Abstract
Aims
Among acute coronary syndromes (ACS), ST-segment elevation myocardial infarction (STEMI) has the most severe early clinical course. Recent randomized clinical trials have demonstrated that novel antithrombotic therapies improve in-hospital outcomes in STEMI patients. We aimed to describe the effectiveness and safety of P2Y12 receptor inhibitors in clinical practice in patients with STEMI based on data from contemporary European ACS registries.
Methods and results
Five registries from the PIRAEUS initiative (AAPCI/ADPAT, ALKK-PIC, AMIS Plus, Belgium STEMI, and EYESHOT) provided data for the assessment of P2Y12 receptor inhibitor-based dual antiplatelet therapy. Registries were heterogeneous in terms of setting, patient characteristics, and treatment selection. Matched pair analysis and propensity score matching were used to assess all-cause in-hospital death rates based on data from 25 250 patients (8577 patients on prasugrel, 5995 on ticagrelor, and 10 678 on clopidogrel). The odds ratio (OR) for the death of any cause when compared with clopidogrel was 0.72 [95% confidence interval (CI) 0.62–0.84, P < 0.001] in favour of the new P2Y12 receptor inhibitors (prasugrel and ticagrelor combined). In the comparison between prasugrel and ticagrelor, there were no relevant differences (OR 0.97, 95% CI 0.77–1.23; P = 0.81). Event rates of cardiovascular death and stroke were also substantially lower for the new P2Y12 receptor inhibitors. The differences between clopidogrel and prasugrel or ticagrelor on major bleeding were numerically in the same order as for death of any cause but were not statistically significant. No differences in ischaemic and bleeding outcomes were observed between prasugrel and ticagrelor.
Conclusion
This analysis suggests that the prasugrel or ticagrelor compared with clopidogrel have favourable outcomes in clinical practice while not being inferior in terms of safety.
Keywords: Effectiveness, Safety, Acute coronary syndromes, ST-segment elevation myocardial infarction, Antiplatelet agents, P2Y12 receptor inhibitors, Clopidogrel, Prasugrel, Ticagrelor, Propensity score matching, Real-world evidence
Introduction
Contemporary European Society of Cardiology (ESC) guidelines highlight the benefit of dual antiplatelet therapy (DAPT) consisting of acetylic salicylic acid (ASA) plus one of the P2Y12 receptor inhibitors, i.e. clopidogrel with a preference for prasugrel, or ticagrelor, with the aim to reduce the risk of both acute ischaemic complications and recurrent atherothrombotic events.1,2 Based on their higher antithrombotic potency and proven superiority in outcome trials, prasugrel and ticagrelor are given preference over clopidogrel.1,2 Recently, a direct comparison of ticagrelor and prasugrel suggested superiority of the latter compared with the former.3
Although prior studies have compared prasugrel or ticagrelor with clopidogrel in a real-world setting, a head-to-head comparison between the three available antiplatelet agents have never been performed.
We aimed to investigate whether in real-word practice, the newer P2Y12 inhibitors prasugrel and ticagrelor are indeed superior to clopidogrel. We used the ‘Platelet Inhibition Registry in ACS EvalUation Study’ (PIRAEUS) platform which was initiated in 2014 to integrate data generated by individual European registries on acute coronary syndromes (ACS) to gain a comprehensive overview on the effectiveness and safety of the P2Y12 receptor inhibitors.4 The participating registries have been described in narrative and tabular form in detail in an earlier review of this group,4 and the non-ST-elevation myocardial infarction (NSTEMI)5 and ST-segment elevation myocardial infarction (STEMI) cohorts.6
This study presents data on STEMI patients from five European registries, with a focus on effectiveness (all-cause deaths) and safety (bleeding) for P2Y12 receptor inhibitor-based DAPT specifically. In the assessment of effectiveness and safety endpoints, to account for differences in patient composition, a matched pair strategy and propensity scoring was used.
Methods
The original PIRAEUS data set consists of 12 ACS registries and studies which are heterogenous in terms of setting, duration, documented variables, and endpoints.4 The criteria to select suitable registries were as follows: European multicentre or single-centre observational studies on real-life experience in the management of ACS within the last 5 years; large unselected STEMI patient cohorts; data on percutaneous coronary intervention (PCI); data on management during initial hospitalization for ACS available; information on specific P2Y12 inhibitor treatment pre-cath or in-cath lab; follow-up data on outcomes (death, cardiac events, and bleedings) available; and willingness of registry owners to share data. Five registries met the criteria: AMIS Plus, Belgium STEMI, ALKK-PCI, APCI/ADAPT, and EYESHOT. A description of these registries that provided STEMI data has been provided previously.6
All participating registries were national projects. Two were a running registry with no specified stop date for inclusion of patients (AMIS Plus and Belgium STEMI database), while the others included cohorts of patients within a defined time frame.
Registry owners were asked to provide detailed current data on STEMI patients treated with the P2Y12 receptor inhibitors prasugrel, ticagrelor, or clopidogrel. While the registries collected individual patient data prospectively, for the purpose of this analysis, aggregate data in tabular format were used. Endpoints of interest comprised in-hospital events, such as all-cause death, cardiovascular (CV) death, stroke, recurrent myocardial infarction (MI), and repeat percutaneous coronary intervention (PCI) for effectiveness, and life-threatening/major and minor bleeding. For bleeding events, the definition (e.g. BARC) was requested from the registry owners but was not always available or sometimes changed during the registry data collection. Registry owners were asked to provide percentages for the various events together with event number and patient number at the various time points.
Statistical analysis
Patients who received either new P2Y12 receptor inhibitors (prasugrel and ticagrelor combined) or clopidogrel during pre-cath or in-cath treatment were included in the analysis. Patients administered more than one kind of P2Y12 receptor inhibitors during pre-cath or in-cath treatment were excluded from the event analysis.
Propensity score matching7 of new P2Y12 receptor inhibitors and clopidogrel was based on one-to-one matching without replacement stratified by registry. The propensity score was estimated using logistic regression on a study-by-study basis including the following 24 baseline covariates: age, gender, diabetes mellitus, heart failure, atrial fibrillation, prior MI, previous stroke/transient ischaemic attack (TIA), previous coronary artery bypass grafting (CABG), arterial hypertension, peripheral arterial disease, hypercholesterolaemia, current smoking, chronic kidney disease, Killip class >2, chronic ASA, chronic P2Y12 receptor inhibitors, chronic oral anticoagulation, multivessel disease, >1 stent implanted, and pre-cath or in-cath lab treatments [glycoprotein IIb/IIIa inhibitors (GPI), unfractionated heparin, low molecular weight heparin (LMWH), and bivalirudin]. The estimated propensity score was the predicted probability of treated new P2Y12 receptor inhibitor from the logistic regression model. A greedy, nearest-neighbour matching algorithm was employed to match patients on the logit of the propensity score using calipers of width equal to 0.1 of the propensity score logit. The balance in baseline characteristics between matched patients was assessed by standardized differences.8 Odds ratio (OR) and 95% confidence interval (CI) of new P2Y12 receptor inhibitors group compared with clopidogrel group on in-hospital events was estimated using the conditional logistic regression model stratified by matched pairs (Analysis 1). In-hospital death, CV death, stroke/TIA, and any major bleeding [intracranial haemorrhage (ICH), fatal bleeding, or Thrombolysis in Myocardial Infarction (TIMI) major bleeding] were used as the outcome variables in the event analysis. Propensity score-matched analyses to compare between prasugrel and ticagrelor were done by using similar models (Analysis 2).
Means and standard deviation were calculated for continuous variables, and percentages were calculated for categorical variables based on the total number of patients excluding missing and unknown data. Statistical significance level was considered as two-sided P-value of 5%. All statistical analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC, USA) software.
Results
Patient disposition
We collected data on a total of 36 553 STEMI patients. The sample size between registries varied considerably (from n = 1066 in EYESHOT to n = 13 916 in AAPCI/ADAPT). Of these, a total of 25 250 patients met the inclusion criteria for in-hospital event analysis (see Methods section) before matching patients. No patient in the BELGIUM STEMI registry met the inclusion criteria, because antiplatelet therapies during pre-cath or in-cath treatment were not recorded in the registry (only antiplatelet therapy at discharge). After matching patients, a total of 15 120 patients (Analysis 1) and 8910 (Analysis 2) were included in the in-hospital event analysis (Figure 1).
Figure 1.
Patient disposition.
Risk factors and comorbidities by registry are shown in Table 1. In all registries, the proportion of males (range 72.0–76.0%) was higher than of females. Mean age ranged between 62.5 and 66.0 years. Comorbidities were frequent, in particular, arterial hypertension (range 39.5–70.9%), hypercholesterolaemia (range 31.0–58.9%), and diabetes mellitus (range 16.4–22.4%). Prior MI was reported in 10.1–20.1% and current smoking in 38.5–47.9%.
Table 1.
Baseline characteristics: risk factors and comorbidities
| Registry acronym | AAPCI/ADAPT | ALKK | AMIS Plus | BELGIUM | EYESHOT |
|---|---|---|---|---|---|
| Patient number, n | 13 916 | 3553 | 10 224 | 7794 | 1066 |
| Year of enrollment, n(%) | 13 916 | 3553 | 10 224 | 7794 | 1066 |
| Before 2009 | 1640 (11.8) | 104 (2.9) | 0 (0.0) | 0 (0.0) | 0 (0.0) |
| 2010 | 1359 (9.8) | 1061 (29.9) | 0 (0.0) | 0 (0.0) | 0 (0.0) |
| 2011 | 1889 (13.6) | 1284 (36.1) | 1461 (14.3) | 0 (0.0) | 0 (0.0) |
| 2012 | 1822 (13.1) | 1104 (31.1) | 1786 (17.5) | 0 (0.0) | 0 (0.0) |
| 2013 | 1770 (12.7) | 0 (0.0) | 1966 (19.2) | 0 (0.0) | 844 (79.2) |
| 2014 | 1392 (10.0) | 0 (0.0) | 1837 (18.0) | 1133 (14.5) | 222 (20.8) |
| 2015 | 1393 (10.0) | 0 (0.0) | 1499 (14.7) | 2685 (34.4) | 0 (0.0) |
| 2016 | 1137 (8.2) | 0 (0.0) | 1202 (11.8) | 2482 (31.8) | 0 (0.0) |
| After 2017 | 1514 (10.9) | 0 (0.0) | 473 (4.6) | 1494 (19.2) | 0 (0.0) |
| Age (years), mean (SD) | 62.5 (13.1) | 62.7 (13.0) | 64.2 (12.8) | 63.3 (13.1) | 66.0 (13.1) |
| Age >75 years, n(%) | 2860/13 916 (20.6) | 754/3553 (21.2) | 2352/10 224 (23.0) | 1718/7794 (22.0) | 308/1066 (28.9) |
| Risk factors and comorbidities, n(%) | |||||
| Males | 10 177/13 916 (73.1) | 2641/3553 (74.3) | 7770/10 224 (76.0) | 5891/7766 (75.9) | 767/1066 (72.0) |
| Diabetes mellitus | 2065/11 875 (17.4) | 746/3330 (22.4) | 1659/9799 (16.9) | 1269/7717 (16.4) | 223/1041 (21.4) |
| Heart failure | 284/2108 (13.5) | 387/3521 (11.0) | 147/8880 (1.7) | NA | 13/1064 (1.2) |
| Atrial fibrillation | 971/10 949 (8.9) | 77/1048 (7.3) | 343/9077 (3.8) | NA | NA |
| Prior myocardial infarction | 1294/12 766 (10.1) | 688/3422 (20.1) | 1220/9998 (12.2) | 1200/7714 (15.6) | 121/1058 (11.4) |
| Previous stroke/TIA | 394/12 155 (3.2) | 140/3290 (4.3) | 386/10 027 (3.8) | NA | 56/1056 (5.3) |
| Previous PCI | 1642/12 956 (12.7) | 533/3457 (15.4) | 1328/10 000 (13.3) | NA | 107/1063 (10.1) |
| Previous CABG | NA | 125/3508 (3.6) | 281/10 000 (2.8) | NA | 24/1064 (2.3) |
| Arterial hypertension | 1395/3528 (39.5) | 2232/3150 (70.9) | 5407/9750 (55.5) | 3673/7719 (47.6) | 620/1029 (60.3) |
| Peripheral arterial disease | 122/3528 (3.5) | 159/3252 (4.9) | 380/10 027 (3.8) | 552/7710 (7.2) | 93/1019 (9.1) |
| Hypercholesterolaemia | 1093/3528 (31.0) | 1283/2768 (46.4) | 5453/9261 (58.9) | 3813/7160 (53.3) | 381/921 (41.4) |
| Current smoking | 5461/12 467 (43.8) | 1457/3040 (47.9) | 4124/9171 (45.0) | 2974/7169 (41.5) | 405/1053 (38.5) |
| Chronic kidney disease | 2050/11 834 (17.3) | 373/3285 (11.4) | 455/10 028 (4.5) | 607/7158 (8.5) | 90/1036 (8.7) |
Percentages are based on the total number of patients excluding missing and unknown data. For BELGIUM, prior MI includes prior PCI and CABG. CKD is defined as either serum creatinine >1.5 mg/dL or eGFR <60/mL/m². For AAPCI-ADAPT, prior PCI includes prior CABG. CABG, coronary artery bypass grafting; NA, not applicable; TIA, transient ischaemic attack.
ST-elevation MI characteristics and pre-treatment are shown in Table 2. The STEMI location was slightly more often in the inferior artery (range 49.8–57.4%) compared with the anterior (41.6–44.3%). Killip Class I prevailed (66.4–85.7%). Chronic pre-treatment was reported for ASA in 20.0–28.7%, clopidogrel in 3.1–10.9%, prasugrel in 0.6–4.4%, and P2Y12 overall in 3.8–15.4%. Chronic pre-treatment with anticoagulation was noted in 3.8–11.6%.
Table 2.
Information on STEMI at admission and pre-treatment
| Registry acronym | AAPCI/ADAPT | ALKK | AMIS Plus | BELGIUM | EYESHOT |
|---|---|---|---|---|---|
| Patient number, n | 13 916 | 3553 | 10 224 | 7794 | 1066 |
| Admission | |||||
| Rhythm at ECG, n(%) | 12 318 | 9077 | . | 1053 | |
| Sinus rhythm | 11 098 (90.1) | NA | 8310 (91.6) | NA | 980 (93.1) |
| Atrial fibrillation | 945 (7.7) | NA | 343 (3.8) | NA | 56 (5.3) |
| Others | 275 (2.2) | NA | 424 (4.7) | NA | 17 (1.6) |
| STEMI location, n(%) | 13 914 | . | 10 178 | 7747 | 1066 |
| Inferior | 6935 (49.8) | NA | 5291 (52.0) | 4449 (57.4) | 584 (54.8) |
| Anterior | 6166 (44.3) | NA | 4288 (42.1) | 3224 (41.6) | 467 (43.8) |
| Undetermined | 813 (5.8) | NA | 599 (5.9) | 74 (1.0) | 15 (1.4) |
| Killip class, n(%) | 10 801 | . | 10 164 | 7681 | 1031 |
| I | 7176 (66.4) | NA | 8706 (85.7) | 6246 (81.3) | 811 (78.7) |
| II | 2123 (19.7) | NA | 722 (7.1) | 662 (8.6) | 151 (14.6) |
| III | 616 (5.7) | NA | 238 (2.3) | 214 (2.8) | 31 (3.0) |
| IV | 886 (8.2) | NA | 498 (4.9) | 559 (7.3) | 38 (3.7) |
| Killip class >2, n(%) | 1502/10 801 (13.9) | NA | 736/10 164 (7.2) | 773/7681 (10.1) | 69/1031 (6.7) |
| Pre-treatment, n(%) | |||||
| Chronic ASA | 579/2893 (20.0) | 520/2276 (22.8) | 2466/8595 (28.7) | NA | 243/1048 (23.2) |
| Chronic clopidogrel | 88/2864 (3.1) | 389/3553 (10.9) | 359/8462 (4.2) | NA | 35/1055 (3.3) |
| Chronic prasugrel | 17/2899 (0.6) | 156/3553 (4.4) | 77/8423 (0.9) | NA | 4/1056 (0.4) |
| Chronic ticagrelor | 10/2815 (0.4) | 1/1069 (0.1) | 64/6972 (0.9) | NA | 8/1057 (0.8) |
| Chronic P2Y12 receptor inhibitor | 115/3026 (3.8) | 546/3553 (15.4) | 499/8489 (5.9) | NA | 47/1057 (4.4) |
| Chronic oral anticoagulation | 110/2862 (3.8) | 46/1104 (4.2) | 1015/8719 (11.6) | NA | 33/1057 (3.1) |
| Vital sign | |||||
| SBP | |||||
| n | 10 626 | 9864 | 4090 | 1066 | |
| Mean (SD) | 133.0 (29.2) | NA | 131.8 (27.6) | 131.0 (30.7) | 130.6 (25.2) |
| DBP | |||||
| n | 10 386 | 9851 | 4110 | 1066 | |
| Mean (SD) | 78.4 (16.5) | NA | 78.3 (17.0) | 78.1 (20.7) | 76.5 (14.8) |
| HR | |||||
| n | 10 584 | 9825 | 1066 | ||
| Mean (SD) | 78.5 (19.3) | NA | 78.5 (18.4) | NA | 79.1 (19.0) |
Percentages are based on the total number of patients excluding missing and unknown data. For vital signs, extreme data with <10 or >300 were excluded. For STEMI location, lateral, inferolateral, and posterior are merged into ‘inferior’. Antero-lateral is merged into ‘anterior’.
Information on the timing and in-hospital management are given in Supplementary material online, Table S1. The timing from first medical contact to PCI was 79–101 min. Coronary angiography was performed in 94.0–100.0%.
Pre-cath and in-cath lab treatment are presented in Supplementary material online, Table S2. The information on pre-cath lab treatment was available for AAPCI/ADPAT, AMIS Plus, and EYESHOT. In these registries, ASA was given in 77.0–96.6%, clopidogrel in 28.8–46.7%, prasugrel in 18.0–21.9%, and ticagrelor in 16.7–24.5%. In-cath lab treatment was available for AAPCI/ADAPT, ALKK, and EYESHOT. Acetylic salicylic acid was routinely used in ALKK (93.0%), and rarely in EYESHOT (13.0%) and AAPCI/ADAPT (0.1%). Clopidogrel use ranged between 10.8% and 60.2%, prasugrel between 8.8% and 43.5%, and ticagrelor between 2.1% and 20.6%.
Table 3 displays discharge treatments, which were based on ASA (96.9– 99.3%), and also included clopidogrel (10.7–48.1%), prasugrel (24.4–31.3%), and ticagrelor (20.6–64.9%).
Table 3.
Discharge treatment and events in-hospital and within 30 days after discharge in propensity-matched populations
| Registry acronym | AAPCI/ADAPT | ALKK | AMIS Plus | BELGIUM | EYESHOT |
|---|---|---|---|---|---|
| Patient number, n | 13 916 | 3553 | 10 224 | 7794 | 1066 |
| Discharge treatment, n (%) | |||||
| ASA | 13 347/13 437 (99.3) | NA | 9629/9801 (98.2) | NA | 993/1025 (96.9) |
| Clopidogrel | 6102/12 680 (48.1) | NA | 2424/9770 (24.8) | 654/6114 (10.7) | 350/1025 (34.1) |
| Prasugrel | 3969/12 679 (31.3) | NA | 4268/9767 (43.7) | 1489/6114 (24.4) | 267/1025 (26.0) |
| Ticagrelor | 2608/12 665 (20.6) | NA | 2797/8691 (32.2) | 3971/6114 (64.9) | 353/1025 (34.4) |
| Oral anticoagulation | 214/2714 (7.9) | NA | 650/9720 (6.7) | NA | 48/1025 (4.7) |
| In-hospital events, n (%) | |||||
| Death | 805/13 916 (5.8) | 124/3553 (3.5) | 416/10 224 (4.1) | 531/7613 (7.0) | 41/1066 (3.8) |
| Cardiovascular death | 111/3528 (3.1) | NA | 195/10 204 (1.9) | NA | 37/1066 (3.5) |
| Stroke/TIA | 74/13 848 (0.5) | 11/3553 (0.3) | 70/10 224 (0.7) | NA | 9/1066 (0.8) |
| ICH | 31/13 916 (0.2) | NA | 2/358 (0.6) | NA | 0/1066 (0.0) |
| Fatal bleeding | NA | NA | 10/9850 (0.1) | NA | 1/1066 (0.1) |
| TIMI major bleeding | 131/13 916 (0.9) | NA | NA | NA | 14/1066 (1.3) |
| Any major bleeding | 131/13 916 (0.9) | NA | 12/9850 (0.1) | NA | 14/1066 (1.3) |
| TIMI minor bleeding | 40/3528 (1.1) | NA | NA | NA | 24/1066 (2.3) |
| Urgent revascularization | 28/3463 (0.8) | NA | NA | NA | 12/1066 (1.1) |
| Events within 30 days after discharge, n (%) | |||||
| Death | NA | NA | NA | 196/3264 (6.0) | NA |
| Cardiovascular death | NA | NA | NA | NA | NA |
| Stroke/TIA | NA | NA | NA | NA | NA |
| ICH | NA | NA | NA | NA | NA |
| Fatal bleeding | NA | NA | NA | NA | NA |
| TIMI major bleeding | NA | NA | NA | NA | NA |
| Any major bleeding | NA | NA | NA | NA | NA |
| TIMI minor bleeding | NA | NA | NA | NA | NA |
| Urgent revascularization | NA | NA | NA | NA | NA |
ICH: intracranial haemorrhage.
Percentages are based on the total number of patients excluding missing and unknown data. Other events within 30 days after discharge except for death are not available in all registries. Any major bleeding includes ICH, fatal bleeding, and TIMI major bleeding.
Data on in-hospital events are presented in Table 3 bottom. Data on death were available in all registries and ranged from 3.5% to 7%, data on CV death from AAPCI/ADAPT, AMIS Plus, and EYESHOT ranging from 1.9% to 3.5%. Stroke/TIA was reported in 0.3–0.8%. Information on bleeding, depending on the definition, was heterogeneous: fatal bleeding as reported in AMIS Plus and EYESHOT was 0.1% each, TIMI major bleeding in AAPCI/ADPAT and EYESHOT was 0.9% and 1.3%. Urgent revascularization in AAPCI/ADPAT and EYESHOT was 0.8% and 1.1%, respectively.
Event analysis before matching patients
Baseline patient characteristics in all registries before matching patients were widely different between new P2Y12 receptor inhibitor (prasugrel and ticagrelor combined) and clopidogrel. After matching patients, these imbalances were largely improved and the standardized differences in most of the patient characteristics in each registry were within 10% (Supplementary material online, Tables S1-1–S1-4 and S2-1–S2-4). It was not possible to match patients in ALKK-PCI for Analysis 2 due to small number of patients who received ticagrelor.
Before matching patients, the overall cumulative event rates of new P2Y12 receptor inhibitor, prasugrel, ticagrelor, and clopidogrel on in-hospital events were 3.2%, 2.6%, 3.9%, and 6.3% on death, 1.5%, 1.3%, 1.7%, and 4.5% on CV death, 0.4%, 0.3%, 0.6%, and 0.8% on stroke/TIA, and 0.4%, 0.4%, 0.4%, and 0.8% on any major bleedings, respectively.
Event analysis after matching procedure
Figure 2presents in-hospital events, by single events, differentiated by new P2Y12 receptor inhibitor vs. clopidogrel as administered as pre-cath or in-cath lab treatment with matched population. As a consistent finding, the event rates appeared substantially lower with the new P2Y12 receptor inhibitors (with an exception for any major bleeding with low event rates). The ORs of the new P2Y12 receptor inhibitors for the death of any cause, CV death, stroke/TIA, and any major bleeding when compared with clopidogrel were 0.72 (95% CI 0.62–0.84, P < 0.001), 0.70 (95% CI 0.52–0.96, P = 0.026), 0.52 (95% CI 0.33–0.80, P = 0.003), and 0.81 (95% CI 0.53–1.24, P = 0.335), respectively.
Figure 2.
Relationship between the type of P2Y12 treatment and events in-hospital in propensity-matched populations (Analysis 1).
The standardized differences for all covariates in the matched patients were <10% after integrating registries (Supplementary material online, Tables S1-5 and S2-5).
Upon further differentiation, event rates of prasugrel and ticagrelor appeared similar across registries (Analysis 2; Figure 3). In the comparison between prasugrel and ticagrelor, there were no significant differences on any event.
Figure 3.
Relationship between the type of P2Y12 treatment and events in-hospital in propensity-matched populations (Analysis 2).
Discussion
The present analysis of contemporary European real-world registries suggests that in STEMI patients the newer P2Y12 receptor inhibitors compared with clopidogrel have more favourable in-hospital outcomes, while not being inferior in terms of safety.
With respect to the setting, the majority of registries, including ALKK and APCI focused on selected (tertiary) hospitals, while others were open to all types of hospitals (AMIS Plus). Thus, the considerable differences in outcomes between the registries are likely explained by their designs. Overall across all analysed registries, however, in-hospital mortality rates associated with STEMI were quite low and homogeneous. Reasons that might account for this finding include selection bias (fewer ill patients might have been preferentially recruited) and improved ACS management over the years.9 Patients in the five registries were similar in age and gender distribution. However, they differed substantially in the prevalence of comorbidities, such as arterial hypertension, diabetes mellitus, or prior MI.
Death rates, ischaemic events, and bleeding rates were overall lower than those reported in Phase III studies with P2Y12 inhibitors. Regarding individual P2Y12 inhibitors, patients on prasugrel, and, to a lesser degree, ticagrelor, had fewer ischaemic and bleeding events at all time points than clopidogrel-treated patients. These findings are partly related to the fact that the newer agents were used in younger and less ill patients.
A thorough comparison of various P2Y12 inhibitors can be only be made in the presence of appropriate adjustment techniques reducing indication biases that registries are prone to. In line with its labelling, the registries consistently documented that prasugrel in everyday practice conditions is used in younger patients, compared with other antiplatelet agents as recommended by the ESC Guidelines.1,2 This age difference reflects the somewhat restricted labelling, as prasugrel is contraindicated in patients with prior TIA or stroke, and the drug is generally not recommended in elderly patients (≥75 years), although maintenance dose adaptation (5 mg instead of 10 mg) may be considered in such patients.10 According to the product labelling, ticagrelor should be used with caution in patients with a history of asthma and/or chronic obstructive pulmonary disease (due to a relatively high incidence of dyspnoea) and also in patients with renal impairment (due to creatinine level increases).11 These side effects have not been systematically assessed in the registries contributing to PIRAEUS.
Trials comparing clopidogrel with either prasugrel or ticagrelor found a superior effectiveness of the novel antiplatelet agents,12,13 even in STEMI patients.14,15
Prior studies have compared prasugrel or ticagrelor with clopidogrel also in a real-world setting, but a head-to-head comparison between the three available antiplatelet agents has never been performed. An AMIS Plus analysis using a propensity score-matched pairs analysis found significantly lower in-hospital mortality with prasugrel (1.8%) vs. clopidogrel (3.1%).16 In MULTIPRAC, a multinational, multicentre, prospective registry enrolling 2053 STEMI patients, after adjustment for differences in baseline characteristics (including a 10-year age difference), treatment with prasugrel was associated with a significantly lower risk of CV death than treatment with clopidogrel (OR 0.248, 95% CI 0.06–0.89; data submitted).17 In the SCAAR registry, the age difference was 2 years only, but the difference in 30-day all-cause mortality was substantial (prasugrel 2.5% vs. clopidogrel 5.0%).18 Accordingly, a prospective cohort study of 45 073 ACS patients enrolled into the SWEDEHEART registry confirmed the results of the PLATO trial as ticagrelor was associated with a lower risk of death, MI, or stroke compared with clopidogrel [11.7 vs. 22.3%; adjusted hazard ratio (HR) 0.85, 95% CI: 0.78–0.93], as well as death alone (5.8 vs. 12.9%; adjusted HR 0.83 95% CI: 0.75–0.92).19 Re-admission for bleeding occurred more frequently with ticagrelor vs. clopidogrel (5.5 vs. 5.2%; adjusted HR 1.20, 95% CI 1.04–1.40).19
Recently, ticagrelor and prasugrel have been directly compared in the multicentre, randomized, open-label ISAR-REACT 5 trial.3 Among the 4018 patients with ACS enrolled in this trial, a primary endpoint event—death from any cause, MI, or stroke at 1 year after randomization—occurred in 9.1% of patients in the ticagrelor group and 6.8% in the prasugrel group (HR 1.36, 95% CI: 1.09–1.70; P = 0.006), without any difference in terms of major bleeding events (BARC Type 3 through 5). Among the pre-specified subgroup of patients with STEMI no difference was observed in terms of primary efficacy (10.1 vs. 7.9; HR 1.31, 95% CI 0.94–1.81) and safety endpoint across all enrolled patients, including those with STEMI.3 These findings are in accordance with our real-world data showing a numerical, not significant reduction in the ischaemic event rate associated with prasugrel compared with ticagrelor, with a similar occurrence of bleeding events during the hospitalization for STEMI.
Limitations
Not all of the previously identified suitable registries6 provided data in the agreed structured format and several registries could, therefore, not be analysed for the purpose of this article. Outcomes analyses were restricted to the short in-hospital period and does not allow for long-term assessment. The concurrent use of oral anticoagulation and other anticoagulants, such as unfractionated heparin, low-molecular-weight heparins, and fondaparinux, needs to be taken into account. Bleeding events were categorized using various definitions and were not standardized across registries. The low rate of bleeding complications is noteworthy, as real-life data are usually more likely to document increased safety hazards, compared with randomized controlled trials.
In conclusion, the present analysis of contemporary European registries suggests that in clinical practice STEMI patients the newer P2Y12 receptor inhibitors compared with clopidogrel have favourable in-hospital outcomes while not being inferior in terms of safety.
Supplementary material
Supplementary material is available at European Heart Journal – Cardiovascular Pharmacotherapy online.
Acknowledgement
Statistical analyses were performed by Dr. Yasuyuki Matsushita, Daiichi Sankyo Europe, Biostatistics and Data Management, Munich, Germany. Meetings of the PIRAEUS group and editorial/medical writing support of the present article by 3P Consulting, Seefeld, Germany, were funded by Daiichi Sankyo GmbH Europe.
Funding
This analysis was funded by Daiichi-Sankyo Europe GmbH, Germany.
Conflict of interest: L.D.L. has received honoraria for advisory boards or as speaker/chairman at scientific congresses from Amgen, Aspen, Astra-Zeneca, Bayer, Boehringer Ingelheim, Chiesi, Daiichi Sankyo, Eli Lilly, Menarini, Pfizer/BMS, Servier and The Medicines Company. J.W.J. has received research grants from and/or was speaker (with or without lecture fees) on (CME accredited) meetings sponsored by Amgen, Astellas, Anthera, Astra-Zeneca, Bayer, Biotronik, Boston Scientific,Correvio, Daiichi Sankyo, Lilly, Genzyme, Medtronic, Merck-Schering-Plough, Pfizer, Orbus Neich, Novartis, Roche, Servier, Sanofi Aventis, The Medicine Company, the Netherlands Heart Foundation, the Interuniversity Cardiology Institute of the Netherlands and the European Community Framework KP7 Programme. M.J.C. has received honoraria for advisory boards or as speaker/chairman at scientific congresses from AstraZeneca, Bayer, Boehringer Ingelheim, Daiichi Sankyo, Eli Lilly, Sanofi and The Medicines Company. J.D. has received consulting fees or fees as speaker from AstraZeneca, Daiichi Sankyo, Eli Lilly, Merck, Server. T.F.L. reports research grants to the institution from AstraZeneca, Bayer, Biosensors, Biotronik, Boston Scientific, Medtronic, MSD, Roche and Servier, including lecture fees. C.M.M. reports research grants to the institution from Eli Lilly, AstraZeneca, Roche, MSD, Medtronic, St. Jude Medical, Sanofi, Pfizer; lecture fees from Eli Lilly, Daiichi-Sankyo, AstraZeneca, Roche, MSD, Amgen, Novartis. F.W. reports speaker honoraria and consultancy fees from Astra Zeneca, Lilly, Daiichi Sankyo, BMS, Pfizer. U.Z. reports personal fees from Astra Zeneca, during the conduct of the study; personal fees from Astra Zeneca, grants and personal fees from Daiichi Sankyo, grants and personal fees from Eli Lilly, personal fees from Bayer Healthcare, personal fees from The Medicines Company, grants and personal fees from Sanofi, grants and personal fees from Novartis, personal fees from Boehringer Ingelheim, personal fees from MSD, outside the submitted work. P.E. and D.R. report no conflict of interest.
Supplementary Material
References
- 1. Steg PG, James SK, Atar D, Badano LP, Blomstrom-Lundqvist C, Borger MA, Di Mario C, Dickstein K, Ducrocq G, Fernandez-Aviles F, Gershlick AH, Giannuzzi P, Halvorsen S, Huber K, Juni P, Kastrati A, Knuuti J, Lenzen MJ, Mahaffey KW, Valgimigli M, van ‘t Hof A, Widimsky P, Zahger D. ESC Guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation. Eur Heart J 2012;33:2569–2619. [DOI] [PubMed] [Google Scholar]
- 2. Ibanez B, James S, Agewall S, Antunes MJ, Bucciarelli-Ducci C, Bueno H, Caforio ALP, Crea F, Goudevenos JA, Halvorsen S, Hindricks G, Kastrati A, Lenzen MJ, Prescott E, Roffi M, Valgimigli M, Varenhorst C, Vranckx P, Widimský P. 2017 ESC Guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation: the Task Force for the management of acute myocardial infarction in patients presenting with ST-segment elevation of the European Society of Cardiology (ESC). Eur Heart J 2018;39:119–177. [DOI] [PubMed] [Google Scholar]
- 3. Schüpke S, Neumann FJ, Menichelli M, Mayer K, Bernlochner I, Wöhrle J, Richardt G, Liebetrau C, Witzenbichler B, Antoniucci D, Akin I, Bott-Flügel L, Fischer M, Landmesser U, Katus HA, Sibbing D, Seyfarth M, Janisch M, Boncompagni D, Hilz R, Rottbauer W, Okrojek R, Möllmann H, Hochholzer W, Migliorini A, Cassese S, Mollo P, Xhepa E, Kufner S, Strehle A, Leggewie S, Allali A, Ndrepepa G, Schühlen H, Angiolillo DJ, Hamm CW, Hapfelmeier A, Tölg R, Trenk D, Schunkert H, Laugwitz KL, Kastrati A. Ticagrelor or prasugrel in patients with acute coronary syndromes. N Engl J Med 2019;381:1524–1534. [DOI] [PubMed] [Google Scholar]
- 4. Jukema JW, Lettino M, Widimský P, Danchin N, Bardaji A, Barrabes JA, Cequier A, Claeys MJ, De Luca L, Dörler J, Erlinge D, Erne P, Goldstein P, Koul SM, Lemesle G, Lüscher TF, Matter CM, Montalescot G, Radovanovic D, Lopez-Sendón J, Tousek P, Weidinger F, Weston CF, Zaman A, Zeymer U. Contemporary registries on P2Y12 inhibitors in patients with acute coronary syndromes in Europe: overview and methodological considerations. Eur Heart J Cardiovasc Pharmacother 2015;1:232–244. [DOI] [PubMed] [Google Scholar]
- 5. Zeymer U, Widimsky P, Danchin N, Lettino M, Bardaji A, Barrabes JA, Cequier A, Claeys MJ, De Luca L, Dörler J, Erlinge D, Erne P, Goldstein P, Koul SM, Lemesle G, Lüscher TF, Matter CM, Montalescot G, Radovanovic D, Sendón JL, Tousek P, Weidinger F, Weston CF, Zaman A, Andell P, Li J, Jukema JW. P2Y12 receptor inhibitors in patients with non-ST-elevation acute coronary syndrome in the real world: use, patient selection, and outcomes from contemporary European registries. Eur Heart J Cardiovasc Pharmacother 2016;2:229–243. [DOI] [PubMed] [Google Scholar]
- 6. Danchin NL, Maddalena Zeymer U, Widimský P, Danchin N, Bardaji A, Barrabes JA, Cequier A, Claeys MJ, De Luca L, Dörler J, Erlinge D, Erne P, Goldstein P, Koul SM, Lemesle G, Lüscher TF, Matter CM, Montalescot G, Radovanovic D, Lopez Sendón J, Tousek P, Weidinger F, Weston CF, Zaman A, Andell P, Li J, Jukema JW. Use, patient selection and outcomes of P2Y12 receptor inhibitor treatment in patients with STEMI based on contemporary European registries. Eur Heart J Cardiovasc Pharmacother 2016;2:152–167. [DOI] [PubMed] [Google Scholar]
- 7. Rosenbaum P, Rubin D. The central role of the propensity score in observational studies for causal effects. Biometrika 1983;70:41–55. [Google Scholar]
- 8. Austin PC. Some methods of propensity-score matching had superior performance to others: results of an empirical investigation and Monte Carlo simulations. Biom J 2009;51:171–184. [DOI] [PubMed] [Google Scholar]
- 9. Bjorklund E, Nielsen SJ, Hansson EC, Karlsson M, Wallinder A, Martinsson A, Tygesen H, Romlin BS, Malm CJ, Pivodic A, Jeppsson A. Secondary prevention medications after coronary artery bypass grafting and long-term survival: a population-based longitudinal study from the SWEDEHEART registry. Eur Heart J 2019; pii: ehz714. doi: 10.1093/eurheartj/ehz714. [Epub ahead of print]. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Efient(R) Summary of Product Characteristics. http://www.ema.europa.eu (20 January 2020).
- 11.European Medicines Agency (EMA). Brillque(R). Ticagrelor Summary of Product Characteristics and Package Leaflet. http://www.emea.europa.eu. (20 January 2020).
- 12. Wiviott SD, Braunwald E, McCabe CH, Montalescot G, Ruzyllo W, Gottlieb S, Neumann F-J, Ardissino D, De Servi S, Murphy SA, Riesmeyer J, Weerakkody G, Gibson CM, Antman EM, TRITON-TIMI 38 Investigators. Prasugrel versus clopidogrel in patients with acute coronary syndromes. N Engl J Med 2007;357:2001–2015. [DOI] [PubMed] [Google Scholar]
- 13. Wallentin L, Becker RC, Budaj A, Cannon CP, Emanuelsson H, Held C, Horrow J, Husted S, James S, Katus H, Mahaffey KW, Scirica BM, Skene A, Steg PG, Storey RF, Harrington RA, Investigators P, Freij A, Thorsén M. Ticagrelor versus clopidogrel in patients with acute coronary syndromes. N Engl J Med 2009;361:1045–1057. [DOI] [PubMed] [Google Scholar]
- 14. Montalescot G, Wiviott SD, Braunwald E, Murphy SA, Gibson CM, McCabe CH, Antman EM; TRITON-TIMI 38 investigators. Prasugrel compared with clopidogrel in patients undergoing percutaneous coronary intervention for ST-elevation myocardial infarction (TRITON-TIMI 38): double-blind, randomised controlled trial. Lancet 2009;373:723–731. [DOI] [PubMed] [Google Scholar]
- 15. Steg PG, James S, Harrington RA, Ardissino D, Becker RC, Cannon CP, Emanuelsson H, Finkelstein A, Husted S, Katus H, Kilhamn J, Olofsson S, Storey RF, Weaver WD, Wallentin L; PLATO Study Group. Ticagrelor versus clopidogrel in patients with ST-elevation acute coronary syndromes intended for reperfusion with primary percutaneous coronary intervention: a Platelet Inhibition and Patient Outcomes (PLATO) trial subgroup analysis. Circulation 2010;122:2131–2141. [DOI] [PubMed] [Google Scholar]
- 16. Kurz DJ, Radovanovic D, Seifert B, Bernheim AM, Roffi M, Pedrazzini G, Windecker S, Erne P, Eberli FR. Comparison of prasugrel and clopidogrel-treated patients with acute coronary syndrome undergoing percutaneous coronary intervention: a propensity score-matched analysis of the Acute Myocardial Infarction in Switzerland (AMIS)-Plus Registry. Eur Heart J Acute Cardiovasc Care 2016;5:13–22. [DOI] [PubMed] [Google Scholar]
- 17. Clemmensen P, Grieco N, Ince H, Danchin N, Goedicke J, Ramos Y, Schmitt J, Goldstein P. MULTIPRAC study investigators. MULTInational non-interventional study of patients with ST-segment elevation myocardial infarction treated with PRimary Angioplasty and Concomitant use of upstream antiplatelet therapy with prasugrel or clopidogrel–the European MULTIPRAC Registry. Eur Heart J Acute Cardiovasc Care 2015;4:220–229. [DOI] [PubMed] [Google Scholar]
- 18. Damman P, Varenhorst C, Koul S, Eriksson P, Erlinge D, Lagerqvist B, James SK. Treatment patterns and outcomes in patients undergoing percutaneous coronary intervention treated with prasugrel or clopidogrel (from the Swedish Coronary Angiography and Angioplasty Registry [SCAAR]). Am J Cardiol 2014;113:64–69. [DOI] [PubMed] [Google Scholar]
- 19. Sahlen A, Varenhorst C, Lagerqvist B, Renlund H, Omerovic E, Erlinge D, Wallentin L, James SK, Jernberg T. Outcomes in patints treated with ticagrelor or clopidogrel after acute myocardial infarction: experiences from SWEDEHEART registry. Eur Heart J 2016;37:3335–3342. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.