Summary
Background
Little information is available on the long-term incidence of bleeding events after ST-segment elevation myocardial infarction (STEMI) with the current antithrombotic strategy.
Aims
To evaluate the effect of bleedings for up to 12 months on clinical events and therapeutic compliance in unselected STEMI patients treated with prasugrel or clopidogrel.
Methods
Patients were treated with clopidogrel or prasugrel according to guidelines. The primary endpoint was first occurrence of a bleeding event from hospital discharge to 12 months, assessed by the Bleeding Academic Research Consortium (BARC) classification using a dedicated questionnaire. Topography of bleedings, causes of premature cessation and ischaemic events were compared between clopidogrel- and prasugrel-treated patients.
Results
A total of 390 patients were enrolled (211 in the prasugrel group, 179 in the clopidogrel group). Elderly, female and low-body weight patients were more likely to receive clopidogrel. At 12 months, the incidence of major bleedings (BARC 3) was lower with prasugrel (1% vs 6%; P = 0.02), mainly due to fewer transfusions. Elderly age was a risk factor for severe bleeding. Premature treatment cessation was related to ischaemic complications ( P = 0.03), and occurred more frequently with prasugrel ( P = 0.001). One-year mortality was very low (1.9 per 100 person-years, 95% confidence interval 0.9–4.0), and was higher in the clopidogrel group ( P = 0.03).
Conclusions
In this unselected STEMI population, the rate of major bleedings with prasugrel at 12 months was low, but nuisance bleedings were frequent and led to more premature cessations than with clopidogrel. Prevention of bleeding complications, even minor, is necessary to prevent disruption of antithrombotic medication.
Résumé
Contexte
Très peu de données sont disponibles sur les complications hémorragiques à long terme après infarctus aigu du myocarde (IDM) avec les nouvelles stratégies antithrombotiques.
Objectifs
L’objectif principal était d’évaluer l’impact clinique des saignements à 12 mois et la compliance de patients non sélectionnés pris en charge pour un IDM et traités par prasugrel ou clopidogrel.
Méthodes
Les patients étaient traités selon les recommandations par du clopidogrel ou du prasugrel. Le critère principal de jugement était la survenue dans les 12 mois d’un saignement évalué par la classification BARC (Bleeding Academic Research Consortium) à l’aide d’un questionnaire dédié. La topographie des saignements, les causes d’un arrêt prématuré et la survenue d’événements ischémiques ont aussi été comparés pour les patients traités soit par clopidogrel, soit par prasugrel.
Résultats
Un total de 390 patients ont été inclus dont 211 sous prasugrel et 179 sous clopidogrel. Les patients âgés, de sexe féminin et de petit poids étaient plus fréquemment traités par clopidogrel. À 12 mois, l’incidence des saignements majeurs (BARC 3) était plus faible avec le prasugrel (1 % vs 6 % ; p = 0,02) tout particulièrement du fait de transfusions plus rares. L’âge était un facteur de risque de saignement majeur. L’arrêt prématuré de traitement était plus fréquent sous prasugrel ( p = 0,001) et était associé à des complications ischémiques plus nombreuses ( p = 0,03). La mortalité à 1 an restait très faible (1,9 par 100 personnes-années, intervalle de confiance à 95 % 0,9–4,0) mais plus importante dans le groupe clopidogrel ( p = 0,03).
Conclusion
Dans cette population non sélectionnée, le taux de saignements majeurs sous prasugrel était faible dans les 12 mois après un IDM. Les saignements minimes restaient cependant fréquents et associés à une plus mauvaise compliance que sous clopidogrel. La prévention des complications hémorragiques, même minimes, apparaît indispensable pour limiter le risque d’interruption des traitements antithrombotiques.
Background
Significant advances have been made recently in the care of patients with acute coronary syndromes, resulting in substantial improvements in clinical outcomes . The P 2 Y 12 inhibitor clopidogrel has been used widely over the past 10 years, because of the reported reduction in ischaemic events . However, clopidogrel is associated with limitations, including a possible increased risk of ischaemic events, such as stent thrombosis , a modest antiplatelet effect, a relatively slow onset of action , substantial interpatient variability in platelet inhibition and potential genetic resistance .
New P 2 Y 12 inhibitors such as prasugrel and, more recently, ticagrelor have been developed to overcome the limitations of clopidogrel, and these are now recommended as first-line therapy in acute coronary syndromes . Prasugrel induces more rapid, potent and predictable platelet inhibition compared with clopidogrel . In the Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition With Prasugrel-Thrombolysis in Myocardial Infarction 38 (TRITON-TIMI 38), prasugrel was associated with a greater reduction in ischaemic events, but with an excess of bleeding events versus clopidogrel in patients with acute coronary syndromes . In the subgroup admitted for ST-segment elevation myocardial infarction (STEMI) , a reduction in major adverse cardiac events was observed without an increase in major bleeding complications.
Little information is available regarding long-term follow-up of clinical events and therapeutic compliance in patients with STEMI treated with prasugrel. The aim of this study was to evaluate the 1-year incidence and clinical impact of bleedings in patients admitted for STEMI and treated with prasugrel or clopidogrel in the “real world”.
Methods
This study was a single-centre prospective cohort study of patients admitted for STEMI to Nîmes University Hospital, France.
All patients admitted to the intensive care unit for STEMI from 15 February 2010 to 30 August 2012, and who presented within 24 hours of symptom onset, were considered for inclusion. STEMI was defined according to recent guidelines . Reperfusion therapy, with percutaneous coronary intervention or thrombolysis, was performed and dual antiplatelet therapy (DAPT) was started. Patients on long-term oral anticoagulants, those who had had an initial cardiac arrest requiring mechanical ventilation, those with advanced cognitive or psychiatric disorders and those who lived abroad were excluded. All patients were given full study information and gave written informed consent. The protocol was approved by the local ethics committee and the institutional regulatory authorities, and was conducted according to the principals of the Declaration of Helsinki.
Patients were treated with clopidogrel or prasugrel (at the discretion of the physicians) and aspirin, in accordance with guidelines and the usual precautions for prasugrel use (i.e. clopidogrel is preferable for patients with a low-body weight or a history of stroke, elderly patients and those with a history of bleeding). Use of glycoprotein IIb/IIIa inhibitors, thromboaspiration and stenting (type and number during the coronary angiogram) was at the discretion of the physicians and in accordance with guidelines .
Treatment group (clopidogrel or prasugrel) was defined by the P 2 Y 12 inhibitor prescription at hospital discharge. Patients were followed up for 12 months after discharge by a trained research assistant supervised by the study investigator, and a dedicated telephone questionnaire was completed for each. Data on bleeding events (topography, severity, time to onset, effect on therapeutic strategy and compliance at 12 months), ischaemic adverse events and DAPT cessation were collected. In cases of bleeding or ischaemic adverse events notified by patients, relevant information was checked in their medical files. If patients stopped DAPT, information on cessation dates, possible restarting and the reasons for cessation and/or interruption was collected. Modes of DAPT cessation were defined according to the classification established by Mehran et al. : discontinuation was premature drug cessation based on physician recommendation; disruption corresponded to cessation of antiplatelet treatment due to bleeding or non-compliance; and interruption was defined as temporary cessation of antiplatelet treatment due to surgical necessity or invasive intervention with reinstitution of DAPT within 14 days. All data were collected using a standardized form and entered into an electronic database. In case of bleeding complications, therapeutic strategy (i.e. switching to another drug, discontinuation of treatment) was left at the discretion of practitioners, without a mandatory strategy. None of the patients received prasugrel 5 mg, as this dose was unavailable at the time of this study. The primary outcome was the first occurrence of a bleeding event between hospital discharge and 12 months. Bleeding was assessed by the Bleeding Academic Research Consortium (BARC) criteria .
Secondary outcomes were: the first occurrence of an ischaemic event using a composite of cardiac death, myocardial infarction, urgent target lesion revascularization and definite or probable stent thrombosis; and all-cause mortality. Stent thrombosis, urgent target lesion revascularization and death (classified as cardiac, vascular or non-cardiovascular) were defined according to the Academic Research Consortium criteria .
Patient characteristics were compared according to treatment group (prasugrel or clopidogrel) using the χ 2 or non-parametric Kruskall-Wallis tests, with Bonferroni correction when necessary. Bleeding topography and severity (BARC classification) were compared according to treatment group and study period (0–3, 3–12 and 0–12 months). Risk factors for bleeding after hospital discharge were modelled using survival analysis (Cox modelling) after checking the proportional hazards assumption. Bleeding events were coded as binary variables: BARC 0 or 1 versus BARC > 2–5 (i.e. moderate to severe bleeding); and BARC 0–2 versus BARC ≥ 3–5. Univariate analyses were performed with the potential variables of interest (age, weight, sex, chronic renal failure, diabetes, history of cancer, gastrointestinal bleeding, stroke and treatment group) and multivariable modelling was built with a backward stepwise procedure. Subsequently, any associations between bleeding events (BARC classification) and variables of interest (including treatment group) were sought by proportional logistic regression modelling (backward stepwise procedure) in two time periods (0–3 and 3–12 months).
Treatment duration and cessation and/or interruption were compared by treatment group and study period (0–3, 3–12 and 0–12 months) using the χ 2 test with Bonferroni correction. Associations between treatment cessation and ischaemic event occurrence (measured by a composite of mortality/stent thrombosis/urgent revascularization/ischaemic stroke) were analysed.
Mortality rates and risk factors for ischaemic events (composite outcome) were studied using Kaplan-Meier curves with the Log-rank test. All P -values were two-sided, with a 0.05 significance level. Analyses were performed using Stata 12/SE software (SAS Institute Inc., Cary, NC, USA).
Results
Among 436 patients admitted for STEMI, 46 were excluded, leaving 390 patients in the study population ( Fig. 1 ). Seven (2%) patients died before hospital discharge (five cardiogenic shocks and two mechanical complications) and eight (2%) were lost to follow-up. Thus, 375 patients were followed for 12 months (205 receiving prasugrel [54.7%] and 170 clopidogrel [45.3%]), including seven patients who died (two cardiogenic shocks, one mechanical complication, one other cardiological cause and three non-cardiovascular causes).
The patients’ baseline characteristics differed between the treatment groups, reflecting current guidelines ( Table 1 ). Patients with an increased risk of bleeding (i.e. elderly, female and low-body weight patients) were more frequently treated with clopidogrel.
| Overall population, ( n = 390) | Prasugrel group, ( n = 211) | Clopidogrel group, ( n = 179) | P a | |
|---|---|---|---|---|
| Women | 92 (24) | 18 (9) | 74 (41) | < 0.001 |
| Age (years) | 61 (52–73) | 57 (50–63) | 60 (50–73) | < 0.001 |
| Age > 75 years | 78 (20) | 3 (1) | 75 (42) | < 0.001 |
| Weight (kg) | 76 (66–85) | 80 (72–90) | 70 (60–81) | < 0.001 |
| Weight < 60 kg | 57 (15) | 11 (5) | 46 (26) | < 0.001 |
| BMI (kg/m 2 ) | 26.2 (23.6–29.1) | 27.0 (24.2–30.1) | 25.0 (22.1–28.1) | < 0.001 |
| BMI > 30 kg/m 2 | 75 (19) | 55 (26) | 20 (11) | < 0.001 |
| Hypertension | 172 (44) | 76 (36) | 96 (54) | < 0.001 |
| Diabetes | 74 (19) | 36 (17) | 38 (21) | 0.3 |
| Current smoking | 173 (44) | 124 (59) | 49 (27) | < 0.001 |
| Family history of CAD | 116 (30) | 71 (34) | 45 (25) | 0.07 |
| Dyslipidaemia | 166 (43) | 91 (43) | 75 (42) | 0.8 |
| Medical history | ||||
| CAD | 41 (11) | 23 (11) | 18 (10) | 0.8 |
| CABG | 1 (0) | 0 (0) | 1 (1) | 0.3 |
| Stroke | 13 (3) | 1 (0) | 12 (7) | < 0.001 |
| Cancer | 37 (9) | 14 (7) | 23 (13) | 0.04 |
| Atrial fibrillation | 3 (1) | 0 (0) | 3 (2) | 0.06 |
| CRF b | 17 (4) | 2 (1) | 15 (8) | < 0.001 |
| Pain duration (minutes) | 180 (120–360) | 180 (120–300) | 180 (120–360) | 0.3 |
| Cardiogenic shock | 11 (3) | 2 (1) | 9 (5) | 0.02 |
| LVEF (%) | 47.5 (45–50) | 50 (45–55) | 45 (40–50) | 0.02 |
| Territory | 0.8 | |||
| Anterior | 155 (40) | 86 (41) | 69 (39) | |
| Inferior | 214 (55) | 113 (53) | 101 (56) | |
| Lateral | 21 (5) | 12 (6) | 9 (5) | |
| Laboratory variables | ||||
| Creatinine (μmol/L) | 75 (67–90) | 76 (69–89) | 74 (66–92) | 0.4 |
| Haemoglobin (g/dL) | 14.6 (13.6–15.4) | 14.9 (14.2–15.6) | 14.0 (12.9–15.0) | < 0.001 |
| Platelets (× 10 3 /μL) | 223 (188–262) | 222 (186–254) | 226 (191–270) | 0.3 |
| Peak CRP c (mg/L) | 20 (10–52) | 15 (8–34) | 30 (12–80) | < 0.001 |
| Peak troponin d (μg/L) | 43 (15–100) | 42 (14–96) | 44 (18–110) | 0.3 |
| Treatment | ||||
| ACEI/ARB | 354 (91) | 203 (96) | 151 (84) | < 0.001 |
| Beta-blocker | 375 (96) | 206 (98) | 169 (94) | 0.1 |
| Aldosterone antagonist | 85 (22) | 47 (22) | 38 (21) | 0.8 |
| Statin | 381 (98) | 207 (98) | 174 (97) | 0.6 |
| Proton pump inhibitor | 325 (83) | 181 (86) | 144 (80) | 0.2 |
a Comparisons between treatment groups were performed using the χ 2 and Kruskall-Wallis tests.
b CRF is defined by glomerular filtration rate < 60 mL/min/1.73 m 2 .
c Normal value for CRP is < 3 mg/L.
Radial access was more frequent in the prasugrel group ( P = 0.001). Tritroncular disease represented 12% of patients without significant difference between the two groups. Achievement of Thrombolysis In Myocardial Infarction (TIMI) 3 flow was more frequent in the prasugrel group ( P < 0.001) ( Table 2 ).
| Overall population, ( n = 390) | Prasugrel group, ( n = 211) | Clopidogrel group, ( n = 179) | P a | |
|---|---|---|---|---|
| Reperfusion strategy | 0.08 | |||
| Primary PCI | 359 (92) | 200 (95) | 159 (89) | |
| Successful thrombolysis | 15 (4) | 5 (2) | 10 (6) | |
| Rescue PCI | 13 (3) | 6 (3) | 7 (4) | |
| Angiography approach | 0.001 | |||
| Radial access | 344 (88) | 198 (94) | 146 (82) | |
| Femoral access | 27 (7) | 6 (3) | 21 (12) | |
| Femoral cross-over | 8 (2) | 3 (1) | 5 (3) | |
| Infarction-related artery | 0.3 | |||
| Left anterior descending | 160 (41) | 92 (44) | 68 (39) | |
| Circumflex | 65 (17) | 38 (18) | 27 (15) | |
| Right coronary artery | 162 (42) | 81 (38) | 81 (46) | |
| Coronary status | 0.9 | |||
| Monotroncular disease | 214 (55) | 117 (55) | 97 (55) | |
| Bitroncular disease | 127 (33) | 70 (33) | 57 (32) | |
| Tritroncular disease | 46 (12) | 24 (11) | 22 (13) | |
| Thromboaspiration | 216 (56) | 122 (58) | 94 (54) | 0.4 |
| Glycoprotein IIb/IIIa inhibitor | 226 (59) | 133 (63) | 93 (54) | 0.05 |
| ≥ 1 DES | 169 (44) | 107 (51) | 62 (35) | 0.002 |
| Number of stents | 1 (1–3) | 1 (1–3) | 1 (1–3) | 0.9 |
| TIMI 3 flow at end of procedure | 361 (93) | 204 (97) | 157 (89) | < 0.001 |
| Angiography access closure | 0.002 | |||
| Elastic compression | 341 (90) | 196 (95) | 145 (84) | |
| Compression wristband | 10 (3) | 3 (1) | 7 (4) | |
| Angioseal | 27 (7) | 7 (3) | 20 (12) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
