; 95% confidence interval [CI]: 1.01 to
1.10) and in those not treated with clopidogrel (hazard
ratio: 1.05; 95% CI: 1.01 to 1.11). The hazard rate ratio for
interaction between CCBs and clopidogrel was 1.08 (95%
CI: 0.99 to 1.18; p ? 0.08), indicating no significant effect
modification of CCBs and clopidogrel treatment on the
primary study outcome. The results from the analyses of the
secondary outcomes (i.e., cardiovascular death, rehospital-
ization for MI, stroke, and all-cause death) gave similar
results (Table 2). For the composite end point, we found no
interaction between any of the CCB subclasses and clopi-
dogrel (Fig. 2). Additionally, we found no interaction of the
risk of the composite end point between clopidogrel and
either high- or low-dose CCB treatment (p value for
interaction was 0.60 and 0.64, respectively). The analyses in
selected patient subgroups yielded similar results (Fig. 3).
Risk Associated With CCB Treatment
Table 2Risk Associated With CCB Treatment
Treatment With Clopidogrel
(n ? 24,923)
No Treatment With Clopidogrel
(n ? 31,877)
Likelihood for Difference With
CCB Treatment
Without CCB
(n ? 19,201)
HR
With CCB
(n ? 5,722)
HR (95% CI)p Value
Without CCB
(n ? 24,219)
HR
With CCB
(n ? 7,658)
HR (95% CI) p ValueHRR (95% CI) p Value*
MI, stroke, or CV death1.00 1.15 (1.07–1.24)
?0.0011.001.05 (1.01–1.10)0.031.08 (0.99–1.18)0.08
MI1.001.35 (1.20–1.52)
?0.0011.00 1.27 (1.16–1.40)
?0.001 1.05 (0.90–1.22) 0.51
Stroke 1.001.14 (0.95–1.36)0.15 1.001.20 (1.07–1.34) 0.0020.98 (0.80–1.20)0.83
CV death1.00 1.05 (0.89–1.24)0.561.00 1.00 (0.92–1.09)0.96 1.11 (0.93–1.34)0.25
All-cause death1.00 0.97 (0.83–1.13)0.671.000.97 (0.90–1.04) 0.361.05 (0.89–1.25) 0.54
*p value for interaction between CCBs and clopidogrel.
CCB ? calcium-channel blocker; CI ? confidence interval; CV ? cardiovascular; HR ? hazard ratio; HRR ? hazard rate ratio; MI ? myocardial infarction.
412 Olesen et al.
Calcium-Channel Blockers and Clopidogrel
JACC Vol. 57, No. 4, 2011
January 25, 2011:409–17

Page 5

Propensity score–matched time-dependent Cox proportional-
hazard analyses. Using logistic regression models condi-
tional on baseline covariates (Table 1), we estimated the
propensity for receiving CBBs for each patient. Logistical
regression was performed in the total study population and
in 2 separate models including patients treated and not
treated with clopidogrel. In the overall regression model,
c statistics were 0.61 both with and without clopidogrel (i.e.,
clopidogrel as an explanatory covariate did not improve the
discriminatory power of the model). In the models of
patients treated and not treated with clopidogrel, the c sta-
tistics were 0.64 and 0.60, respectively, indicating acceptable
discriminatory power of the models. Additional results of
the logistic regression used for the propensity score-
matching are given in the Online Appendix. In patients
receiving clopidogrel, the median propensity score for ini-
tiating CCB treatment was 23.5% (interquartile range: 18.6
to 31.5) for both patients treated and those not treated with
CCBs, respectively. In patients not receiving clopidogrel,
the median propensity score was 25.0% (interquartile range:
20.4 to 30.3) for both the CCB and non-CCB treated,
respectively. By individual propensity scores, 5,692 patients
receiving both clopidogrel and CCB treatment were
matched 1:1 with patients treated with clopidogrel and no
CCB. In the same manner, 7,658 patients treated with
CCBs and no clopidogrel were matched with patients not
treated with either CCBs or clopidogrel. Baseline charac-
teristics of the propensity score–matched populations and
p values for differences between groups are displayed in
Table 3. Results of the univariate Cox analyses of the
Figure 2Risk of Cardiovascular Death, Myocardial Infarction, or Stroke According to CCB Subgroups and Clopidogrel Treatment
Results from adjusted time-dependent Cox proportional hazard analyses. Error bars illustrate 95% confidence intervals. p values for the difference between patients
treated and not treated with clopidogrel were 0.08, 0.65, 0.37, and 0.08 for all CCBs, verapamil, diltiazem, and dihydropyridines, respectively. CCB ? calcium-channel
blocker; HR ? hazard ratio.
Figure 3Risk of Cardiovascular Death, Myocardial Infarction, or Stroke Associated With CCBs in Selected Subpopulations
Results from adjusted time-dependent Cox proportional-hazard analyses. Error bars illustrate 95% confidence intervals. In both groups (clopidogrel or no clopidogrel), the
p value for the difference between corresponding paired subpopulations (i.e., subjects 70 years of age and older vs. those younger than 70 years of age, men vs.
women, and so on) was not significant. PCI ? percutaneous coronary intervention; other abbreviations as in Figure 2.
413
JACC Vol. 57, No. 4, 2011
January 25, 2011:409–17
Olesen et al.
Calcium-Channel Blockers and Clopidogrel

Page 6

propensity score–matched groups were concordant with the
findings of the nonmatched primary analyses (Table 4).
Adjusting the propensity score–matched Cox analyses for
covariates with minor differences between the CCB and
non-CBB treated (i.e., year of inclusion, vitamin K antag-
onist, and angiotensin-converting enzyme inhibitors) did
not alter the results.
Discussion
In this nationwide cohort of patients discharged after a
first-time MI, treatment with a CBB was not associated with
significant changes in clopidogrel efficacy across a range of
adverse cardiovascular end points and all-cause mortality in
either multivariable or propensity score–matched time-
dependent Cox proportional-hazard analyses. The results were
consistent for all the main CCB pharmacological subclasses, in
both high- and low-dose CCB treatment and in different
patient subpopulations. We found that the use of CCBs was
associated with increased cardiovascular risk after MI indepen-
dent of concomitant treatment with clopidogrel. However,
amlodipine was the most commonly used CCB, and results
with the other CCBs were conducted with less power. The
results suggest the absence of a clinically significant interaction
between CCBs and clopidogrel and that the increased risk
associated with CCBs may be explained by unmeasured
confounders. Along this line, we found that CCBs were more
frequently administered to patients with diabetes mellitus or
renal failure.
Baseline Characteristics for the Propensity Score–Matched Population With a First-Time Myocardial Infarction
Table 3Baseline Characteristics for the Propensity Score–Matched Population With a First-Time Myocardial Infarction
Treatment With Clopidogrel
(n ? 11,384)
No Treatment With Clopidogrel
(n ? 15,316)
Without CCB
(n ? 5,692)
With CCB
(n ? 5,692)
p Value for
Difference
Without CCB
(n ? 7,658)
With CCB
(n ? 7,658)
p Value for
Difference
Overall age, yrs 66.2 ? 12.7 66.4 ? 12.20.36 71.1 ? 13.0 71.0 ? 12.50.44
Men3,594 (63.1) 3,622 (63.6)0.594,273 (55.8) 4,300 (56.2)0.66
Age, yrs64.3 ? 12.2 64.8 ? 11.90.05 68.4 ? 12.968.5 ? 12.3 0.81
Women2,098 (36.9)2,070 (36.4) 0.593,385 (44.2) 3,358 (43.9)0.66
Age, yrs70.0 ? 12.769.3 ? 12.20.4174.6 ? 12.374.2 ? 11.90.18
Year of inclusion0.150.02
2000–20021,354 (23.8)1,433 (25.1)5,010 (65.4)4,885 (63.8)
2003–20042,203 (38.7) 2,122 (37.3)1,712 (22.4)1,732 (22.6)
2005–2006 2,135 (37.5) 2,137 (37.5)936 (12.2) 1,041 (13.6)
Income group0.73 0.46
0 (lowest income quintile)682 (12.0)719 (12.6)1,489 (19.4) 1,465 (19.1)
1893 (15.7)903 (15.9)1,849 (24.1) 1,782 (23.3)
2 1,264 (22.2) 1,228 (21.6)1,501 (19.6)1,584 (20.7)
3 1,307 (23.0)1,251 (22.0)1,505 (19.7) 1,510 (19.7)
4 (highest income quintile)1,546 (27.2)1,591 (28.0) 1,314 (17.2)1,317 (17.2)
PCI3,282 (57.7)3,326 (58.4)0.40 597 (7.8) 631 (8.2)0.31
Comorbidity
Shock 32 (0.6)28 (0.5) 0.6081 (1.1) 88 (1.2) 0.59
DM with complications 325 (5.7)321 (5.6) 0.87482 (6.3)505 (6.6)0.45
Peptic ulcer 63 (1.1)68 (1.2)0.66 141 (1.8)162 (2.1)0.22
Pulmonary edema 26 (0.5)33 (0.6)0.36 88 (1.2)92 (1.2) 0.76
Peripheral vascular disease93 (1.6)93 (1.6)1.00219 (2.9)215 (2.8) 0.85
Cerebrovascular disease208 (3.7) 230 (4.0)0.28 460 (6.0) 467 (6.1) 0.81
Cancer 20 (0.4) 16 (0.3)0.5023 (0.3) 25 (0.3)0.77
Cardiac dysrhythmias406 (7.1) 449 (7.9) 0.13 1,069 (14.0)1,115 (14.6)0.29
Acute renal failure 36 (0.6) 39 (0.7)0.73114 (1.5)127 (1.7)0.40
Chronic renal failure81 (1.4)103 (1.8)0.10216 (2.8)215 (2.8)0.96
Concomitant medication
Loop diuretic1,732 (30.4)1,758 (30.9)0.603,527 (46.1)3,586 (46.8)0.34
Spironolactone397 (7.0)404 (7.1)0.80 762 (10.0)803 (10.5)0.27
Aspirin3,744 (65.8)3,695 (64.9)0.333,590 (46.9)3,615 (47.2)0.69
Vitamin K antagonist261 (4.6)275 (4.8)0.54676 (8.8) 774 (10.1)0.007
Statin4,980 (87.5) 4,948 (86.9)0.37 4,072 (53.2)4,120 (53.8) 0.44
Beta-blocker4,435 (77.9)4,435 (77.9)1.004,791 (62.6)4,735 (61.8)0.35
ACE inhibitor3,175 (55.8)3,050 (53.6)0.023,254 (42.5)3,228 (42.2)0.67
Glucose-lowering drug 846 (14.9)835 (14.7)0.771,128 (14.7)1,141 (14.9)0.77
Values are mean ? SD or n (%). “With CCB” refers to patients treated with a CCB at some point in the 1-year follow-up period.
Abbreviations as in Table 1.
414Olesen et al.
Calcium-Channel Blockers and Clopidogrel
JACC Vol. 57, No. 4, 2011
January 25, 2011:409–17

Page 7

No interaction was found between CCBs and the year of
patient inclusion and the risk of the composite end point.
Nevertheless, in patients included after 2004, a nonsignifi-
cant trend for a higher risk with CCBs in patients treated
with clopidogrel was observed (Fig. 3). Interpretation of this
result is difficult because after 2004, patients treated or not
treated with clopidogrel probably displayed more confound-
ing differences in baseline covariates than patients included
in the earlier study period because after 2004, guideline-
based recommendations for clopidogrel in patients with MI
had been more firmly adopted. Moreover, the trend for
increased risk with CCBs in patients receiving clopidogrel
after 2004 was not reproduced in the propensity score–
matched analysis.
Hepatic metabolism is necessary to generate the active
clopidogrel metabolite through a 2-step oxidative process
driven by the CYP system. The first step that leads to
2-oxo-clopidogrel is catalyzed by 3 enzymes (CYP1A2,
CYP2B6, and CYP2C19), and the second step that gener-
ates the pharmacologically active metabolite involves 4
enzymes (CYP3A4, CYP2B6, CYP2C9, and CYP2C19)
(22). CYP2C19 is active in both oxidative steps, but a recent
study found that a polymorphism of CYP2C19 (CYP2C19*2)
had only little influence on the platelet-inhibiting effect of
clopidogrel(23).AllCBBsaremetabolizedbyCYP3A4(7–9),
and the importance of CYP3A4 for clopidogrel bioactivation
has been amply demonstrated by significant variations in
clopidogrel responsiveness as measured by ex vivo platelet
activation in subjects with CYP3A4 polymorphisms (3,4).
Furthermore, other CYP enzymes that contribute to clopi-
dogrel bioactivation participate in CBB degradation (e.g.,
CYP1A2 and CYP2C9 for verapamil) (24). In potential
agreement with the shared metabolic mechanisms between
clopidogrel and CCBs, recent observational studies found
reduced clopidogrel responsiveness as determined by increased
ex vivo residual platelet activation in patients with MI who
were treated with CCBs, and in 2 of these studies, the reduced
effect was associated with adverse short-term cardiovascular
outcomes (10–12). On the other hand, the absence of acute
effects of intravenous verapamil on the platelet inhibition
achieved by a 600-mg loading dose of clopidogrel has been
reported, suggesting that a CCB-clopidogrel interaction may
not be clinically relevant (25).
Previously, concerns of inhibited clopidogrel bioactiva-
tion by drug interactions with CYP enzymes have been
entertained with statins and proton pump inhibitors (26–28).
Many statins, including simvastatin and atorvastatin, are
metabolized by CYP3A4, but early reports of reduced
antiplatelet effects of clopidogrel as measured by various ex
vivo assays during cotreatment with these statins were
subsequently countered by numerous studies showing that
such interaction, if present, was not clinically significant
(29–32). Proton pump inhibitors are metabolized by
CYP2C19, and reports of reduced platelet responsiveness to
clopidogrel during cotreatment with proton pump inhibi-
tors were first supported by observational studies in selected
populations indicating an association with adverse cardio-
vascular events (33,34). Recently, these initial findings were
questioned by post hoc analyses of large randomized clinical
trials in which proton pump inhibitors were associated with
increased risk of adverse cardiovascular events irrespective of
the use of clopidogrel (35). Our study supports a similar
conclusion for the potential CCB-clopidogrel interaction in
patients after MI (i.e., treatment with CCBs is associated
with an adverse prognosis independent of cotreatment with
clopidogrel), indicating that a specific pharmacological in-
teraction between these agents is not clinically important
and that the increased risk associated with CCBs may be
explained by confounding. In this context, it should also be
emphasized that although reduced clopidogrel responsive-
ness as determined by ex vivo platelet function tests appears
to be associated with increased risk of adverse cardiovascular
events, the power of these tests to predict such events at
present seems to be limited (36–38).
Study strengths and limitations. The main strength of
this study was the nationwide complete data sources. No-
tably, we had accurate and complete information on all
prescriptions dispensed nationwide in Denmark. Further-
more, the study population represented an unselected cohort
of post-MI patients in a contemporary clinical setting. The
main study limitation was inherent to its observational
design. We had no information on the precise indication for
Risk Associated With CCB Treatment
Table 4 Risk Associated With CCB Treatment
Treatment With ClopidogrelNo Treatment With Clopidogrel
Likelihood for Difference With
CCB Treatment
Without CCB
(n ? 5,692)
HR
With CCB
(n ? 5,692)
HR (95% CI)
Without CCB
(n ? 7,658)
HR
With CCB
(n ? 7,658)
HR (95% CI) p Value p ValueHRR (95% CI)p Value*
MI, stroke, or CV death 1.001.17 (1.08–1.28)
?0.0011.001.14 (1.09–1.21)
?0.001 1.00 (0.90–1.10)0.93
MI 1.001.30 (1.14–1.49)
?0.0011.001.21 (1.09–1.34)
?0.0011.03 (0.87–1.22)0.71
Stroke1.001.31 (1.08–1.59) 0.0071.00 1.24 (1.09–1.40)
?0.0011.06 (0.84–1.33) 0.63
CV death 1.001.14 (0.94–1.37)0.181.001.25 (1.14–1.38)
?0.001 0.93 (0.75–1.14)0.46
All-cause death1.001.09 (0.91–1.29)0.351.00 1.21 (1.11–1.32)
?0.0010.91 (0.75–1.10) 0.34
Results from propensity score–matched, time-dependent Cox proportional hazard models. *p value for interaction between CCBs and clopidogrel.
Abbreviations as in Table 2.
415
JACC Vol. 57, No. 4, 2011
January 25, 2011:409–17
Olesen et al.
Calcium-Channel Blockers and Clopidogrel

Page 8

CCB treatment (i.e., leaving room for effects of unmeasured
confounders and confounding by indication). For example,
residual confounding may have contributed to the increased
risk observed with CCBs, and patients treated or not treated
with clopidogrel might have differed with respect to unmea-
sured confounders (e.g., perceived bleeding risk), which may
have influenced the results. Amlodipine was the most used
CCB, and therefore, results with dihydropyridines had
more power than the results with the 2 other CCB sub-
classes. Furthermore, it was not possible to check whether
the patients actually used the dispensed medication, and we
were not able to account for important prognostic factors
related to cardiovascular disease such as smoking, physical
activity, left ventricular ejection fraction, and body mass
index. Moreover, clopidogrel resistance is linked to CYP
genotype polymorphisms that display ethnic variations, and
generalization of the results from our study population
composed of almost exclusively white patients to other racial
and ethnic groups should be made with caution (39,40).
Conclusions
In patients with MI, we found no evidence of a clinically
significant effect modification between CCBs and clopi-
dogrel regarding the risk of new adverse cardiovascular
events. The results of this study do not support safety
concerns with concomitant treatment with CCBs and
clopidogrel.
Reprint requests and correspondence: Dr. Jonas B. Olesen,
Department of Cardiology, Post 67, Copenhagen University
Hospital Gentofte, Niels Andersens Vej 65, 2900 Hellerup,
Denmark. E-mail: jo@heart.dk.
REFERENCES
1. Anderson JL, Adams CD, Antman EM, et al. ACC/AHA 2007
guidelines for the management of patients with unstable angina/non–
ST-elevation myocardial infarction: a report of the American College
of Cardiology/American Heart Association Task Force on Practice
Guidelines (Writing Committee to Revise the 2002 Guidelines for the
Management of Patients With Unstable Angina/Non-ST-Elevation
Myocardial Infarction). J Am Coll Cardiol 2007;50:e1–157.
2. Kushner FG, Hand M, Smith SC Jr., et al. 2009 focused updates:
ACC/AHA guidelines for the management of patients with ST-
elevation myocardial infarction (updating the 2004 guideline and 2007
focused update) and ACC/AHA/SCAI guidelines on percutaneous
coronary intervention (updating the 2005 guideline and 2007 focused
update): a report of the American College of Cardiology Foundation/
American Heart Association Task Force on Practice Guidelines. J Am
Coll Cardiol 2009;54:2205–41.
3. Angiolillo DJ, Fernandez-Ortiz A, Bernardo E, et al. Contribution of
gene sequence variations of the hepatic cytochrome P450 3A4 enzyme
to variability in individual responsiveness to clopidogrel. Arterioscler
Thromb Vasc Biol 2006;26:1895–900.
4. Mega JL, Close SL, Wiviott SD, et al. Cytochrome p-450 polymor-
phisms and response to clopidogrel. N Engl J Med 2009;360:354–62.
5. Pereillo JM, Maftouh M, Andrieu A, et al. Structure and stereochem-
istry of the active metabolite of clopidogrel. Drug Metab Dispos
2002;30:1288–95.
6. Savi P, Pereillo JM, Uzabiaga MF, et al. Identification and biological
activity of the active metabolite of clopidogrel. Thromb Haemost
2000;84:891–6.
7. Katoh M, Nakajima M, Shimada N, Yamazaki H, Yokoi T. Inhibition
of human cytochrome P450 enzymes by 1,4-dihydropyridine calcium
antagonists: prediction of in vivo drug-drug interactions. Eur J Clin
Pharmacol 2000;55:843–52.
8. Nishio S, Watanabe H, Kosuge K, Uchida S, Hayashi H, Ohashi K.
Interaction between amlodipine and simvastatin in patients with
hypercholesterolemia and hypertension. Hypertens Res 2005;28:
223–7.
9. Watanabe H, Kosuge K, Nishio S, et al. Pharmacokinetic and
pharmacodynamic interactions between simvastatin and diltiazem in
patients with hypercholesterolemia and hypertension. Life Sci 2004;
76:281–92.
10. Harmsze AM, Robijns K, van Werkum JW, et al. The use of
amlodipine, but not of P-glycoprotein inhibiting calcium channel
blockers is associated with clopidogrel poor-response. Thromb Hae-
most 2010;103:920–5.
11. Gremmel T, Steiner S, Seidinger D, Koppensteiner R, Panzer S, Kopp
CW. Calcium-channel blockers decrease clopidogrel-mediated platelet
inhibition. Heart 2010;96:186–9.
12. Siller-Matula JM, Lang I, Christ G, Jilma B. Calcium-channel
blockers reduce the antiplatelet effect of clopidogrel. J Am Coll Cardiol
2008;52:1557–63.
13. Andersen TF, Madsen M, Jorgensen J, Mellemkjoer L, Olsen JH. The
Danish National Hospital Register. A valuable source of data for
modern health sciences. Dan Med Bull 1999;46:263–8.
14. Gaist D, Sorensen HT, Hallas J. The Danish prescription registries.
Dan Med Bull 1997;44:445–8.
15. Sorensen R, Gislason GH, Fosbol EL, et al. Initiation and persistence
with clopidogrel treatment after acute myocardial infarction: a nation-
wide study. Br J Clin Pharmacol 2008;66:875–84.
16. Madsen M, Davidsen M, Rasmussen S, Abildstrom SZ, Osler M. The
validity of the diagnosis of acute myocardial infarction in routine
statistics: a comparison of mortality and hospital discharge data with
the Danish MONICA registry. J Clin Epidemiol 2003;56:124–30.
17. Krarup LH, Boysen G, Janjua H, Prescott E, Truelsen T. Validity of
stroke diagnoses in a National Register of Patients. Neuroepidemiol-
ogy 2007;28:150–4.
18. Tu JV, Austin PC, Walld R, Roos L, Agras J, McDonald KM.
Development and validation of the Ontario acute myocardial infarc-
tion mortality prediction rules. J Am Coll Cardiol 2001;37:992–7.
19. Rasmussen S, Zwisler AD, Abildstrom SZ, Madsen JK, Madsen M.
Hospital variation in mortality after first acute myocardial infarction in
Denmark from 1995 to 2002: lower short-term and 1-year mortality in
high-volume and specialized hospitals. Med Care 2005;43:970–8.
20. Gislason GH, Jacobsen S, Rasmussen JN, et al. Risk of death or
reinfarction associated with the use of selective cyclooxygenase-2
inhibitors and nonselective nonsteroidal antiinflammatory drugs after
acute myocardial infarction. Circulation 2006;113:2906–13.
21. Greedy Matching Macro. Available at: http://www2.sas.com/
proceedings/sugi26/p214-26.pdf. Accessed December 3, 2010.
22. Kazui M, Nishiya Y, Ishizuka T, et al. Identification of the human
cytochrome P450 enzymes involved in the two oxidative steps in the
bioactivation of clopidogrel to its pharmacologically active metabolite.
Drug Metab Dispos 2010;38:92–9.
23. Hochholzer W, Trenk D, Fromm MF, et al. Impact of cytochrome
P450 2C19 loss-of-function polymorphism and of major demographic
characteristics on residual platelet function after loading and mainte-
nance treatment with clopidogrel in patients undergoing elective
coronary stent placement. J Am Coll Cardiol 2010;55:2427–34.
24. Klotz U. Antiarrhythmics: elimination and dosage considerations in
hepatic impairment. Clin Pharmacokinet 2007;46:985–96.
25. Gladding P, Webster M, Zeng I, et al. The antiplatelet effect of higher
loading and maintenance dose regimens of clopidogrel: the PRINC
(Plavix Response in Coronary Intervention) trial. J Am Coll Cardiol
Intv 2008;1:612–9.
26. Clarke TA, Waskell LA. The metabolism of clopidogrel is catalyzed
by human cytochrome P450 3A and is inhibited by atorvastatin. Drug
Metab Dispos 2003;31:53–9.
27. Lau WC, Waskell LA, Watkins PB, et al. Atorvastatin reduces the
ability of clopidogrel to inhibit platelet aggregation: a new drug-drug
interaction. Circulation 2003;107:32–7.
28. Gilard M, Arnaud B, Cornily JC, et al. Influence of omeprazole on the
antiplatelet action of clopidogrel associated with aspirin: the random-
416Olesen et al.
Calcium-Channel Blockers and Clopidogrel
JACC Vol. 57, No. 4, 2011
January 25, 2011:409–17

Page 9

ized, double-blind OCLA (Omeprazole CLopidogrel Aspirin) study.
J Am Coll Cardiol 2008;51:256–60.
29. Blagojevic A, Delaney JA, Levesque LE, Dendukuri N, Boivin JF,
Brophy JM. Investigation of an interaction between statins and
clopidogrel after percutaneous coronary intervention: a cohort study.
Pharmacoepidemiol Drug Saf 2009;18:362–9.
30. Saw J, Brennan DM, Steinhubl SR, et al. Lack of evidence of a
clopidogrel-statin interaction in the CHARISMA trial. J Am Coll
Cardiol 2007;50:291–5.
31. Saw J, Steinhubl SR, Berger PB, et al. Lack of adverse clopidogrel-
atorvastatin clinical interaction from secondary analysis of a random-
ized, placebo-controlled clopidogrel trial. Circulation 2003;108:
921–4.
32. Lim MJ, Spencer FA, Gore JM, et al. Impact of combined pharma-
cologic treatment with clopidogrel and a statin on outcomes of patients
with non-ST-segment elevation acute coronary syndromes: perspec-
tives from a large multinational registry. Eur Heart J 2005;26:1063–9.
33. Ho PM, Maddox TM, Wang L, et al. Risk of adverse outcomes
associated with concomitant use of clopidogrel and proton pump
inhibitors following acute coronary syndrome. JAMA 2009;301:
937–44.
34. Juurlink DN, Gomes T, Ko DT, et al. A population-based study of the
drug interaction between proton pump inhibitors and clopidogrel.
CMAJ 2009;180:713–8.
35. O’Donoghue ML, Braunwald E, Antman EM, et al. Pharmacody-
namic effect and clinical efficacy of clopidogrel and prasugrel with or
without a proton-pump inhibitor: an analysis of two randomised trials.
Lancet 2009;374:989–97.
36. Snoep JD, Hovens MM, Eikenboom JC, van der Bom JG, Jukema
JW, Huisman MV. Clopidogrel nonresponsiveness in patients under-
going percutaneous coronary intervention with stenting: a systematic
review and meta-analysis. Am Heart J 2007;154:221–31.
37. Combescure C, Fontana P, Mallouk N, et al. Clinical implications of
clopidogrel non-response in cardiovascular patients: a systematic re-
view and meta-analysis. J Thromb Haemost 2010;8:923–33.
38. Breet NJ, van Werkum JW, Bouman HJ, et al. Comparison of platelet
function tests in predicting clinical outcome in patients undergoing
coronary stent implantation. JAMA 2010;303:754–62.
39. Simon T, Verstuyft C, Mary-Krause M, et al. Genetic Determinants
of Response to Clopidogrel and Cardiovascular Events. N Engl J Med
2009;360:363–75.
40. Varenhorst C, James S, Erlinge D, et al. Genetic variation of
CYP2C19 affects both pharmacokinetic and pharmacodynamic re-
sponses to clopidogrel but not prasugrel in aspirin-treated patients
with coronary artery disease. Eur Heart J 2009;30:1744–52.
Key Words: calcium-channel blocker y clopidogrel y effect
modification y interaction y myocardial infarction y stroke.
APPENDIX
For supplemental tables, please see online version of this article.
417
JACC Vol. 57, No. 4, 2011
January 25, 2011:409–17
Olesen et al.
Calcium-Channel Blockers and Clopidogrel