Skip to main content
European Heart Journal. Cardiovascular Pharmacotherapy logoLink to European Heart Journal. Cardiovascular Pharmacotherapy
. 2022 Apr 30;8(8):777–785. doi: 10.1093/ehjcvp/pvac032

Cost–effectiveness of ticagrelor in patients with type 2 diabetes and coronary artery disease: a European economic evaluation of the THEMIS trial

Philippe Gabriel Steg 1, Deepak L Bhatt 2, Stefan K James 3, Oliver Darlington 4, Louise Hoskin 5, Tabassome Simon 6, Kim M Fox 7, Lawrence A Leiter 8, Shamir R Mehta 9, Robert A Harrington 10, Anders Himmelmann 11, Wilhelm Ridderstråle 12, Marielle Andersson 13, Héctor Bueno 14,15, Leonardo De Luca 16, Amarjeet Tank 17, Carl Mellström 18, Phil McEwan 19,
PMCID: PMC9716869  PMID: 35488865

Abstract

Aims

To conduct a health economic evaluation of ticagrelor in patients with type 2 diabetes and coronary artery disease (CAD) from a multinational payer perspective. Cost–effectiveness and cost–utility of ticagrelor were evaluated in the overall effect of Ticagrelor on Health Outcomes in Diabetes Mellitus Patients Intervention Study (THEMIS) trial population and in the predefined patient group with prior percutaneous coronary intervention.

Methods and results

A Markov model was developed to extrapolate patient outcomes over a lifetime horizon. The primary outcome was incremental cost–effectiveness ratios (ICERs), which were compared with conventional willingness-to-pay thresholds [€47 000/quality-adjusted life-year (QALY) in Sweden and €30 000/QALY in other countries].

Treatment with ticagrelor resulted in QALY gains of up to 0.045 in the overall population and 0.099 in patients with percutaneous coronary intervention (PCI). Increased costs and benefits translated to ICERs ranged between €27 894 and €42 252/QALY across Sweden, Germany, Italy, and Spain in the overall population. In patients with prior PCI, estimated ICERs improved to €18 449, €20 632, €20 233, and €13 228/QALY in Sweden, Germany, Italy, and Spain, respectively, driven by higher event rates and treatment benefit.

Conclusion

Based on THEMIS results, ticagrelor plus aspirin compared with aspirin alone may be cost–effective in some European countries in patients with T2DM and CAD and no prior myocardial infarction (MI) or stroke. Additionally, ticagrelor is likely to be cost–effective across European countries in patients with a history of PCI.

Keywords: Coronary artery disease, Cost–effectiveness, Percutaneous coronary intervention, Ticagrelor, Type 2 diabetes

Introduction

Type 2 diabetes mellitus (T2DM) had an estimated global prevalence of ∼500 million in 2019.1 T2DM is associated with an increased risk of vascular disease, with around one third of patients diagnosed with cardiovascular disease (CVD) and a two- to four-fold increased risk of developing events such as cardiovascular (CV) death, myocardial infarction (MI), stroke, and amputation.2–4 Patients with coronary artery disease (CAD) and T2DM represent a group at particularly high risk of atherothrombotic events.5–7 Even in patients without a prior ischemic event, the combination of CAD and T2DM substantially increases the risk of suffering CV events,8 with some estimates placing the risk of patients with T2DM and CAD but without prior MI in the range of patients without T2DM and with prior MI.9 Furthermore, classic CV risk factors such as hypertension and elevated cholesterol are common and often not at goal in patients with atherothrombosis, further increasing the risk of CV events in this population.10

Current standard antithrombotic preventive therapy in patients with established CAD and T2DM is low-dose aspirin. Ticagrelor is currently indicated to reduce the rate of CV death, MI, and stroke in patients with acute coronary syndrome or a history of MI by the North American and European regulatory agencies,11,12 and in patients with CAD and T2DM who have a history of percutaneous coronary intervention (PCI) by Health Canada.13 The clinical benefits of adding ticagrelor to background preventive therapy with aspirin were investigated in patients with CAD and T2DM but without prior MI or stroke in the effect of Ticagrelor on Health Outcomes in Diabetes Mellitus Patients Intervention Study (THEMIS; NCT01991795).14–20 This multinational, randomized, double-blind, placebo-controlled trial evaluated the effect of 90 mg ticagrelor (lowered to 60 mg partway through the trial in line with product labelling) twice daily on the incidence of CV death, MI, or stroke in 19 220 patients with T2DM.14,21 In THEMIS, adding ticagrelor to aspirin reduced the number of events of the composite endpoint of CV death, MI, and stroke, although with an increase in major bleeding compared with aspirin alone.14–20 Net clinical benefit was assessed using a prespecified irreversible harm composite endpoint of all-cause death, MI, stroke, fatal bleed, or intracranial haemorrhage. While there was no net clinical benefit in patients without a history of PCI, net clinical benefit was favourable in patients with a history of PCI14–20 with a 15% relative reduction (P-value for interaction = 0.012).

Given these findings, it is important to determine the long-term cost–effectiveness of ticagrelor in these populations. As such, the objective of this study is to conduct a health economic evaluation of ticagrelor added to standard therapy, including aspirin, vs. standard therapy alone from a multinational payer perspective. The analysis was conducted in the overall trial population, in patients with prior PCI, and, given that ‘modern’ PCI typically involves placement of a stent, in patients with prior PCI and stenting. Given the heterogeneity of costs and willingness-to-pay thresholds in Europe, we conducted the analysis in four major European countries.

Methods

Decision problem and model

A de novo Markov state-transition model was developed to model the health economic outcomes associated with the use of ticagrelor for the prevention of CV outcomes in CAD patients with T2DM. The health states consisted of event-free (patients not experiencing a MI or stroke event), non-fatal MI event, non-fatal stroke event, and all-cause mortality in addition to health states capturing the long-term impacts of the incidence of MI and/or stroke. In addition to efficacy outcomes, safety outcomes, including Thrombolysis In Myocardial Infarction (TIMI) Major bleeding, TIMI Minor bleeding, or bleeding requiring medical attention and dyspnoea, were also included in the model, and assumed to have a transient impact on cost and quality of life.22 In addition, the incidence of major amputation was also modelled. The model employed a lifetime perspective to accommodate the chronic nature of CAD, with a six-monthly cycle length and a half-cycle correction. The primary model outcome was the incremental cost–effectiveness ratio (ICER), expressed as the cost per quality-adjusted life-year (QALY) gained. Analysis was conducted from German, Italian, Spanish and Swedish healthcare payer perspectives, chosen as large European countries that frequently require cost–effectiveness evaluations for reimbursement. Explicit willingness-to-pay (WTP) thresholds are not established in Germany, Italy, Spain, or Sweden.23 The ICERs for each setting were compared with WTP thresholds commonly used in published literature [an assumption of €30 000/QALY for Germany, Italy, and Spain24 and €47 000/QALY (500 000 kr/QALY) for Sweden].25 For Sweden, an exchange rate of €1 = 10.57 kr was used. Differences in WTP thresholds between countries are a result of differences in costs associated with treatment and event management, and different economic valuations of health benefits. Future costs and benefits were discounted at a rate of 3.0% per annum.26 An overview of the model structure is presented in Figure 1.

Figure 1.

Figure 1

Cost–effectiveness model schematic.

Analysis

The base case analysis reflected the enrolled patient population of THEMIS. In addition, the analysis was repeated in prespecified subgroups of patients with a history of PCI, and patients with a history of PCI and stent, representing >90% of the patients enrolled in THEMIS with a history of PCI. Table 1 shows baseline profiles for the THEMIS trial patient population as well as patients with a history of PCI and those who were stented. One-way deterministic sensitivity analysis was used to explore the impact of varying input parameters over specified ranges on model outcomes. Probabilistic sensitivity analysis characterized overall parameter uncertainty using Monte Carlo simulation, with 1000 iterations, on modelled outcomes.

Table 1.

Baseline profiles for the overall population and the history of PCI, and the history of PCI with stent subgroups data are presented as mean ± standard deviation or n (%)

Clinical characteristics Full THEMIS population (N = 19 220) History of PCI subgroup (N = 11 154) History of PCI with stent subgroup (N = 10 295)
Baseline demographics
 Age (years) 66.31 ± 7.76 66.36 ± 7.76 66.3 ± 7.77
 Aged 65 ≤ 75 8247 (42.9%) 4799 (43.0%) 4414 (42.9%)
 Aged ≥ 75 3039 (15.8%) 1762 (15.8%) 1616 (15.7%)
 Male 13 189 (68.6%) 7718 (69.2%) 7147 (69.4%)
 BMI > 30 kg/m2 8206 (42.7%) 4587 (41.1%) 4203 (40.8%)
 Current smoker 2094 (10.9%) 1334 (12.0%) 1237 (12.0%)
 Central and South America 2178 (11.3%) 1166 (10.5%) 1061 (10.3%)
 Europe and South Africa 9759 (50.8%) 5427 (48.7%) 5007 (48.6%)
 North America 2995 (15.6%) 1667 (15.0%) 1491 (14.5%)
 Asia and Australia 4288 (22.3%) 2894 (26.0%) 2736 (26.6%)
 T2DM ≤ 10 years 9702 (50.5%) 5595 (50.2%) 5176 (50.3%)
 HbA1c ≤ 7% 9108 (47.4%) 5315 (47.7%) 4915 (47.7%)
 eGFR < 60 mL/min/1.73 m2 4549 (23.7%) 2589 (23.2%) 2363 (23.0%)
 T2DM complications 4910 (25.5%) 2734 (24.5%) 2493 (24.2%)
Prior medical history
 CABG 5537 (28.8%) 1346 (12.1%) 1106 (10.7%)
 Proportion with coronary arterial revascularization 15 345 (79.8%) 11 154 (100.0%) 10 295 (100.0%)
 Multivessel CAD 11 935 (62.1%) 6310 (56.6%) 5808 (56.4%)
 Polyvascular disease 2579 (13.4%) 1339 (12.0%) 1201 (11.7%)
 PAD 1687 (8.8%) 905 (8.1%) 810 (7.9%)
 PCI (any) 11 154 (58.0%) 11 154 (100.0%) 10 295 (100.0%)
 PCI with stent 10 295 (53.6%) 10 295 (92.3%) 10 295 (100.0%)
 Angina 10 801 (56.2%) 6606 (59.2%) 6087 (59.1%)
 Insulin use 5508 (28.7%) 3036 (27.2%) 2800 (27.2%)
 PPI use 4901 (25.5%) 2986 (26.8%) 2779 (27.0%)
 Statin use 17 266 (89.8%) 10 107 (90.6%) 9341 (90.7%)

BMI, body mass index; CABG, coronary artery bypass graft; CAD, coronary arterial disease; eGFR, estimated glomerular filtration rate; HbA1c, haemoglobin A1c; PAD, peripheral arterial disease; PCI, percutaneous coronary intervention; PPI, proton-pump inhibitor; T2DM, type 2 diabetes mellitus.

Mortality and cardiovascular events

The incidence of MI, stroke and mortality are described by covariate-adjusted survival equations derived from the analysis of individual patient level data from the THEMIS trial. This allows the model to capture changing hazards over time due to changes in risk factors such as age or the impact of different patient characteristics on modelled outcomes. In order to estimate the probability of an event occurring in each model cycle, the relative decrease in survival was estimated as a function of the probability of survival in the previous cycle. The benefits of ticagrelor in terms of prevention of CV events and mortality are captured through the application of hazard ratios to a survival equation describing the control arm of the trial; this approach was taken to ensure plausible differences in outcomes when extrapolating beyond the end of trial-follow up. Equation covariates for all-cause mortality and non-fatal MI and stroke are available in the Supplemental Material. Hazard ratios for each endpoint and for each of the trial subgroups are presented in Table 2. The impact of CV events on the incidence of mortality is also captured through the application of hazard ratios, where patients having experienced a non-fatal MI or stroke in the preceding 6 months (one model cycle) are at an increased risk of mortality; additionally, patients with a history of CV events are also at an increased risk of mortality.

Table 2.

Hazard ratios associated with ticagrelor treatment

Hazard ratio
Event Mean SE
Full THEMIS population
 Non-fatal MI 0.83 0.0858
 Non-fatal stroke 0.81 0.1082
 All-cause mortality 0.99 0.0477
 Increased risk of mortality in cycle of CV event 1.18
 Increased risk of mortality in cycles following CV event 2.41
THEMIS-PCI population
 Non-fatal MI 0.78 0.106
 Non-fatal stroke 0.75 0.1433
 All-cause mortality 0.91 0.0649
 Increased risk of mortality in cycle of CV event 1.24
 Increased risk of mortality in cycles following CV event 2.25
THEMIS-PCI subpopulation with stent
 Non-fatal MI 0.74 0.1096
 Non-fatal stroke 0.75 0.1475
 All-cause mortality 0.88 0.0676
 Increased risk of mortality in cycle of CV event 1.31
 Increased risk of mortality in cycles following CV event 2.15

CV, cardiovascular; MI, myocardial infarction; PCI, percutaneous coronary intervention; SE, standard error.

Health-related quality of life

Each of the modelled health states is assigned a utility weight, and the proportion of patients residing within each health state informs the accrual of QALYs over time. Patients in the event-free health state are subject to the utility value specified for people with CAD and T2DM without other complications, which is derived from a mixed-effects regression model fitted to individual patient data from THEMIS and adjusted for population baseline characteristics. Patients in each of the remaining health states are subject to a decrement in utility, relative to the utility associated with CAD and T2DM without complications. Mean utility values in THEMIS were consistent between treatment groups and over time. Using Swedish tariffs, for example, which resulted in the highest utility indices, mean utility values of 0.900 and 0.896 for ticagrelor and placebo, respectively, after 3 years, vs. 0.901 and 0.902 at baseline, were determined. Similar results were seen for Italy, with the lowest utility indices among the four countries, with ticagrelor and placebo means of 0.842 and 0.837 after 3 years and 0.845 and 0.847 at baseline. As such, utility estimates were derived from a pooled analysis of individual patient data from the THEMIS clinical trial. Linear mixed-effects regression models were fitted to predict patient reported utility values derived from EQ-5D-5 L questionnaires, which were collected at randomization and at six monthly intervals throughout the trial. In total, 140 461 EQ-5D-5 L questionnaires were completed, with missing questionnaires assumed to be missing at random. At baseline, 19 219 (99.99% of study participants) EQ-5D-5 L questionnaires were completed, decreasing to 9787 (50.92% of randomized study participants, or 82.19% of the patients with trial follow-up at 3 years) questionnaires at 3 years from baseline.

EQ-5D-5 L responses were converted to utility index scores using the most recently identified mapping algorithms for each of the countries studied. Values for Germany27 and Spain28 were derived from published hybrid models consisting of composite time-trade off and discrete choice experiment methods mapped directly to EQ-5D-5 L response data. Values for Sweden29 were derived from a similar direct mapping of EQ-5D-5 L response data, but using the time trade-off method. For Italy, no such report of direct mapping of EQ-5D-5 L responses to utility weights was identified. In this case, the EQ-5D-5 L data were first mapped to EQ-5D-3 L applying the mapping function developed by van Hout,30 in line with NICE technology assessment guidelines, and assuming that reported domain scores within individual questionnaires were uncorrelated. The converted Italian EQ-5D-3 L individual response data were then converted to utility weights using the most recent Italian model.31 Utility estimates were adjusted for patient characteristics, study follow-up, the incidence of MI, stroke, TIMI bleeding events, and adverse events. Model health state utility values are presented in Table 3.

Table 3.

Health-state utility inputs derived from analysis of EQ-5D-5 L responses mapped to EQ-5D-3 L levels

Country Sweden Germany Italy Spain
Mean (SE) Mean (SE) Mean (SE) Mean (SE)
Utility
CAD and T2DM, no complications 0.8984 (0.0006) 0.8886 (0.0009) 0.8406 (0.0011) 0.8463 (0.0010)
Utility decrement
MI event –0.0272 (0.0032) –0.0310 (0.0052) –0.0419 (0.0055) –0.0346 (0.0053)
Stroke event –0.0738 (0.0041) –0.1023 (0.0066) –0.1393 (0.0071) –0.1080 (0.0067)
Post MI –0.0155 (0.0025) –0.0201 (0.0040) –0.0269 (0.0043) –0.0226 (0.0041)
Post stroke –0.050 (0.0034) –0.0718 (0.0054) –0.0996 (0.0058) –0.0815 (0.0056)
Post stroke & MI –0.0655 (0.0042) –0.0919 (0.0068) –0.1265 (0.0073) –0.1042 (0.0069)
Amputationb –0.1750 (–0.0175)a –0.1750 (–0.0175)a –0.1750 (–0.0175)a –0.1750 (–0.0175)a
TIMI minor bleed/bleeding requiring medical attention –0.0069 (0.0016) –0.0093 (0.0027) –0.0120 (0.0028) –0.0094 (0.0027)
TIMI major bleed –0.0325 (0.0038) –0.0418 (0.0061) –0.0621 (0.0065) –0.0526 (0.0062)
Mild/moderate dyspnoea –0.0060 (0.0012) –0.0067 (0.0019) –0.0102 (0.0021) -0.0087 (0.0020)
Severe dyspnoea –0.0150 (0.0053) –0.0316 (0.0086) –0.0265 (0.0092) –0.0317 (0.0088)

CAD, coronary artery disease; MI, myocardial infarction; SE, standard error; TIMI, thrombolysis in myocardial infarction; T2DM, type 2 diabetes mellitus.

a

Assumed 10% of mean value.

b

Amputation was not monitored in trial as an independent event; value taken from literature.36

Resource use

All costs are reported in euros (€) and refer to the year 2019. One-off event costs are applied in the cycle of incidence for non-fatal MI, non-fatal stroke, and CV mortality events, in addition to any costs associated with treating treatment-related adverse events in order to capture the associated increase in healthcare resource utilization. Maintenance costs are defined as annual costs and are applied in all subsequent years for patients with a history of MI, stroke, or amputation. TIMI major bleeding events are defined to include intracranial bleeding, overt signs of haemorrhage associated with a drop in haemoglobin of ≥ 5 g/dL, or fatal bleeding events, and TIMI minor bleeding events are defined as overt signs of haemorrhage associated with a drop in haemoglobin of 3 to < 5 g/dL.32 As these events are serious and likely to require inpatient care, cost inputs were conservatively based on inpatient estimates of the costs associated with managing bleeding events. Less serious bleeding events requiring medical attention but not meeting the criteria for TIMI major and minor bleeds were assumed to require a single primary care visit. Similarly, although patients with mild or moderate dyspnoea may not require any additional medical attention, these events were conservatively assumed to require a single primary care visit; severe events were assumed to incur a cost associated with hospitalization.

Patients’ time on treatment was informed by unadjusted survival curves describing time to premature discontinuation of ticagrelor derived from the THEMIS clinical trial, and discontinuation in the placebo arm of the trial was not considered. All patients were assumed to discontinue treatment with ticagrelor after 4 years, aligned to the follow-up of the THEMIS trial. Ticagrelor treatment costs are applied while the cohort remains on therapy, with patients who have discontinued ticagrelor incurring costs associated with standard care alone. Country-specific costs are reported in Supplementary material online, Table S1. Where appropriate, costs were inflated to 2019 values using relevant consumer price indices.33–35 Costs reported in Swedish Krona (SEK) were converted to euros using an exchange rate of 10.57.

Results

In the overall THEMIS population, the ICER was below established WTP thresholds in Sweden and Spain (Table 4), with QALY gains of 0.045 and 0.043 per treated patient, respectively, based on a 4-year treatment period aligned to the duration of the trial, suggesting that ticagrelor may be a cost–effective treatment option for THEMIS-like patients in these countries. In Germany and Italy, the estimated ICER was above commonly used WTP thresholds, suggesting that ticagrelor may not be a cost–effective treatment option in these countries.

Table 4.

Cost–effectiveness result

THEMIS THEMIS-PCI THEMIS-PCI with stent
Outcome Ticagrelor Placebo Incremental Ticagrelor Placebo Incremental Ticagrelor Placebo Incremental
Total LYs 11.063 11.017 0.046 12.012 11.907 0.105 11.904 11.768 0.136
Sweden (WTP €47 000/QALY)
Total QALYs 9.892 9.848 0.045 10.744 10.645 0.099 10.647 10.521 0.126
Total costs €27 537 €25 820 €1718 €29 201 €27 381 €1820 €29 015 €27 153 €1862
Treatment €1914 €209 €1705 €1949 €226 €1723 €1949 €224 €1725
CV events €4806 €5035 –€229 €4535 €4831 –€296 €4596 €4940 –€344
Adverse events €858 €831 €28 €955 €935 €20 €943 €923 €20
T2DM management €19 958 €19 744 €214 €21 762 €21 389 €373 €21 527 €21 066 €461
ICER €38 428/QALY €18 449/QALY €14 751/QALY
Germany (WTP €30 000/QALY)
Total QALYs 9.771 9.727 0.045 10.614 10.516 0.098 10.518 10.392 0.126
Total costs €40 166 €38 348 €1818 €42 499 €40 471 €2028 €42 184 €40 052 €2132
Treatment €1901 €132 €1769 €1930 €143 €1787 €1930 €141 €1789
CV events €5894 €6130 –€236 €5238 €5528 –€290 €5307 €5636 –€329
Adverse events €809 €862 –€53 €918 €976 –€59 €904 €961 –€56
T2DM management €31 562 €31 224 €338 €34 414 €33 824 €591 €34 043 €33 314 €729
ICER €40 628/QALY €20 632/QALY €16 967/QALY
Italy (WTP €30 000/QALY)
Total QALYs 9.222 9.179 0.043 10.020 9.926 0.094 9.929 9.809 0.120
Total costs €20 060 €18 242 €1818 €21 106 €19 204 €1902 €20 981 €19 044 €1937
Treatment €2132 €319 €1812 €2177 €345 €1832 €2175 €341 €1834
CV events €3685 €3838 –€154 €3385 €3579 –€193 €3430 €3651 –€221
Adverse events €538 €526 €12 €600 €593 €7 €593 €585 €8
T2DM management €13 705 €13 558 €147 €14 943 €14 687 €256 €14 782 €14 466 €317
ICER €42 252/QALY €20 233/QALY €16 140/QALY
Spain (WTP €30 000/QALY)
Total QALYs 9.297 9.254 0.043 10.100 10.006 0.094 10.008 9.888 0.120
Total costs €21 636 €20 441 €1195 €23 048 €21 804 €1243 €22 920 €21 667 €1253
Treatment €1629 €364 €1266 €1673 €393 €1280 €1670 €388 €1282
CV events €4409 €4649 –€239 €4350 €4669 –€319 €4410 €4787 –€378
Adverse events €674 €665 €9 €753 €750 €3 €744 €740 €4
T2DM management €14 923 €14 763 €160 €16 272 €15 993 €279 €16 096 €15 752 €345
ICER €27 894/QALY €13 228/QALY €10 431/QALY

CV, cardiovascular; ICER, incremental cost–effectiveness ratio; QALY, quality-adjusted life-year; T2DM, type 2 diabetes mellitus; WTP willingness-to-pay.

In patients with a history of PCI or in patients with a history of PCI and stent, the estimated ICERs were consistently below the commonly cited WTP thresholds (€47 000/QALY in Sweden and €30 000/QALY in Germany, Italy, and Spain), with differences in cost–effectiveness estimates between countries primarily driven by the acquisition cost of ticagrelor. The reduced cost per QALY in patients with a history of PCI and stent is driven by the improved treatment effect observed in THEMIS.16

Treatment with ticagrelor was associated with QALY gains ranging between 0.094 and 0.099 in patients with a history of PCI and between 0.120 and 0.126 in patients with a history of PCI and stent. Ticagrelor, in addition to standard therapy, appeared to be a cost–effective treatment for the prevention of MI, stroke, and death in patients with CAD, T2DM with a history of PCI (with or without stent) over a lifetime horizon (Table 4).

Sensitivity analysis

Deterministic, or one-way, sensitivity analysis showed that cost–effectiveness was robust to the choice of model input parameters, with no scenarios or population subgroups resulting in an ICER over established WTP thresholds in any of the modelled countries in the THEMIS-PCI and THEMIS-PCI with stent populations (Supplementary material). Higher ICERs in the analysis representing the overall THEMIS trial inclusion criteria mean that conclusions of cost–effectiveness may vary based on the choice of model input parameters in this patient population.

Changes in patient characteristics had the largest impact on cost–effectiveness results, with improved cost–effectiveness in smokers and patients with estimated glomerular filtration rate < 60 mL/min/1.73 m2. The history of coronary artery bypass graft (CABG) and baseline HbA1c level were also influential to cost–effectiveness. In general, ticagrelor was more cost–effective in higher-risk patient populations, such as smokers and patients with impaired renal function, as a result of increased event incidence, leading to a greater number of avoided events than in low-risk patient populations.

Probabilistic sensitivity analysis showed that ticagrelor was cost–effective in 59% of simulations in the overall THEMIS population at a WTP threshold of €47 000/QALY in Sweden, and 32%, 30%, and 54% of simulations at a WTP threshold of €30 000/QALY in Germany, Italy, and Spain, respectively. In subgroup analysis, 92%, 78%, 78%, and 91% of simulations were cost–effective in the THEMIS-PCI subgroup and 98%, 91%, 91%, and 97% of simulations were cost–effective in the THEMIS-PCI with stent subgroup, for analysis in Sweden, Germany, Italy, and Spain, respectively (Figure 2).

Figure 2.

Figure 2

Cost–effectiveness acceptability curve.

Discussion

This analysis demonstrates that ticagrelor may be a cost-effective treatment option for the prevention of CV events and death when used in addition to standard care in patients with stable CAD and T2DM in Sweden and Spain and is likely to be cost–effective in patients with prior PCI (with or without stent) across the four countries studied (Sweden, Germany, Italy, and Spain). Results were principally driven by improved life expectancy as a result of lower mortality within trial follow-up and secondary reductions in mortality as a result of reduced CV event incidence, with larger benefits being observed in patients with prior PCI, with the magnitude of the health benefit comparing favourably to previously published estimates for antithrombotic treatments.36 This translated into significant QALY gains for patients treated with ticagrelor. In the patient population with prior PCI, ticagrelor reduced the incidence of CV events, with fewer non-fatal MI and stroke events, fewer CV deaths, but increased TIMI major bleeds. The avoidance of CV events was associated with modest QALY gains but contributed to cost savings, which partially offset the additional cost of treatment with ticagrelor. In order to assess the cost–effectiveness of treatment with ticagrelor over a lifetime horizon, there is a requirement to extrapolate outcomes beyond the observations of the THEMIS clinical trial, and this inherently introduces uncertainty. Deterministic sensitivity analysis, however, showed that the cost–effectiveness of ticagrelor was robust to the choice of model input parameters and influential patient characteristics.

The analysis may also underestimate incremental costs for patients treated with ticagrelor plus standard therapy, as consistent with health technology assessment guidelines, future costs unrelated to T2DM or CAD are not captured; for example, increased life expectancy will lead to more time at risk of hospitalization for other non-CV causes. Conversely, this analysis is likely to overestimate the costs associated with bleeding events, as model cost inputs conservatively assume that all TIMI major and minor bleeding events will require inpatient care, which is unlikely to be the case in real-world clinical practice. An additional consideration is that while CV events use risk equations that adjust for age, events such as TIMI major/minor bleeds, dyspnoea, and amputation are modelled as non-time-dependent incident events, where individuals are at equivalent risk irrespective of age. This could be viewed as a limitation if events were more likely to be fatal/high-risk in an elderly individual.

The probabilistic analysis also showed that when accounting for the uncertainty inherent in health economic modelling, ticagrelor is likely to be cost-effective in patients with a history of PCI, with or without stent. As a result of the size of the patient population with CAD and T2DM in Europe and the complex nature of managing the disease and its sequelae, these comorbidities impose a significant burden on patients and healthcare payers. Any reduction in the burden of disease through avoided CV events will have significant consequences on society in terms of reduced indirect costs, which were not captured within this study. Furthermore, this study did not explore the potential impact of avoided CV events on service delivery and the potential for the reduction of adverse clinical events to ameliorate pressure on strained healthcare services. The objective of this study was to evaluate ticagrelor across major European countries, and as such, the results of this study do not necessarily generalize to the rest of the world, particularly those countries that have different healthcare systems or reimbursement frameworks. However, the consistency of the findings across the included countries suggests that conclusions of cost–effectiveness may be applicable to other countries with similar healthcare systems.

Of note, this cost–effectiveness analysis was not designed to address the clinical value of ticagrelor in patients with type 2 diabetes and CAD, which has been described in detail in the overall trial population15 and in the subgroup with a history of prior PCI.16 In the overall population,15 the absolute risk increase in TIMI major bleeding was numerically higher than the absolute benefit in ischemic outcomes. The incidence of a prespecified composite outcome of irreversible harm (death from any cause, myocardial infarction, stroke, fatal bleeding, or intracranial haemorrhage) was similar in the ticagrelor group and the placebo group. In patients with prior PCI,16 the benefit to risk balance was more favourable, and there was a significant interaction between the history of prior PCI and the balance of risk and benefit measured by the net irreversible harm outcome. It is noteworthy that, while ticagrelor has been approved in patients with stable CAD (with or without diabetes) in the USA (and the Food and Drug Administration detailed their analysis for doing so37) and in patients with T2DM and prior PCI in Canada, it is not approved in Europe, and the present analysis should not be viewed as encouraging off-label use in Europe. The European Medicines Agency concluded that ‘the benefit-risk balance in the new indication proposed remains currently negative for the population with CAD and T2DM without any history of MI or stroke’.38 Yet, it is important that the present cost–effectiveness analyses pertaining to European countries be available in the public domain.

In conclusion, this analysis suggests that the cost–effectiveness of ticagrelor in addition to standard care in patients with stable chronic CAD and T2DM varies according to country and WTP threshold. However, in T2DM patients with prior PCI, ticagrelor is likely to represent a cost–effective option for the prevention of CV events across Swedish, German, Italian, and Spanish healthcare systems.

Supplementary Material

pvac032_Supplemental_File

Acknowledgment

Model development support was provided by Ryan Miller of Health Economics and Outcomes Research Ltd funded by AstraZeneca.

Medical writing support consisting of preparation, critical review, and revision was provided by Vanessa Gross of Health Economics and Outcomes Research Ltd funded by AstraZeneca.

Contributor Information

Philippe Gabriel Steg, Université de Paris, AP-HP, Hôpital Bichat, FACT (French Alliance for Cardiovascular trials) and INSERM-U1148, Paris, France.

Deepak L Bhatt, Department of Medicine, Brigham and Women's Hospital Heart and Vascular Center and Harvard Medical School, Boston, MA 02115, USA.

Stefan K James, Department of Medical Sciences and Uppsala Clinical Research Center, Uppsala University Hospital, 752 37 Uppsala, Sweden.

Oliver Darlington, Health Economics and Outcomes Research Ltd, Unit A Copse Walk, Cardiff Gate Business Park, Cardiff CF23 8RB, UK.

Louise Hoskin, Health Economics and Outcomes Research Ltd, Unit A Copse Walk, Cardiff Gate Business Park, Cardiff CF23 8RB, UK.

Tabassome Simon, Department of Clinical Pharmacology, Hôpital Saint Antoine, Unité de Recherche Clinique, 75012 Paris, France.

Kim M Fox, National Heart and Lung Institute, Imperial College and Royal Brompton Hospital, London SW3 6NP, UK.

Lawrence A Leiter, Li Ka Shing Knowledge Institute of St. Michael's Hospital, University of Toronto, Toronto, ON M5B 1T8, Canada.

Shamir R Mehta, Hamilton Health Sciences, Population Health Research Institute, McMaster University, Hamilton, ON L8S 4L8, Canada.

Robert A Harrington, Department of Medicine, Stanford University, Stanford, CA 94305, USA.

Anders Himmelmann, BioPharmaceuticals R&D, Cardiovascular, Renal and Metabolic, AstraZeneca, 431 50 Gothenburg, Sweden.

Wilhelm Ridderstråle, BioPharmaceuticals R&D, Cardiovascular, Renal and Metabolic, AstraZeneca, 431 50 Gothenburg, Sweden.

Marielle Andersson, BioPharmaceuticals R&D, Cardiovascular, Renal and Metabolic, AstraZeneca, 431 50 Gothenburg, Sweden.

Héctor Bueno, Department of Cardiology, Hospital Doce de Octubre, 28007 Madrid, Spain; MTCR Group, Centro Nacional de Investigaciones Cardiovasculares, 28029 Madrid, Spain.

Leonardo De Luca, Department of Cardiosciences, A.O. San Camillo-Forlanini, 00152 Rome, Italy.

Amarjeet Tank, AstraZeneca, Cambridge, CB2 8PA, UK.

Carl Mellström, BioPharmaceuticals R&D, Cardiovascular, Renal and Metabolic, AstraZeneca, 431 50 Gothenburg, Sweden.

Phil McEwan, Health Economics and Outcomes Research Ltd, Unit A Copse Walk, Cardiff Gate Business Park, Cardiff CF23 8RB, UK.

Funding

This work was supported by AstraZeneca.

Conflict of interest: Dr P.G.S. has received research grants from Amarin, Bayer, Sanofi, and Servier, and honoraria from Amarin, Amgen, AstraZeneca, Bayer, BristolMyersSquibb, Boehringer-Ingelheim, Idorsia, Novartis, NovoNordisk, Pfizer, Regeneron, Sanofi, and Servier. Dr D.L.B. serves as co-Chair of THEMIS and THEMIS-PCI, with research funding paid by AstraZeneca to Brigham and Women's Hospital. Dr Bhatt discloses the following relationships—Advisory Board: Bayer, Boehringer Ingelheim, Cardax, CellProthera, Cereno Scientific, Elsevier Practice Update Cardiology, Janssen, Level Ex, Medscape Cardiology, Merck, MyoKardia, NirvaMed, Novo Nordisk, PhaseBio, PLx Pharma, Regado Biosciences, and Stasys; Board of Directors: Boston VA Research Institute, DRS.LINQ (stock options), Society of Cardiovascular Patient Care, TobeSoft; Chair: Inaugural Chair, American Heart Association Quality Oversight Committee; Data Monitoring Committees: Acesion Pharma, Assistance Publique-Hôpitaux de Paris, Baim Institute for Clinical Research (formerly Harvard Clinical Research Institute, for the PORTICO trial, funded by St. Jude Medical, now Abbott), Boston Scientific (Chair, PEITHO trial), Cleveland Clinic (including for the ExCEED trial, funded by Edwards), Contego Medical (Chair, PERFORMANCE 2), Duke Clinical Research Institute, Mayo Clinic, Mount Sinai School of Medicine (for the ENVISAGE trial, funded by Daiichi Sankyo; for the ABILITY-DM trial, funded by Concept Medical), Novartis, Population Health Research Institute; Rutgers University (for the NIH-funded MINT Trial); Honoraria: American College of Cardiology (Senior Associate Editor, Clinical Trials and News, ACC.org; Chair, ACC Accreditation Oversight Committee), Arnold and Porter law firm (work related to Sanofi/Bristol-Myers Squibb clopidogrel litigation), Baim Institute for Clinical Research (formerly Harvard Clinical Research Institute; RE-DUAL PCI clinical trial steering committee funded by Boehringer Ingelheim; AEGIS-II executive committee funded by CSL Behring), Belvoir Publications (Editor-in-Chief, Harvard Heart Letter), Canadian Medical and Surgical Knowledge Translation Research Group (clinical trial steering committees), Cowen and Company, Duke Clinical Research Institute (clinical trial steering committees, including for the PRONOUNCE trial, funded by Ferring Pharmaceuticals), HMP Global (Editor-in-Chief, Journal of Invasive Cardiology), Journal of the American College of Cardiology (Guest Editor; Associate Editor), K2P (Co-Chair, interdisciplinary curriculum), Level Ex, Medtelligence/ReachMD (CME steering committees), MJH Life Sciences, Piper Sandler, Population Health Research Institute (for the COMPASS operations committee, publications committee, steering committee, and USA national co-leader, funded by Bayer), Slack Publications (Chief Medical Editor, Cardiology Today's Intervention), Society of Cardiovascular Patient Care (Secretary/Treasurer), WebMD (CME steering committees), Wiley (steering committee); Other: Clinical Cardiology (Deputy Editor), NCDR-ACTION Registry Steering Committee (Chair), VA CART Research and Publications Committee (Chair); Research Funding: Abbott, Afimmune, Aker Biomarine, Amarin, Amgen, AstraZeneca, Bayer, Beren, Boehringer Ingelheim, Boston Scientific, Bristol-Myers Squibb, Cardax, CellProthera, Cereno Scientific, Chiesi, CSL Behring, Eisai, Ethicon, Faraday Pharmaceuticals, Ferring Pharmaceuticals, Forest Laboratories, Fractyl, Garmin, HLS Therapeutics, Idorsia, Ironwood, Ischemix, Janssen, Javelin, Lexicon, Lilly, Medtronic, Merck, Moderna, MyoKardia, NirvaMed, Novartis, Novo Nordisk, Owkin, Pfizer, PhaseBio, PLx Pharma, Recardio, Regeneron, Reid Hoffman Foundation, Roche, Sanofi, Stasys, Synaptic, The Medicines Company, and 89Bio; Royalties: Elsevier (Editor, Braunwald's Heart Disease); Site Co-Investigators: Abbott, Biotronik, Boston Scientific, CSI, Endotronix, St. Jude Medical (now Abbott), Philips, and Svelte; Trustee: American College of Cardiology; Unfunded Research: FlowCo, Takeda. K.F. has received fees, honoraria and/or travel expenses from Servier, Astra Zeneca, CellAegis, Celixir, TauRx, UCB, and Broadview Ventures. He was a director of Vesalius Trials Ltd L.A.L. reports grants or contracts from Esperion, Kowa, Lexicon, Novartis, and The Medicines Company; consulting fees from Amgen, Amarin, AstraZeneca, Bayer, Boehringer Ingelheim, Eli Lilly, Esperion, GSK, HLS, Merck, Novartis, Novo Nordisk, Pfizer, and Sanofi; and payment or honoraria from Amgen, Amarin, AstraZeneca, Boehringer Ingelheim, Eli Lilly, HLS, Merck, Novartis, Novo Nordisk, and Servier. A.H. reports being an employee of AstraZeneca. W.R. reports being an employee of AstraZeneca. M.A. reports being an employee of AstraZeneca. C.M. reports being an employee of AstraZeneca. L.D.L. declares receiving consulting fees or honoraria from Amgen, Aspen, AstraZeneca, Bayer, Boehringer Ingelheim, Chiesi, Daiichi Sankyo, Eli Lilly, Menarini, Pfizer/Bristol-Myers Squibb, Sanofi, and Servier. S.J. reports Research grants to his institution from AstraZeneca, the Medicines Company, Bayer, and Jansen. Honoraria to his institution from AstraZeneca, Chiesi. A.M. reports being an employee of AstraZeneca.

Data availability

The data underlying this article cannot be shared publicly for the privacy of individuals that participated in the study. The data will be shared on reasonable request to the corresponding author.

References

  • 1. Kaiser AB, Zhang N, van der Pluijm W. Global prevalence of type 2 diabetes over the next ten years (2018-2028). Diabetes 2018;67:202–LB. [Google Scholar]
  • 2. Einarson TR, Acs A, Ludwig C, Panton UH. Prevalence of cardiovascular disease in type 2 diabetes: a systematic literature review of scientific evidence from across the world in 2007-2017. Cardiovasc Diabetol 2018;17:83. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3. Fox CS, Coady S, Sorlie PD, Levy D, Meigs JB, D'Agostino RBet al. Trends in cardiovascular complications of diabetes. JAMA 2004;292:2495–2499. [DOI] [PubMed] [Google Scholar]
  • 4. Sarwar N, Gao P, Seshasai SR, Gobin R, Kaptoge S, Di Angelantonio Eet al. Diabetes mellitus, fasting blood glucose concentration, and risk of vascular disease: a collaborative meta-analysis of 102 prospective studies. Lancet 2010;375:2215–2222. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Rapsomaniki E, Shah A, Perel P, Denaxas S, George J, Nicholas Oet al. Prognostic models for stable coronary artery disease based on electronic health record cohort of 102 023 patients. Eur Heart J 2013;35:844–852. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Bhatt DL, Eagle KA, Ohman EM, Hirsch AT, Goto S, Mahoney EMet al. Comparative determinants of 4-year cardiovascular event rates in stable outpatients at risk of or with atherothrombosis. JAMA 2010;304:1350–1357. [DOI] [PubMed] [Google Scholar]
  • 7. Steg PG, Bhatt DL, Wilson PWF, D'Agostino R, Ohman EM, Röther Jet al. One-year cardiovascular event rates in outpatients with atherothrombosis. JAMA 2007;297:1197–1206. [DOI] [PubMed] [Google Scholar]
  • 8. Cavender MA, Steg PG, Smith SC Jr, Eagle K, Ohman EM, Goto Set al. Impact of diabetes mellitus on hospitalization for heart failure, cardiovascular events, and death: outcomes at 4 years from the reduction of atherothrombosis for continued health (reach) registry. Circulation 2015;132:923–931. [DOI] [PubMed] [Google Scholar]
  • 9. Hasvold LP, Erlinge D, Svennblad B, Janzon M, Lindholm D, Sundell KAet al. Cardiovascular outcome in THEMIS -like type 2 diabetes patients in sweden: a nationwide observational study. Eur Heart J 2019;40:1286. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Bhatt DL, Steg PG, Ohman EM, Hirsch AT, Ikeda Y, Mas J-Let al. International prevalence, recognition, and treatment of cardiovascular risk factors in outpatients with atherothrombosis. JAMA 2006;295:180–189. [DOI] [PubMed] [Google Scholar]
  • 11. Food and Drug Administration . Brilinta® (ticagrelor) 2013 https://www.accessdata.fda.gov/drugsatfda_docs/label/2013/022433s010lbl.pdf (19 January 2021). [Google Scholar]
  • 12. European Medicines Agency . Brilique: Epar—product information 2011 https://www.ema.europa.eu/en/documents/product-information/brilique-epar-product-information_en.pdf (1 March 2021). [Google Scholar]
  • 13. AstraZeneca Canada Inc . Product monograph—brilinta®- ticagrelor tablets, 60 and 90 mg, platelet aggregation inhibitor. 2020. [Google Scholar]
  • 14. Bhatt DL, Fox K, Harrington RA, Leiter LA, Mehta SR, Simon Tet al. Rationale, design and baseline characteristics of the effect of ticagrelor on health outcomes in diabetes mellitus patients intervention study. Clin Cardiol 2019;42:498–505. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Steg PG, Bhatt DL, Simon T, Fox K, Mehta SR, Harrington RAet al. Ticagrelor in patients with stable coronary disease and diabetes. N Engl J Med 2019;381:1309–1320. [DOI] [PubMed] [Google Scholar]
  • 16. Bhatt DL, Steg PG, Mehta SR, Leiter LA, Simon T, Fox Ket al. Ticagrelor in patients with diabetes and stable coronary artery disease with a history of previous percutaneous coronary intervention (THEMIS-PCI): a phase 3, placebo-controlled, randomised trial. Lancet 2019;314:1169–1180. [DOI] [PubMed] [Google Scholar]
  • 17. Leiter LA, Bhatt DL, McGuire DK, Teoh H, Fox K, Simon Tet al. Diabetes-related factors and the effects of ticagrelor plus aspirin in the THEMIS and THEMIS-PCI trials. J Am Coll Cardiol 2021;77:2366–2377. [DOI] [PubMed] [Google Scholar]
  • 18. Abtan J, Bhatt DL, Held C, Simon T, Fox K, Mehta SRet al. Incidence of myocardial infarction types in patients treated with ticagrelor in the THEMIS trial. Circ Cardiovasc Interv 2021;14:e011035. [DOI] [PubMed] [Google Scholar]
  • 19. Bhatt DL, Steg PG. THEMIS and THEMIS-PCI. Eur Heart J 2019;40:3378–3381. [DOI] [PubMed] [Google Scholar]
  • 20. Steg PG, Bhatt DL. Is there really a benefit to net clinical benefit in testing antithrombotics? Circulation 2018;137:1429–1431. [DOI] [PubMed] [Google Scholar]
  • 21. Dobesh PP, Patel M. The parthenon clinical development program: the role of ticagrelor in patients with atherothrombotic disease. Cardiovasc Drugs Ther 2017;31:433–444. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Ducrocq G, Schulte PJ, Budaj A, Cornel JH, Held C, Himmelmann Aet al. Balancing the risk of spontaneous ischemic and major bleeding events in acute coronary syndromes. Am Heart J 2017;186:91–99. [DOI] [PubMed] [Google Scholar]
  • 23. Skoupá J, Annemans L, Hájek P. Health economic data requirements and availability in the european union: results of a survey among 10 european countries. Value Health Reg Issues 2014;4:53–57. [DOI] [PubMed] [Google Scholar]
  • 24. Vallejo-Torres L, García-Lorenzo B, Serrano-Aguilar P. Estimating a cost-effectiveness threshold for the spanish NHS. Health Econ 2018;27:746–761. [DOI] [PubMed] [Google Scholar]
  • 25. Svensson M, Nilsson FO, Arnberg K. Reimbursement decisions for pharmaceuticals in sweden: the impact of disease severity and cost effectiveness. Pharmacoeconomics 2015;33:1229–1236. [DOI] [PubMed] [Google Scholar]
  • 26. Dental and Pharmaceutical Benefits Agency (TLV) . General guidelines for economic evaluations from the pharmaceutical benefits board. 2003https://www.tlv.se/download/18.2e53241415e842ce95514e9/1510316396792/Guidelines-for-economic-evaluations-LFNAR-2003-2.pdf (1 March 2021). [Google Scholar]
  • 27. Ludwig K, Graf von der Schulenburg JM, Greiner W. German value set for the EQ-5D-5 L. Pharmacoeconomics 2018;36:663–674. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Ramos-Goni JM, Craig BM, Oppe M, Ramallo-Farina Y, Pinto-Prades JL, Luo Net al. Handling data quality issues to estimate the spanish EQ-5D-5 L value set using a hybrid interval regression approach. Value Health 2018;21:596–604. [DOI] [PubMed] [Google Scholar]
  • 29. Burstrom K, Teni FS, Gerdtham UG, Leidl R, Helgesson G, Rolfson Oet al. Experience-based swedish TTO and VAS value sets for EQ-5D-5 L health states. Pharmacoeconomics 2020;38:839–856. [DOI] [PubMed] [Google Scholar]
  • 30. van Hout B, Janssen MF, Feng YS, Kohlmann T, Busschbach J, Golicki Det al. Interim scoring for the EQ-5D-5L: mapping the EQ-5D-5 L to EQ-5D-3 L value sets. Value Health 2012;15:708–715. [DOI] [PubMed] [Google Scholar]
  • 31. Scalone L, Cortesi PA, Ciampichini R, Belisari A, D'Angiolella LS, Cesana Get al. Italian population-based values of EQ-5D health states. Value Health 2013;16:814–822. [DOI] [PubMed] [Google Scholar]
  • 32. Mehran R, Rao SV, Bhatt DL, Gibson CM, Caixeta A, Eikelboom Jet al. Standardized bleeding definitions for cardiovascular clinical trials: a consensus report from the bleeding academic research consortium. Circulation 2011;123:2736–2747. [DOI] [PubMed] [Google Scholar]
  • 33. Federal Statistical Office of Germany (Destatis Statistisches Bundesamt VD . Consumer price indices. https://www-genesis.destatis.de/genesis/online/data (29 July 2019). [Google Scholar]
  • 34. Insituto Nacional de Estadistica . Consumer price index https://www.ine.es/jaxiT3/Tabla.htm?t=22553&L=1 (26 July 2019). [Google Scholar]
  • 35. The Organisation for Economic Co-operation and Development . Consumer price indices. https://stats.oecd.org/Index.aspx?DataSetCode=PRICES_CPI (29 July 2019). [Google Scholar]
  • 36. Chen J, Bhatt DL, Dunn ES, Shi C, Caro JJ, Mahoney EMet al. Cost-effectiveness of clopidogrel plus aspirin versus aspirin alone for secondary prevention of cardiovascular events: results from the charisma trial. Value Health 2009;12:872–879. [DOI] [PubMed] [Google Scholar]
  • 37. Lackey LG, Garnett CE, Senatore F. Applying decision analysis to inform the US food and drug administration's benefit-risk assessment of ticagrelor for primary prevention of myocardial infarction or stroke based on THEMIS. Circulation 2021;144:655–658. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38. Withdrawal assessment report for Brilique: EMA/459475/2019 . https://www.ema.europa.eu/en/medicines/human/withdrawn-applications/brilique#overview-section (24 April 2022). [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

pvac032_Supplemental_File

Data Availability Statement

The data underlying this article cannot be shared publicly for the privacy of individuals that participated in the study. The data will be shared on reasonable request to the corresponding author.


Articles from European Heart Journal — Cardiovascular Pharmacotherapy are provided here courtesy of Oxford University Press

RESOURCES