http://orcid.org/0000-0001-9837-2977
Figures
Abstract
Background
Major bleedings other than gastrointestinal (GI) and intracranial (ICH) and mortality rates associated with antiplatelet drugs in real-world clinical practice are unknown. The objective was to estimate major bleeding risk and mortality among new users of antiplatelet drugs in real-world clinical practice.
Methods and findings
A population-based prospective cohort using the French national health data system (SNIIRAM), identified 69,911 adults living within five well-defined geographical areas, who were new users of antiplatelet drugs in 2013–2015 and who had not received any antithrombotics in 2012. Among them, 63,600 started a monotherapy and 6,311 a dual regimen. Clinical data for all adults referred for bleeding was collected from all emergency departments within these areas, and medically validated. Databases were linked using common key variables. The main outcome measure was time to major bleeding (GI, ICH and other bleedings). Secondary outcomes were death, and event-free survival (EFS). Hazard ratios (HR) were derived from adjusted Cox proportional hazard models. We used Inverse Propensity of Treatment Weighting as a stratified sensitivity analysis according to the antiplatelet monotherapy indication: primary prevention without cardiovascular (CV) risk factors, with CV risk factors, and secondary prevention. We observed 250 (0.36%) major haemorrhages, 81 ICH, 106 GI and 63 other types of bleeding. Incidences were twice as high in dual therapy as in monotherapy. Compared to low-dose aspirin (≤ 100 mg daily), high-dose (> 100 up to 325 mg daily) was associated with an increased risk of ICH (HR = 1.80, 95%CI 1.10 to 2.95). EFS was improved by high-dose compared to low-dose aspirin (1.41, 1.04 to 1.90 and 1.32, 1.03 to 1.68) and clopidogrel (1.30, 0.73 to 2.3 and 1.7, 1.24 to 2.34) respectively in primary prevention with and without CV risk factors.
Citation: Bouget J, Balusson F, Viglino D, Roy P-M, Lacut K, Pavageau L, et al. (2020) Major bleeding risk and mortality associated with antiplatelet drugs in real-world clinical practice. A prospective cohort study. PLoS ONE 15(8): e0237022. https://doi.org/10.1371/journal.pone.0237022
Editor: Timir Paul, East Tennessee State University, UNITED STATES
Received: April 23, 2020; Accepted: July 18, 2020; Published: August 7, 2020
Copyright: © 2020 Bouget et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: Under French law and regulations, patient-level data from the SNIIRAM database cannot be made publicly available. Data access requests may be directed to info@indsante.fr. A list of relevant data set names to request the minimal data set are provided in the Supporting Information file. The statistical code is available from Figshare (DOI: 10.6084/m9.figshare.12581999.v1).
Funding: This study was supported by the National Clinical Research Hospital Program of the French Ministry of Health (PHRC-12-009-0243). The funder had no role in the study design, data collection, analysis, or interpretation, in the writing of the article, or in the decision to submit it for publication.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Antiplatelet drugs are used for primary and secondary prevention of ischemic cardiovascular events [1]. Long-term low-dose aspirin (<325 mg daily) or clopidogrel are recommended for patients with symptomatic peripheral artery disease, before and after peripheral arterial bypass surgery or percutaneous transluminal angioplasty [2], and for patients with non-cardio-embolic ischemic stroke and transient ischemic attack [3,4]. For patients with acute coronary syndromes who undergo a percutaneous coronary intervention with stent placement, dual antiplatelet therapy for the first year is recommended, with low-dose aspirin in combination with clopidogrel, ticagrelor or prasugrel. Among patients with coronary stents, low-dose aspirin plus clopidogrel is recommended for 12 months [5,6]. Aspirin is also used for primary prevention among patients with an increased risk of cardiovascular events or among patients with diabetes mellitus, although these indications appear more controversial [7,8].
The longer life expectancy of the population, and consequently the increase in the prevalence of these ischemic cardiovascular diseases, have led to an increase in the proportion of patients who require treatment with antiplatelet drugs, especially among the elderly [9].
Whatever their indications and the regimen used, antiplatelet drugs are associated with a risk of bleeding, leading to substantial morbidity and mortality. In monotherapy, the increased risk of gastrointestinal bleeding has been commonly reported [10], in contrast to the risk of intracranial bleeding which appears more controversial [11]. In dual therapy, the risk of major non-fatal bleeding is increased whatever the dual antiplatelet regimen [12]. Major bleedings other than GI and intracranial have almost never been reported.
Mortality rates associated with antiplatelet drugs seem dependent on the indications: in two meta-analyses of randomized clinical trials [7,13], and in contemporary US cohorts [14], a similar all-cause mortality was reported with aspirin alone versus control in primary prevention. A decrease in cardiovascular mortality with aspirin versus control was observed in secondary prevention [13]. Various mortality rates have been reported when several antiplatelet drugs were compared to each other in randomized trials [15–17]. To our knowledge, mortality rates associated with antiplatelet treatment in real-world clinical practice are not known.
Our aim was to estimate the incidence of major bleeding and mortality associated with antiplatelet drugs, irrespective of the indication, in real-world clinical practice. Only new users of antiplatelet drugs were considered in this prospective study.
Methods
Study design and participants
The study design has previously been reported [18]. Briefly, a prospective population-based cohort study was set up linking the French Health Insurance Database (SNIIRAM) to data from all emergency departments located within five well-defined areas around five large French cities (Angers, Brest, Grenoble, Nantes and Rennes). The study received regulatory approval (Commission Nationale de l'Informatique et des Libertés CNIL, DR-2013-488 with subsequent substantial changes DR-2016-489). The need for informed consent was waived.
The SNIIRAM anonymously and comprehensively links the universal healthcare reimbursement database (DCIR) to the French hospital discharge database (PMSI). The DCIR contains anonymous individual data on all reimbursements for health expenditure, including drugs. This database does not provide any direct information on the medical indication for each reimbursement. The PMSI provides hospital discharge diagnoses (ICD-10 code) as well as details on medical acts. Thus, all adult subjects (> 18 years) living in the five above-mentioned areas, affiliated to the French universal Health Insurance System, and with at least one reimbursement for an antiplatelet drug (aspirin, clopidogrel, prasugrel, ticagrelor) in 2013–2015 were identified.
From emergency departments, clinical data for all adult subjects living in the five above-mentioned areas and referred between January 1, 2013 and December 31, 2015 for bleeding while receiving an antiplatelet drug was prospectively collected, focusing on the type of bleeding. Their demographics were collected (month and year of birth, gender) as well as their date of hospital admission. In each emergency department, the medical referent validated the final inclusion of all records of severe bleeding detected.
The SNIIRAM and clinical databases were linked using common key variables (date of birth (month, year), gender, date of hospital entry and discharge, type of antiplatelet drug, and care facility involved).
Only "new users" defined as having had neither an antiplatelet drug delivery nor any other antithrombotic drug delivery in 2012 were analysed.
Patient characteristics were collected first. We calculated the adapted Charlson’s index (see S1 Table in S1 File) from co-morbidities and co-medication retrieved from the SNIIRAM database [19].
Outcomes
The primary outcome was major bleeding. Secondary outcomes were death and major-bleeding-event-free survival.
Major bleeding was defined from at least one of the following criteria: unstable haemodynamics (systolic arterial pressure < 90 mmHg or mean arterial pressure < 65 mm Hg) or haemorrhagic shock, uncontrollable bleeding, need for transfusion or haemostatic procedure (embolization, endoscopic procedure, surgery). The location or the symptoms then defined the type: intracranial haemorrhage (ICH), acute gastrointestinal (GI) bleeding, and other major bleeding in life-threatening locations—intraspinal, intraocular, retroperitoneal, pericardial, thoracic, intra-articular, or intramuscular haematoma with compartment syndrome. We also considered the following as "other" major bleeding events: epistaxis with at least two procedures of nasal packing, and hematuria when bleeding lasted more than 12 hours despite bladder washing. There was a slight alteration with respect to the International Society on Thrombosis and Haemostasis (ISTH) classification of major bleeding events, because in our dataset no information was available on haemoglobin levels [20]. Major bleeding events were further classified into three classes: intracranial haemorrhage (ICH), GI bleeding and other bleeding events. For each type of bleeding event, censoring was based on death or any other type of major bleeding event or end of study follow-up, whichever came first.
Major-bleeding-event-free survival was defined as time to major bleeding or death or end of study follow-up, whichever came first.
Main exposure to antiplatelet drugs
Antiplatelet drug exposure was defined as the first observed delivery, and patients were categorised as new users of either low-dose aspirin (acetylsalicylic acid, ASA ≤ 100 mg daily), high-dose ASA (more than 100 up to 325 mg daily), or clopidogrel for mono-therapies, and ASA + clopidogrel or prasugrel or ticagrelor for dual therapies. Patients were censored if they discontinued antiplatelet therapy (using the date of end of supply) or if they switched to another antithrombotic drug (oral anticoagulant, parenteral anticoagulant). The date of first delivery of the new treatment following a switch was considered as the end of the supply of the antiplatelet drug. The same applied in case dual prescription. Patients were likewise censored in case of death or major bleeding, or if they had moved away from the area, or had reached the end of study follow-up, whichever occurred first.
The indication for antiplatelet prescription was derived from the main discharge diagnosis and/or medical acts performed in the 30 days before the first observed issue of antiplatelet therapy (see S2 Table in S1 File). Three groups of patients according to antiplatelet monotherapy indications were defined: 1) patients with antiplatelet drug for secondary prevention (coronary disease, carotid artery disease, peripheral artery disease, stroke, valvular heart disease, miscellaneous diseases), 2) patients with antiplatelet drugs for primary prevention with identified cardiovascular (CV) risk factors (arterial hypertension, diabetes, renal disorders, lipid-lowering drugs), and 3) patients with antiplatelet drugs for primary prevention without any identified CV risk factors (no factors identified by the SNIIRAM database).
Statistical analysis
Crude incidence rates (IR) were calculated for the first bleeding episode per 10,000 person-months according to the type of bleeding, the type of antiplatelet regimen, and the duration of use (under 6 months, 6 to 12 months or over one year for monotherapy, and under 3 months, 3 to 6 months and over 6 months for dual therapy).
Cox proportional hazard regression analyses were conducted for each type of bleeding event to determine hazard ratios for different antiplatelet regimens compared to low-dose ASA (≤100 mg) used as a reference for monotherapy, and compared to ASA + clopidogrel as a reference for dual therapy, adjusted on known patient characteristics: age, gender, diabetes, arterial hypertension, previous bleeding, medication use (lipid-lowering drug, proton pump inhibitor, NSAID) and co-morbidities using the adapted Charlson index.
For secondary outcomes, Cox proportional hazard regression models were run to estimate death and event-free survival for any antiplatelet drug in mono or dual therapy, irrespective of bleeding type.
For antiplatelet monotherapy, a sensitivity analysis stratified by indication was conducted. We first used the stabilized inverse probability of treatment weighting (SIPTW) based on the propensity score [21]. Separate analyses were performed for the three main indications of antiplatelet monotherapy: primary prevention without CV factors, primary prevention with CV factors and secondary prevention. The weights were truncated by resetting the value of weights greater (or lower) than the 99th (1st) percentile to the value of 99th (1st) percentile [22]. Weighted Kaplan-Meier survival functions were generated for death and event-free survival for these 3 indications.
All statistical tests were two-tailed and P-values <0.05 were considered significant. Statistical analyses were performed using SAS software 9.4 (SAS Institute, Cary, N.C., USA).
Results
Cohort characteristics
Between 2013 and 2015, 246,272 patients aged 18 or over living within the defined areas had at least one delivery of antiplatelet drugs. Only 69,992 patients were new users of antiplatelet drugs and constituted the study cohort (Fig 1, flow chart). Antiplatelet monotherapy accounted for 91% of the cohort. Sixty-eight patients receiving combinations of antiplatelet drugs other than mono or dual therapy were excluded from the final analysis.
Characteristics of the population according to antiplatelet therapy (mono and dual therapy) are reported in Table 1. Among patients under different antiplatelet monotherapy regimens, the median age ranged from 65 to 67 years and 19% to 21% of the patients were over 80. Out of 5,674 patients classified as having coronary artery disease as presumed indication for antiplatelet monotherapy, 1,026 had been hospitalized (out of them 705 suffered from ACS) and 4,048 had ambulatory care for angina pectoris identified through stress tests. Then 705 out of 5,674 patients (12.4%) suffered from ACS. Whatever the antiplatelet regimen, 85% to 95% of patients had an adapted Charlson comorbidity index of 2 or under. Patients with dual therapy including prasugrel or ticagrelor were younger, and respectively 96% or 97% had a coronary disease with current use of lipid-lowering drugs. Twenty-seven to 32% of the patients with different antiplatelet monotherapy regimens were using concomitant antiulcer medication, and 56% to 59% of the patients with different dual therapy regimens.
The distribution of CV risk factors according to antiplatelet therapy among patients treated for primary prevention is shown in S3 Table in S1 File. Among patients with low-dose ASA, high-dose ASA and clopidogrel monotherapy, 53%, 63% and 59% respectively had one risk factor, 36%, 29% and 32% respectively had 2 risk factors and 11%, 7% and 9% had at least 3 risk factors.
During the study period, a total of 250 (0.36%) patients experienced a first major bleeding event: 81 patients (0.12%) with ICH, 106 (0.15%) with GI bleeding, and 63 (0.09%) with other major bleeding.
Bleeding sites according to antiplatelet drugs are reported in the supplementary material in S4 Table in S1 File Major bleeding was about twice as frequent among patients with dual therapy compared to monotherapy, in particular with the ASA + prasugrel or ticagrelor combinations. GI was the main site of bleeding, whatever the antiplatelet regimen, especially the upper GI tract. Among the 81 ICH observed, 43 were intracerebral haemorrhages and 31 extra- or subdural hematomas. Hematuria and oropharyngeal bleeding were the most frequent other major types of bleeding.
Incidence rates for major bleeding events according to exposure to antiplatelet monotherapies
Crude incidence rates of major bleeding according to antiplatelet monotherapy and time period are reported in Fig 2 (Panel A). Values are detailed in the supplementary material in S5A Table in S1 File. Altogether, incidence rates were low. The incidence rates of intracranial haemorrhage (ICH) and gastrointestinal (GI) bleeding were higher in the first six months of treatment with high-dose ASA, and decreased regularly with time. This observation was not found with low-dose ASA nor with clopidogrel.
Panel A: Monotherapy. Panel B: Dual antiplatelet regimen.
Associations between exposure to antiplatelet monotherapy and outcomes
Forest plots showing adjusted hazard ratios (HRs) for each first bleeding episode, death and event-free survival for antiplatelet monotherapy, compared to low-dose ASA as a reference are presented in Fig 3.
Compared to low-dose ASA, high-dose ASA was associated with a significantly increased risk of ICH (HR = 1.80, 95%CI 1.10 to 2.95) and a non-significant increased risk of GI bleeding (HR = 1.54, 95%CI 0.93 to 2.55). Clopidogrel was associated with a non-significant increased risk of other major bleeding events (HR = 2.03, 95%CI 0.90 to 4.57). No significant differences were reported concerning death or EFS (Fig 3).
Stratified sensitivity analyses were conducted across three indications: primary prevention without CV risk factors (n = 19,761), primary prevention with CV risk factors (n = 22,690), and secondary prevention (n = 21,149). Baseline characteristics are reported in supplementary S6 Table in S1 File.
Association estimates between major bleeding, death or event-free survival and high-dose ASA or clopidogrel compared to low-dose ASA as a reference are reported as follows: in Fig 4 for primary prevention without CV risk factors, in Fig 5 for primary prevention with CV risk factors and in Fig 6 for secondary prevention. EFS was significantly improved by high-dose ASA (HR = 1.41, 95%CI 1.04 to 1.90 and HR = 1.32, 95%CI 1.03 to 1.68) and clopidogrel (HR = 1.30, 95%CI 0.73 to 2.3 and HR = 1.7, 95%CI 1.24 to 2.34) in primary prevention among patients without CV risk factor and in primary prevention among patients with CV risk factor respectively, compared to low-dose ASA. In secondary prevention, clopidogrel was significantly better than low-dose ASA in terms of EFS (HR = 0.58, 95%CI 0.38 to 0.88) but there were a few bleeding events (Fig 6). In primary prevention without CV risk factors (Fig 4), high-dose ASA was associated with a higher risk of ICH and GI bleeding. In primary prevention with CV risk factors (Fig 5), high-dose ASA and clopidogrel were associated with a higher risk of death and lower EFS. In secondary prevention (Fig 6), clopidogrel was associated with a lower risk of death and better EFS.
Weighted Kaplan-Meyer survival functions for EFS according to antiplatelet therapy and indications are shown in Fig 7. In primary prevention without CV risk factor, low-dose ASA had a significantly better survival function than high-dose ASA and clopidogrel (p = 0.051) (Fig 7 panel A). Clopidogrel had a significantly better EFS in secondary prevention (p = 0.042) (Fig 7 panel C) and a poorer survival function in primary prevention with CV risk factors (p<0.001) (Fig 7 panel B). Weighted Kaplan-Meyer survival functions for death according to antiplatelet monotherapy and indications are reported in S1 Fig. There was significantly poorer survival with clopidogrel compared to low-dose ASA in primary prevention with CV risk factors (p<0.001).
Panel A: for primary prevention without cardiovascular risk factors. Panel B: for primary prevention with CV risk factors. Panel C: for secondary prevention.
Incidence rates for major bleeding according to exposure to dual antiplatelet regimens
Crude incidence rates for major bleeding according to dual antiplatelet regimens and time period are reported in Fig 2 (Panel B). Values are detailed in supplementary S5B Table in S1 File. Incidence rates for all major bleeding events were higher in dual regimens than in monotherapy. There were no significant variations in the incidence rates of any major bleeding event with the different dual regimens, whatever the duration of the treatment.
Association between exposure to dual antiplatelet regimen and outcomes
Forest plots showing adjusted hazard ratios (HR) for each first bleeding episode, death and EFS for dual antiplatelet regimens compared to ASA + clopidogrel as a reference are reported in Fig 8. No significant differences in the different types of major bleeding were reported between the combinations of ASA + prasugrel or ticagrelor.
Discussion
Principal findings
Using an administrative healthcare database linked to prospectively collected clinical data from emergency departments, our population-based cohort study showed low incidence rates of major bleeding and low mortality rates among new antiplatelet drug users. A 2- to 3-fold increased risk of gastrointestinal bleeding and other major bleeding was observed in dual regimens compared to monotherapy, with a better safety profile for low-dose ASA (≤ 100 mg) among monotherapies. There were no significant differences across dual regimens. We observed very few events in patients prescribed with dual regimens, keeping in mind only major bleeding were accounted for; given these low incidences we had a very low power to detect any difference.
Strengths and weaknesses
This large, comprehensive study is original for its ability to directly compare different antiplatelet drugs, including the different doses and regimens, encompassing all indications, to explore important, common safety outcomes. Patients exposed to anticoagulants combined with antiplatelet therapy were not included in this analysis. Analysing major bleeding minimizes the bias that affects the external validity of studies focusing on hospital data only, where patients diagnosed and managed in primary care are not included. An important point is that this study retrieved and linked data from a prospective multi-centre clinical study and from a public healthcare system that covers all residents within the defined area. As a result, the dataset provided a complete picture of all hospitalizations and prescriptions dispensed, minimizing selection bias. There is indeed a risk of misclassification related to coding errors at the time of hospital admission, although this may not be very likely for serious conditions like bleeding and with prospective medical validation. We analyzed a population exposed to antiplatelet agents, irrespective of the indication, then including a larger scope than coronary artery disease. This is the reason why we did not consider bleeding scores devoted to coronary artery disease. Moreover, we focused on major bleeding occurring after discharge when antiplatelet agents were prescribed in secondary prevention, that is after an hospitalization for an acute thrombotic event. We did not collected early major bleeding after medical or surgical procedures. All bleeding events were medically validated using pre-specified criteria which were somewhat consistent with the "severe or life-threatening' item from GUSTO. The systematic medical validation of all bleeding limited the overestimation of incidence rates of major bleeding [23]. Exposure in our study was based on reimbursement claims data. We studied drug exposure on the basis of pharmacy dispensation but we had no information on the patients’ actual intake. The lack of information on patient adherence could have led to incorrect estimations of exposure, but the clinical validation of major bleeding made it possible to check that patients were still receiving antiplatelet drugs at the time of bleeding. Causes of death were unfortunately not available. The design and linkage to Health Insurance database allowed to state the study was exhaustive within five well-defined areas around five large French cities. But translating the result to French population is more hazardous. However, we could hypothesize that the reasons why those patients were referred to emergency department for major bleeding should not be different in other French regions. Of note, racial data cannot be collected in France. We hypothesized our population was mostly of Caucasian descent.
Differences with other studies
Several studies have reported higher rates of major bleeding associated with antiplatelet monotherapy, especially when used for secondary prevention [9,24,25]. The meta-analysis by the Antithrombotic Trialist collaboration reported similar observations, but the authors noted heterogeneity across the studies included [13]. Several reasons could explain the discrepancy between our results and those from earlier studies.
Firstly, there was inconsistency across studies in the definition of major bleeding, whatever the type of bleeding, and in the inclusion criteria; and there were differences between randomized clinical trials (RCT) and observational studies [10]. For example, Vazquez et al. in a meta-analysis of 15 trials reported rates of major bleeding with aspirin ranging from 0% in 5 trials up to 19.1% [26]. it can be noted that most of the trials included used different definitions of bleeding and 4 trials did not define major bleeding. Moreover, Mehran et al. [27] suggested standardized definitions for bleeding for cardiovascular clinical trials that were different from those defined by Kaatz et al. [19], which are closer to our criteria. GI bleeding is the main reported bleeding risk with antiplatelet drugs [28]. This observation was confirmed in the present study for all antiplatelet drugs, in monotherapy or dual therapy, but with overall lower incidence rates. This could be explained by the strict definition of the severity of major bleeding used in our study and by the systematic medical validation of all events [23].
Secondly, most studies were set in a specific clinical setting–or indication—for antiplatelet drugs, such as coronary disease [29], stroke prevention [25], peripheral artery disease [30], or primary prevention [7]. We hypothesised that bleeding risk related to antiplatelet drugs was mostly related to patient characteristics and not to their indication. Our population had an overall low comorbidity score on the adapted Charlson index, which could explain low rates of major bleeding. In primary prevention with aspirin, Zheng et al. in his meta-analysis reported an overall incidence of major bleeding of 23.1 events per 10,000 participant-years with a daily dose of aspirin from 50 mg to 500 mg [7]. We report higher incidence rates with high-dose ASA according to exposure time, whatever the indication.
A decrease in the bleeding risk over time has been previously reported. In a post-hoc analysis of individual data from 6 RCTs on antiplatelet therapy after transient ischemic attack or ischemic stroke, Hilkens et al. [31] showed high early risk of major bleeding with antiplatelet monotherapy and dual therapy, and high early risk of GI bleeding with dual therapy. No significant variations in incidence rates were found for ICH with mono or dual therapy, unlike Li et al. [9] who reported a reduced risk of bleeding events over time among both patients on monotherapy and patients treated initially with ASA + clopidogrel. In the present study, no similar variations were observed with clopidogrel nor with the different dual therapy regimens, probably because of a lack of power related to a small number of events.
Most studies have reported a higher risk of major bleeding with dual therapy compared to monotherapy, especially for GI bleeding [32–38]. In the meta-analysis by Savarese et al. [12] including 15 randomized studies, dual therapy induced a significant increase in non-fatal bleeding without an increase in ICH or fatal bleeding. These results were independent from the type of dual therapy. In our study, ICH did not significantly increase with dual therapy compared to monotherapy. We did not detect significant differences between the combinations used. These results are in line with pivotal studies on dual therapy in various indications [12,36,37,39].
Low-dose aspirin is usually defined as a daily dose from 75 mg to 325 mg. We defined two dosage regimens for ASA in monotherapy to evidence any dose-dependent bleeding risk. We thus report a lower risk of ICH and GI bleeding with ASA ≤100 mg daily compared to ASA >100 up to 325 mg daily. Similar results were reported by Peters et al. among patients with acute coronary syndromes, with an increased risk of major bleeding with increasing doses of aspirin [40]. For ICH, the present results have not been reported elsewhere, in particular neither by Hilkens et al. [31] nor by Cea Soriano et al. [11] Like low-dose ASA, clopidogrel increased the risk of upper or lower GI bleeding [41]. However, the CAPRIE study reported a risk of severe GI bleeding that was significantly lower among patients treated with clopidogrel (0.49%) than among patients treated with ASA 325 mg daily (0.71%) (p<0.05) [42]. This finding is in line with our observation of similar incidence rates of GI bleeding between low-dose ASA (≤100 mg) and clopidogrel.
Other major bleedings were not reported in most studies. In our study they accounted for 25% of all major bleeding events observed, mainly hematuria, oropharyngeal bleeding and epistaxis.
In addition EFS can have an interest over and above overall survival. Patients who experience major bleeding may have a lower quality of life. Surprisingly RCTs and observational studies have reported on major bleeding events and overall mortality, but not on EFS. It should be noted that the ASPREE study used a composite primary end-point including death, dementia and persistent physical disability to encompass more than bleeding events as such [43].
Whatever the antiplatelet drug used, in monotherapy or in dual therapy, we did not show any differences in mortality or EFS. The prospective registry in Dresden reported considerable in-hospital mortality after GI bleeding under antiplatelet drugs (11.9%) [44]. In most studies assessing dual therapy, no significant fatal bleedings were reported across the different regimens [12]. We evidenced lower mortality with antiplatelet monotherapy than other studies [7,8,15,16,30,42]. We do not have any rational explanation for the lower death rate in our study, especially as the characteristics of our population seem similar to those in these other studies in terms of median age and co-morbidities.
Conclusion
Our observation in real-world clinical practice supports the best safety profile of low-dose ASA (≤100 mg daily) for secondary and primary prevention of cardiovascular diseases, and clopidogrel as an alternative for secondary prevention. We did not detect significant differences between dual antiplatelet regimens in terms of major bleeding events or death.
Supporting information
S1 Fig. Weighted Kaplan-Meyer survival functions for death according to antiplatelet monotherapy and indications.
Panel A: Primary prevention without risk factors. Panel B: Primary prevention with risk factors. Panel C: Secondary prevention.
https://doi.org/10.1371/journal.pone.0237022.s001
(TIF)
References
- 1. Eikelboom JW, Hirsh J, Spencer FA, Baglin TP, Weitz JI. Antiplatelet drugs. Antithrombotic therapy and prevention of thrombosis, 9th ed: American college of chest physicians evidence-based clinical practice guidelines. Chest. 2012;141(2 suppl):e89S–e119S. pmid:22315278
- 2. Alonso-Coello P, Bellmunt S, McGorrian C, et al. Antithrombotic therapy in peripheral artery disease. Antithrombotic therapy and prevention of thrombosis, 9th ed: American college of chest physicians evidence-based clinical practice guidelines. Chest. 2012; 141(2 suppl):e669S–90S. pmid:22315275
- 3. Lansberg MG, O’Donnell MJ, Khatri P, et al. Antithrombotic and thrombolytic therapy for ischemic stroke. Antithrombotic therapy in peripheral artery disease. Antithrombotic therapy and prevention of thrombosis, 9th ed: American college of chest physicians evidence-based clinical practice guidelines. Chest. 2012; 141(2 suppl):e601S–36S. pmid:22315273
- 4. Kernan WN, Ovbiagele B, Black HR, et al, on behalf of the American heart association stroke council, Council on cardiovascular and stroke nursing, Council on clinical cardiology, and Council on peripheral vascular disease. Guidelines for the prevention of stroke in patients with stroke and transient ischemic attack. A guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke. 2014;45:2160–236. pmid:24788967
- 5. Vandvik PO, Lincoff AM, Gore JM, et al. Primary and secondary prevention of cardiovascular disease. Antithrombotic therapy and prevention of thrombosis, 9th ed: American college of chest physicians evidence-based clinical practice guidelines. Chest. 2012; 141(2 suppl):e637S–68S. pmid:22315274
- 6. Roffi M, Patrono C, Collet JP, et al. 2015 ESC guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: task force for the management of acute coronary syndromes in patients without persistent ST-segment elevation of the European society of cardiology (ESC). Eur Heart J. 2016;37(3):267–315. pmid:26320110
- 7. Zheng SL, Roddick AJ. Association of aspirin use for primary prevention with cardiovascular events and bleeding events. A systematic review and meta-analysis. JAMA. 2019;321(3):277–87. pmid:30667501
- 8. McNeil JJ, Wolfe R, Woods RL, et al for the ASPREE investigator group. Effect of aspirin on cardiovascular events and bleeding in the healthy elderly. N Engl J Med. 2018;379(16):1509–18. pmid:30221597
- 9. Li L, Geraghty OC, Mehta Z, Rotwell PM, on behalf of the Oxford Vascular study. Age-specific risks, severity, time course, and outcome of bleeding on long-term antiplatelet treatment after vascular events: a population-based cohort study. Lancet. 2017;390:490–9. pmid:28622955
- 10. Valkhoff VE, Sturkenboom M, Hill C, Veldhuyzen van Zanten S, Kuipers EJ. Low-dose acetylsalicylic acid use and the risk of upper gastrointestinal bleeding: a meta-analysis of randomized clinical trials and observational studies. Can J Gastroenterol. 2013;27(3):159–67. pmid:23516680
- 11. Cea Soriano L, Gaist D, Soriano-Gabarro M, Bromley S, Garcia-Rodriguez LA. Low-dose aspirin and risk of intracranial bleeds. An observational study in UK general practice. Neurology. 2017;89:2280–7. pmid:29093065
- 12. Savarese G, Savonitto S, Lund LH, et al. Efficacy and safety of prolonged dual antiplatelet therapy: a meta-analysis of 15 randomized trials enrolling 85 265 patients. Eur Heart J Cardiovasc Pharmacother. 2016;2(4):218–28. pmid:27533949
- 13. Antithrombotic Trialists’ (ATT) Collaboration. Aspirin in the primary and secondary prevention of vascular disease: collaborative meta-analysis of individual participant data from randomized trials. Lancet. 2009;373:1849–60. pmid:19482214
- 14. Huang WY, Singer DE, Wu YL, et al. Association of intracranial haemorrhage risk with non-vitamin K antagonist oral anticoagulant use vs. aspirin use: a systematic review and meta-analysis. JAMA Neurol. 2018;75:1511–8. pmid:30105396
- 15. Diener HC, Bogousslavsky J, Brass LM, et al., on behalf of the MATCH investigators. Aspirin and clopidogrel compared with clopidogrel alone after recent ischaemic stroke or transient ischaemic attack in high-risk patients (MATCH): a randomized, double-blind, placebo-controlled trial. Lancet. 2004;364:331–7. pmid:15276392
- 16. Bhatt DL, Fox KAA, Hacke W, et al. for the CHARISMA investigators. Clopidogrel and aspirin versus aspirin alone for the prevention of atherothrombotic events. N Engl J Med. 2006;354:1706–17. pmid:16531616
- 17. Mauri L, Elmariah S, Yeh RW, et al., for the DAPT study investigators. Causes of late mortality with dual antiplatelet therapy after coronary stents. Eur Heart J. 2016;37:378–85 pmid:26586780
- 18. Bouget J, Balusson F, Scailteux LM, et al. Major bleeding with antithrombotic drugs: a 2012–2015 study using the French nationwide Health Insurance database linked to emergency department records within five areas. Rationale and design of SACHA study. Fund Clin Pharmacol. 2019;33:443–62.
- 19. Bannay A, Chaignot C, Blotière PO, et al. The best use of the Charlson comorbidity index with electronic health care database to predict mortality. Med Care. 2016;54:188–94. pmid:26683778
- 20. Kaatz S, Ahmad D, Spyropoulos AC, Schulman S for the Subcommittee on Control of Anticoagulation. Definition of clinically relevant non-major bleeding in studies of anticoagulants in atrial fibrillation and venous thromboembolic disease in non-surgical patients: communication from the SSC of the ISTH. J Thromb Haemost. 2015;13:2119–26. pmid:26764429
- 21. Austin PC. An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behav Res. 2011;46:399–424. pmid:21818162
- 22. Cole SR, Hernán MA. Constructing inverse probability weights for marginal structural models. Am J Epidemiol. 2008;168:656–64. pmid:18682488
- 23. Oger E, Botrel MA, Juchault C, Bouget J. Sensitivity and specificity of an algorithm based on medico-administrative data to identify hospitalized patients with major bleeding presenting to an emergency department. BMC Med Res Methodol. 2019 Oct 18;19(1):194. pmid:31627721
- 24. The SALT collaborative group. Swedish Aspirin Low-dose Trial (SALT) of 75 mg aspirin as secondary prophylaxis after cerebrovascular ischemic events. Lancet. 1991;338:1345–9. pmid:1682734
- 25. Sandercock PAG, Counsell C, Tseng MC, Cecconi E. Oral antiplatelet therapy for acute ischaemic stroke. Cochrane database Syst Rev. 2014;26.
- 26. Vazquez FJ, Gonzales JP, LeGal G, Carrier M, Gẚndara E. Risk of major bleeding in patients receiving vitamin K antagonists or low doses of aspirin. A systematic review and meta-analysis. Thromb Res. 2016;138:1–6. pmid:26826501
- 27. Mehran R, Rao SV, Bhatt DL, et al. Standardized bleeding definitions for cardiovascular clinical trials. A consensus report from the bleeding academic research consortium. Circulation. 2011;123:2736–47. pmid:21670242
- 28. Lanas A, Carrera-Lasfuentes P, Arguedas Y, et al. Risk of upper and lower gastrointestinal bleeding in patients taking nonsteroidal anti-inflammatory drugs, antiplatelet agents, or anticoagulants. Clin Gastroenterol Hepatol. 2015;13:906–12. pmid:25460554
- 29. Spoendlin J, Gagne JJ, Lexey JJ, Patorno E, Schneeweiss S, Desai RJ. Comparative effectiveness and safety of antiplatelet drugs in patients with diabetes mellitus and acute coronary syndrome. Pharmacoepidemiol Drug Saf. 2018;27:1361–70. pmid:30379372
- 30. Mahmoud AN, Elgendy AY, Rambarat C, Mahtta D, Elgendy IY, Bavry AA. Efficacy and safety of aspirin in patients with peripheral vascular disease: an updated systematic review and meta-analysis of randomized controlled trials. PLoS One. 2017;12(4):e0175283. pmid:28403216
- 31. Hilkens NA, Kappelle LJ, Bath PM, et al. on behalf of the CAT collaboration. Early time course of major bleeding on antiplatelet therapy after TIA or ischemic stroke. Neurology. 2018;90:e683–9. pmid:29374102
- 32. The SPS3 investigators. Effects of clopidogrel added to aspirin in patients with recent lacunar stroke. N Engl J Med. 2012;368:817–25.
- 33. Ding L, Peng B. Efficacy and safety of dual antiplatelet therapy in the elderly for stroke prevention: a systematic review and meta-analysis. Eur J Neurol. 2018;25:1276–84. pmid:29855121
- 34. Johnston SC, Easton JD, Farrant M, et al, for the Clinical Research Collaboration Neurological Emergencies Treatment Trials Network and the POINT Investigators. Clopidogrel and aspirin in acute ischemic stroke and high-risk TIA. N Engl J Med. 2018;379:215–25. pmid:29766750
- 35. Berger JS, Bhatt DL, Steg PG, et al. Bleeding, mortality, and antiplatelet therapy: results from the clopidogrel for high atherothrombotic risk and ischemic stabilization, management and avoidance (CHARISMA) trial. Am Heart J. 2011;162:98–105. pmid:21742095
- 36. Wiviott SD, Braunwald E, McCabe CH, et al., for the TRITON-TIMI 38 investigators. Prasugrel versus clopidogrel in patients with acute coronary syndromes. N Engl J Med. 2007;357:2001–15. pmid:17982182
- 37. Wallentin L, Becker RC, Budaj A, et al., for the PLATO investigators. Ticagrelor versus clopidogrel in patients with acute coronary syndromes. N Engl J Med. 2009;361:1045–57. pmid:19717846
- 38. Tillman H, Johnston C, Farrant M, et al. Risk for major hemorrhages in patients receiving clopidogrel and aspirin compared with aspirin alone after transient ischemic attack or minor ischemic stroke. A secondary analysis of the POINT randomized clinical trial. JAMA Neurology. 2019;76:774–82. pmid:31034032
- 39. Garcia Rodriguez LA, Lin KJ, Hernandez-Diaz S, Johansson S. Risk of upper gastrointestinal bleeding with low-dose acetylsalicylic acid alone and in combination with clopidogrel and other medications. Circulation. 2011;123:1108–15. pmid:21357821
- 40. Peters RJG, Mehta SR, Fox KAA, et al, for the Clopidogrel in Unstable angina to prevent Recurrent events (CURE) trial investigators. Effects of aspirin dose when used alone or in combination with clopidogrel in patients with acute coronary syndromes. Observations from the Clopidogrel in Unstable angina to prevent Recurrent Events (CURE) study. Circulation. 2003;108:1682–7. pmid:14504182
- 41. Di Minno A, Spadarella G, Prisco D, Scalera A, Ricciardi E, Di Minno G. Antithrombotic drugs, patient characteristics, and gastrointestinal bleeding: clinical translation and areas of research. Blood Reviews. 2015. pmid:25866382
- 42. Caprie Steering Committee. A randomized, blinded, trial of clopidogrel versus aspirin in patients at risk of ischaemic events (CAPRIE). Lancet. 1996;348:1329–39. pmid:8918275
- 43. McNeil JJ, Nelson MR, Woods RL, et al. for the ASPREE investigator group. Effect of aspirin on all-cause mortality in the healthy elderly. N Engl J Med. 2018;379:1510–28.
- 44. Pannach S, Goetze J, Marten S, Schreier T, Tittl L, Beyer-Westendorf J. Management and outcome of gastrointestinal bleeding in patients taking oral anticoagulants or antiplatelet drugs. J Gastroenterol. 2017 pmid:28210867