OUP user menu

High on-thienopyridine platelet reactivity in elderly coronary patients: the SENIOR-PLATELET study

Johanne Silvain, Guillaume Cayla, Jean-Sébastien Hulot, Jonathan Finzi, Mathieu Kerneis, Stephen A. O'Connor, Anne Bellemain-Appaix, Olivier Barthélémy, Farzin Beygui, Jean-Philippe Collet, Gilles Montalescot
DOI: http://dx.doi.org/10.1093/eurheartj/ehr407 1241-1249 First published online: 8 November 2011


Aims The aim of this study was to compare on-thienopyridine platelet reactivity of elderly patients (≥75 years) vs. younger patients (<75 years). Elderly patients represent a growing and challenging segment of the coronary population for whom the effect of dual antiplatelet therapy on platelet inhibition has not been specifically addressed.

Methods and results The SENIOR-PLATELET study included 1331 coronary patients chronically (>14 days) treated with aspirin and a thienopyridine (clopidogrel 75 mg, n= 1027; clopidogrel 150 mg, n= 139; or prasugrel 10 mg, n= 165). Platelet response to clopidogrel and prasugrel was assessed by the VerifyNow assay and light transmission aggregrometry (LTA). Response to treatment, rate of high platelet reactivity (HPR), and inhibition (HPI) were compared in the two age categories. On-treatment platelet reactivity with clopidogrel 75 mg, 150 mg or prasugrel 10 mg was higher in elderly patients (n= 205) than in younger patients (n= 1126) whichever the test used. The difference in P2Y12 reaction units (PRU) between the two populations was +45 in patients treated with clopidogrel 75 mg (P< 0.0001), +30 in patients treated with clopidogrel 150 mg (P= 0.17), and +20 with prasugrel 10 mg (P= 0.10). Differences in residual platelet aggregation were consistent when measured by LTA. Elderly patients treated with clopidogrel 75 mg were more likely to have HPR than younger patients (38.2 vs. 18.2%, OR: 2.58, 95% CI: 1.76–3.79; P< 0.0001) even after adjustment for potential confounders (adj OR: 1.83, 95% CI: 1.16–2.87; P= 0.009).

Conclusion Elderly patients present an impaired response to clopidogrel with a high rate of HPR. Clopidogrel 150 mg or prasugrel 10 mg blunt, but do not eliminate the difference in response observed between old and young patients.

  • Platelet
  • Inhibitors
  • Angioplasty


The elderly represent a growing segment of the coronary population treated by dual antiplatelet therapy for percutaneous coronary intervention (PCI).1,2 These patients bear a higher risk of both ischaemic events and bleeding complications than younger patients, with a subsequently higher rate of mortality.3

Recent progress in antithrombotic treatment demonstrated the efficacy of adding a P2Y12 receptor antagonist to low-dose aspirin.47 Whether this benefit is also present in the elderly remains a debated issue due to the lack of specific data in this sub-population.

The platelet response to clopidogrel in the general population is widely documented and seems to be well correlated to the risk of recurrent ischaemic event812 and to the risk of bleeding complication, although data are scarce for the latter.13,14 However, the response to thienopyridines (clopidogrel or prasugrel) in elderly is poorly documented.15

The present study was realized to provide specific data on platelet response to clopidogrel, standard dose (75 mg) or double dose (150 mg) or prasugrel 10 mg in elderly patients (≥75 years old) in comparison with a younger population (<75 years old).


Study population

The SENIOR-PLATELETS study is a cross-sectional observational, prospective study conducted in patients with documented coronary artery disease (CAD) and treated by PCI in a single high-volume PCI centre in Paris, France. Inclusion criteria were as follows: patients had to be >18 years old, they had to be chronically treated (for >14 days) with dual antiplatelet therapy including aspirin (75–320 mg/day) and a thienopyridine, being either clopidogrel (75 or 150 mg/day) or prasugrel (10 mg/day) and lastly, they had to verbally declare being observant to their treatment. The indication of higher dose of clopidogrel (150 mg) or prasugrel 10 mg was left to the discretion of the physician. However, as prasugrel is generally not recommended in elderly patients (>75 years old) unless clinically needed (high thrombotic risk), elderly patients who were treated by prasugrel were patients who presented either a ST-elevation myocardial infarction or diabetic patients with a non-ST-elevation acute coronary syndrome (ACS). These patients received prasugrel at the dose of 10 mg, the only dosage approved in France. Prasugrel was not used in patients with history of stroke as recommended.

The exclusion criteria's in this study, included: (i) being treated with other antiplatelet agents (dypiridamole, ticlopidine, and cilostazol), (ii) having received a loading dose of thienopyridine or IIb/IIIa antagonist within the last 14 days, (iii) being treated with non-steroid anti-inflammatory drugs, (iv) having a low-platelet count <100*106/L or a haematocrit <25%.

The patients were recruited during follow-up visits in the outpatient clinic at the Institut de Cardiologie—Pitié-Salpêtrière Hospital, Paris, France from January 2005 to February 2011. The Institutional Review Board of the Pitié-Salpêtrière Hospital in Paris, France approved this study and written informed consent was provided before participation. The unit 937 of the Institut National de la Santé et de la Recherche Médicale (INSERM) conducted the study, with funding from the ACTION Group and a public grant from the Société and Fédération Française de Cardiologie.

Data collection

All patients were entered into a prospective web-based registry regrouping clinical and biological data on platelet monitoring of CAD patients treated with dual antiplatelet therapy. Clinical and biological data were collected as well as the identification of drug intake to evaluate drug-to-drug interaction.

Blood samples

During a dedicated consultation, blood samples were drawn by venipuncture on fasting patients who verbally confirmed taking their medication. Blood was collected into Becton–Dickinson 3.2% citrate vacuette tubes after having discarded the first 2–4 mL of blood to avoid spontaneous platelet activation. Blood samples were processed for platelet function test within 2 h after drawing.

Light transmission aggregometry

Platelet-rich plasma (PRP) was obtained by centrifugation of citrated whole blood at 100 g for 10 min at room temperature. Platelet poor plasma was obtained by further centrifugation at 4500 g for 15 min. In vitro platelet aggregation in PRP was measured at 37°C by light transmission aggregometry (LTA) (model 490–4D, Chrono-Log Corp, Kordia, the Netherlands) and was induced by the addition of adenosine diphosphate (ADP) (Sigma-Aldrich, Saint Quentin Fallavier, France) at final concentrations of 20 µmol/L for clopidogrel testing. Maximal platelet aggregation (MPA) and residual platelet aggregation (RPA) measured 6 min after induction of aggregation by agonist were recorded for both agonists. Pre-specified criteria used to define non-evaluable samples were: lack of sufficient signal, haemolysis, and PRP platelet count <150 000/mL and an unstable baseline.

VerifyNow P2Y12 assay

For the VerifyNow P2Y12 (VN-P2Y12) assay, samples were run locally according to the package insert (Accumetrics Corporation, San Diego, CA, USA) at the same time as for LTA. Results were expressed in P2Y12 reaction units (PRU) in response to iso-thrombin receptor activating peptide (iso-TRAP) and in response to ADP-PGE1. Iso-TRAP strongly activates platelets despite of P2Y12 receptor blockage by thienopyridine or aspirin and reflects the platelet reactivity without treatment.16 The device provides an estimated inhibition (in percentage) without pre-clopidogrel sample by reporting the ratio of the results of the ADP-PGE1 and iso-TRAP channels.


Genomic DNA was extracted from peripheral blood and was genotyped for the loss of function CYP2C19 variant (*2) using TaqMan Validated SNP assays (C_25986767_70) with the 7900HT sequence Detection System (Applied Biosystems, Courtaboeuf, France). DNA for genotyping and a specific informed consent was available for 59.1% of the total population only.


Our main objective was to evaluate platelet reactivity by the VN-P2Y12 and LTA assays under different regimens of thienopyridines (clopidogrel 75 mg, 150 mg or prasugrel 10 mg) in elderly patients (≥75 years old) and to compare it to the response obtained in younger patients (<75 years old). Secondary endpoints were (i) the rate of high platelet reactivity (HPR) in response to P2Y12 receptor antagonist using established cut-off values (PRU ADP-PGE1> 235 with the VN-P2Y12 assay and RPA > 46.2% with 20 µM ADP in LTA)8,17 and the newly suggested PRU threshold (PRU ADP-PGE1> 208);18 (ii) the percentage of patients with a PRU ≤ 30 to estimate the rate of patients with a hyper-response to P2Y12 inhibitor and high on-treatment platelet inhibition.19

Statistical analysis

Based on previous platelet function studies,8,2022 we calculated that we needed to include at least 163 elderly patients to yield 95% power with an alpha-risk error of 0.05, assuming a 40% standard deviation to demonstrate that elderly patients have a 20% higher level of platelet reactivity measured by the VN-P2Y12.

Continuous variables were analysed for a normal distribution with the Shapiro–Wilk test. Continuous variables following a normal distribution are presented as mean ± SD and were compared with Student's unpaired t-test. Variables not following a normal distribution were compared with Mann–Whitney rank-sum test. The Kruskall–Wallis test was used for multiple comparisons across decades and to generate P-values for trend tests of non-normally distributed variables. Categorical variables are presented as counts and percentages and were compared by means of the χ2-test or Fisher's exact test. Correlations between quantitative variables were assessed with Pearson correlation coefficients.

A stepwise multivariable logistic regression analysis was used to identify independent variables associated with HPR (PRU ADP-PGE1> 235).

After univariate analysis, variables that presented a significant association with HPR (P-value < 0.20) were entered in the multivariate model. All P-values are two-sided and a value of P< 0.05 was considered significant. All analyses were performed with the use of SAS software, version 9.1 (SAS Institute, USA). All the authors had full access to and take full responsibility for the integrity of the data.


Study population

The entire cohort of this study comprised 1331 patients, divided into 205 elderly patients and 1126 younger patients. Thienopyridine treatment was a maintenance dose (MD) of clopidogrel 75 mg in 1027 patients, clopidogrel 150 mg in 139 patients, and prasugrel 10 mg in 165 patients. Patients' baseline characteristics are provided in Table 1. The indication of prasugrel 10 mg in elderly patients was STEMI presentation (81.9%) and NSTE-ACS with diabetic status (18.1%). The mean age of the elderly population was 80.2 ± 4 years and as expected the elderly group was marked by a three-fold increase in the proportion of women (28.7%), a three-fold increase in the rate of renal impairment (clearance <60 mL/min) and lower body weight. In the elderly group of patients, there was a higher utilization of concomitant medications; including calcium inhibitors, statins, and proton pump inhibitors (PPI). The more frequent PPI and dose used were pantoprazol 20 mg (33.3%) or 40 mg (13.4%), esomeprazol 20 mg (30.1%) or 40 mg (6.0%), and omeprazole 20 mg (17.2%). In contrast, the rates of diabetes were similar in both groups (27 and 23%, P= 0.3) and there was a similar genotype distribution of the CYP2C19*2 loss of function allele with 66.8 vs. 66.7% for *1/*1, 29.7 vs. 29.8% for *1/*2 and 3.6 vs. 3.6% for *2/*2 in elderly and younger patients, respectively (P= NS for all). Platelet response was available in 91% of patients with the VN-P2Y12 platelet function test and in 90.1% of patients with the LTA; all patients had a least one measure of platelet reactivity. Results of platelet response for the entire study population measured by the VN-P2Y12 (ADP-PGE1) are presented in Figure 1 and reported in Table 2 for both platelet function tests.

View this table:
Table 1

Patients characteristics

View this table:
Table 2

Platelet response to thienopyridines

Figure 1

Platelet reactivity assessed by the VerifyNow (P2Y12) test (PRU ADP-PGE1) in elderly and younger coronary artery disease patients under maintenance dose of clopidogrel 75 mg/day, clopidogrel 150 mg/day, and prasugrel 10 mg/day. P-value indicates the comparison of the P2Y12 reaction units in young vs. elderly patients. VN-P2Y12 values were available for 91% of patients and the number of patients (n) is indicated below each plot.

Age and platelet response to clopidogrel 75 mg

Platelet reactivity under an MD of 75 mg of clopidogrel was higher in elderly patients than younger patients, whichever the test used (Table 2). The rate of HPR (PRU > 235) was two-fold higher in the elderly patients than in younger patients (38.2 vs. 18.2%; P< 0.0001), with similar and consistent results when the population was analysed in decades of age (Figure 2A). When HPR was defined with a more sensitive definition (PRU > 208), the rate of non-responders was higher but still differed between elderly and younger patients (50.4 vs. 26.6%; P< 0.0001). Platelet reactivity correlated weakly but significantly with age (Figure 3). This correlation was not linear and we observed that platelet reactivity increased across decades of age, with a marked increase for patients aged between 65 and 75 years and over 75 years old for PRU in response to ISO-TRAP and to ADP-PGE1 resulting in a lower platelet inhibition with clopidogrel in older patients (Figure 2B).

Figure 2

Rate of high platelet reactivity (PRU>235) (Figure 2A) and mean inhibition (%) (Figure 2B) in patients treated with an MD of 75 mg of clopidogrel according to decades of age identified with the VN-P2Y12 assay. Mean inhibition corresponds to the ratio PRU iso-TRAP/PRU ADP-PGE1. Asterisks indicate P<0.05 with Kruskall–Wallis test for multiple comparison.

Figure 3

Platelet reactivity with a MD of 75 mg of clopidogrel measured with the VN-P2Y12 (PRU ADP-PGE1) (Y-axis) correlated with age (X-axis).

Platelet response measured by LTA showed similar differences between elderly and younger patients (Table 2) with a similar pattern across decades of age (P= 0.001) (data not shown). When using the RPA definition for HPR results were consistent, although not significant (36.4 vs. 28.4%; P= 0.07).

At the other end of the spectrum of platelet inhibition, the rate of elderly patients with high on-treatment platelet inhibition (PRU ≤ 30) was only 2.3% when compared with 8.7% in younger patients (P=0.007) (Figure 4).

Figure 4

Rate of high on-treatment platelet inhibition assessed by the VN-P2Y12 (PRU≤30) in elderly and young coronary artery disease patients treated with a maintenance dose of clopidogrel 75 mg/day, clopidogrel 150 mg/day, and prasugrel 10 mg/day.

Age and platelet response to clopidogrel 150 mg or prasugrel 10 mg

Interestingly, patients treated with clopidogrel 150 mg or prasugrel 10 mg displayed a non-significant trend towards higher platelet reactivity in elderly compared with younger ones with a smaller difference of PRU between the two age categories (+30; P=0.17 and +20; P=0.10) than the difference measured on clopidogrel 75 mg (+45; P<0.0001). Measurements of RPA were consistent showing higher levels of platelet reactivity in elderly than in younger patients when treated with clopidogrel 150 mg (37.2±21 vs. 31.0±23%; P=0.3) or prasugrel 10 mg (16±17 vs. 10.4 ±14%; P=0.05).

The rate of HPR was numerically higher in elderly patients than in younger patients both with clopidogrel 150 mg (39.1 vs. 26.5%, P=0.3) and with prasugrel 10 mg (6.9 vs. 3.1% P=0.3).

The rate of patients with HPI was 5.0% with clopidogrel 150 mg and 40.3% with prasugrel 10 mg. Elderly patients always had, at least numerically, lower rates of HPI when compared with younger patients (Figure 4).

Independent factor of high platelet reactivity in patients treated with a maintenance dose of clopidogrel 75 mg

Factors associated with HPR in the group of 1027 patients treated with a 75 mg MD of clopidogrel are presented in Table 3. After adjustment for factors of HPR, age>75 years old remained an independent correlate of HPR (OR: 1.83, 95% CI: 1.23–2.64; P=0.02). Renal impairment defined by a creatinine clearance <60mL/min, female gender, diabetes status, and the use of PPI were the other independent correlates of HPR. As expected, the loss of function allele CYP2C19*2 impacted the response to clopidogrel and was associated with a higher level of PRU in a stepwise manner; 153±86 PRU for *2/*2, 182±73 PRU for *1/*2 genotype, and 223±78 PRU for *1/*1 genotype; P<0.0001 for multiple comparison. The impact of the loss of function allele was present in younger and elderly patients but was not included into the multivariate analysis as only 59.1% of patient provided written informed consent for genetic analysis.

View this table:
Table 3

Multivariable logistic regression for predictors of high on-treatment platelet reactivity (PRU < 235) in the population treated with clopidogrel 75 mg


Elderly patients are generally more vulnerable to adverse effects of antithrombotic drugs, and the estimation of the risk–benefit balance in this population is difficult. It is only recently, that the American Heart Association in collaboration with the Society of Geriatric Cardiology has recommended enrolling elderly subjects proportionate to their prevalence in the treated population.1,2 Moreover, registries have found that elderly patients presenting with an ACS receive less frequently evidence-based therapies,3 highlighting the need for studies targeted specifically to the elderly.

The result of the SENIOR-PLATELETS study brings new insight on platelet response in this population. We report that HPR in elderly patients treated with an MD of 75 mg of clopidogrel is frequent, with a two- to three-fold higher rate than in younger patients. These results are consistent across the two different platelet function tests used and there is a similar trend towards higher platelet reactivity in patients treated with prasugrel 10 mg or in those on a higher MD of clopidogrel although the differences were smaller and not statistically significant. Along the same lines, elderly patients have a lower rate of high on-treatment platelet inhibition (PRU≤30) when compared with younger patients.

It is now well established that clopidogrel response is highly variable, and many components of this variability have been identified.21,23,24 We can confirm that age is a strong and independent factor of high on-treatment platelet reactivity,25 even if in previous publication other factors like diabetes or genetic traits such as CYP2C19*2 appear to be even stronger factors of resistance.21,24 Our study suggests that advanced age integrates several factors associated with HPR, such as PPI use and renal impairment, leading to the higher unadjusted OR for HPR. Adjustment for these confounding factors confirms, however, that age itself is an independent correlate of HPR.

The first explanation for these finding is that elderly patients bear a higher rate of platelet reactivity at baseline than younger ones, which is documented by the higher level of PRU in response to iso-TRAP observed in elderly patients. However, the P2Y12-related platelet reactivity measured by PRU in response to the ADP-PGE1 agonist increased even more with age, resulting in a lower level of inhibition (ADP-PGE1/iso-TRAP ratio). These findings indicate that the higher rate of HPR is partially due to a higher baseline level of platelet reactivity in the elderly but also to a lower response to the drug, which itself raises questions about the degree of exposure to the clopidogrel active metabolite in elderly patients. Indeed, the activity of hepatic enzymes such as CYP450 2C19 or paraxonase 1, a debated determinant of response to clopidogrel efficacy,26 appear to be influenced by age.27,28 Therefore a lower response could be due to the physiological reduction in liver function in the elderly. However, to our knowledge there is no published pharmacological data on clopidogrel active metabolite exposure in elderly patients that could confirm or not this hypothesis.

The second explanation is that elderly patients have more frequent co-morbidities as demonstrated by the higher rate of renal insufficiency and the higher rate of PPI use, both independent factors of biological hypo-response in the present study as in previous reports.3033 As identified in previous studies,21,34,35 diabetes was found to be associated with HPR in our study; wherever increased BMI,36 a factor associated with a higher rate of HPR,22,3739 was more frequent in our elderly population but not independently associated with HPR. Logically, the genotype distribution of cytochrome P450 genetic variants associated with HPR21,24 was similar between elderly and younger patients.

Another possible explanation for the poorer response to clopidogrel could be accelerated platelet turnover due to more advanced vascular atherosclerosis in this group;29 however, this hypothesis remains to be demonstrated.

If age is an integration factor of high on-treatment platelet reactivity, higher doses of clopidogrel or more potent P2Y12 antagonists in elderly patients could potentially overcome this issue.4042 Nevertheless, in our study elderly patients chronically treated with higher doses of clopidogrel (150 mg a day) or with prasugrel (10 mg a day, a high dose for the elderly) still demonstrate higher levels of platelet reactivity than younger patients, although the difference was smaller than for clopidogrel 75 mg.

Higher level of platelet inhibition has been shown to increase bleeding events1 and treating elderly patients, in whom the risk of bleeding is generally higher,43 with more intensive therapy raises concerns among physicians. In this study, using a cut-off value of PRU of 30 to detect high on-treatment platelet inhibition, we observed that elderly patients are still less frequently exposed to over-treatment than younger patients, whichever the thienopyridine used. These findings remain difficult to interpret in terms of clinical implication. Indeed, as elderly constitute a well-known high-risk group for bleeding events the relationship between bleeding events and platelet reactivity seems to be a more complex than the one between HPR and ischaemic events. However, it suggests that the increase in bleeding risk in elderly patients is not entirely attributable to platelet function and is also likely due to the characteristics and co-morbidities of these patients. Therefore, more data are needed to understand the link between platelet inhibition and the bleeding risk in the elderly.

Of interest it has to be noted that doubling the MD of clopidogrel had only a limited impact on platelet inhibition in our study as in the GRAVITAS44 randomized study and might explain the lack of impact on clinical events also found in the CURRENT45 randomized study. These observations emphasize the need for more clinical studies such as the ARCTIC trial46 on the potential clinical additive value of platelet function tests for adjustment of antiplatelet therapy, which is currently largely unproven. Regarding the dose of prasugrel 10 mg, the variability of response observed in the present work is much higher than the one observed in the blinded randomized trial ACAPULCO,41 highlighting the differences in platelet response between highly selected patients in randomized trial and high-risk real-life patient.

This study has the limitations and biases inherent to registries, but represents a ‘real-life’ population. Despite the large number of patients included in this study, the limited size of the elderly population taking the same treatment and the lack of follow-up limits our conclusion. The choice of one of the three regimens of DAPT was left to the discretion of the physician in charge of the patients in the absence of randomization and difference in characteristics are therefore detailed in Table 1 and commented in the ‘Results’ section. However, the multivariate analysis in the largest group of patients (clopidogrel 75 mg) allows evaluating the independent impact of age over characteristics and minimized this bias. Another limitation is that in a recent consensus on HPR,17 three different tests were proposed for the evaluation of HPR and we used only two of these three tests, the third being the vasodilator-stimulated phosphoprotein (VASP) index. Our study has technical limitations, the first being the absence of baseline measurement and measurement of the active metabolite of clopidogrel and prasugrel, which is very challenging to obtain outside of a limited sized randomized pharmacological study; the second being that there was no objective way to rule out a difference of compliance between younger and elderly patients. The comparison of the biological response to different thienopyridine regimens in the elderly population would have been of high interest but could not be done in our study due to the absence of randomization (direct comparison) and the limited number of elderly patients under prasugrel 10 mg (indirect comparison); however, this question is actually addressed in the ongoing randomized GENERATIONS study (NCT01107912) including a comparison between 5 and 10 mg in elderly patients. Finally, in the absence of a consensus on high on-treatment platelet inhibition we cannot extrapolate on these results, although they are consistent and supportive of the main results of our study.

In conclusion, advanced age integrates several risk factors for high on-treatment platelet reactivity and is independently associated with an impaired response to clopidogrel 75 mg. Clopidogrel 150 mg or prasugrel 10 mg blunt, but do not eliminate the difference in response observed between old and young patients. More research needs to be conducted to confirm these findings, and ideally the impact of stronger and more consistent P21Y12 inhibition in elderly patients and the role of platelet function testing in this high-risk population should be addressed in a dedicated randomized trial.


This study was supported by the ACTION (Allies in Cardiovascular Trials Initiatives and Organized Networks) Group and a research grant from the Société and Fédération Française de Cardiologie.

Conflict of interest: Dr J.S. has received research grants from Sanofi-Aventis, Daiichi-Sankyo, Eli Lilly, INSERM, Fédération Française de Cardiologie and Société Française de Cardiologie; consultant fees from Daiichi-Sankyo and Eli Lilly; and lecture fees from AstraZeneca, Daiichi-Sankyo, and Eli Lilly. Pr Montalescot discloses the following relationships: Research Grants (to the Institution) from Abbott Vascular, AstraZeneca, BMS, Boston Scientific, Cordis, Eli Lilly, Fédération Française de Cardiologie, Fondation de France, Guerbet Medical, INSERM, ITC Edison, Medtronic, Pfizer, Sanofi-Aventis, Servier, Société Française de Cardiologie, Stago; consulting or lecture fees from AstraZeneca, Bayer, Boehringer-Ingelheim, Cardiovascular Research Foundation, Cleveland Clinic Research Foundation, Daiichi-Sankyo, Duke Institute, Eli Lilly, Europa, Lead-Up, GSK, Institut de Cardiologie de Montreal, Menarini, Nanospheres, Novartis, Pfizer, Portola, Sanofi-Aventis, The Medicines Company, TIMI study group. Pr Collet has received research grants from Bristol-Myers Squibb, Sanofi-Aventis, Eli Lilly, Guerbet Medical, Medtronic, Boston Scientific, Cordis, Stago, Centocor, Fondation de France, INSERM, Fédération Française de Cardiologie, and Société Française de Cardiologie; consulting fees from Sanofi-Aventis, Eli Lilly, and Bristol-Myers Squibb; and lecture fees from Bristol-Myers Squibb, Sanofi-Aventis, and Eli Lilly. Dr F.B. has received lecture fees from Roche, Sanofi-Aventis, Pfizer, and Astellas. Dr S.A.O'C. has received research grants from Menarini and the European Society of Cardiology. Dr J.-S.H. reports receiving research grant support from Fondation de France, INSERM, Fédération Française de Cardiologie, Biotronik, Medco Research Institute; consulting fees from Biotronik, Medco Health Solutions; and, lecture fees from Sanofi-Aventis, Daiichi Sankyo, Eli Lilly, Bristol-Myers Squibb.


We would like to thank Ghalia Anzaha, Delphine Brugier and Sophie Galier for their technical assistance.


  • Both authors contributed equally to this work and are considered as first co-authors.

  • See page 1187 for the editorial comment on this article (doi:10.1093/eurheartj/ehr458)


View Abstract