Article Text
PDF
Abstract
Objective Patients with acute coronary syndrome (ACS) remain at high risk for recurrent ischaemic and bleeding events during follow-up. Our study aimed to quantify and compare the impact of these adverse events on quality of life (QoL).
Methods Data from patients with ACS prospectively enrolled in the FORCE-ACS registry between January 2015 and December 2019 were used for this study. The primary ischaemic and bleeding events of interest were hospital readmission for ACS and Bleeding Academic Research Consortium type 2 or 3 bleeding during 12 months follow-up. QoL was measured using the EQ-5D Visual Analogue Scale (VAS) score and the 12-item Short Form Survey version 2 derived Physical Component Summary (PCS) and Mental Health Component Summary (MCS) scores at 12 months follow-up.
Results In total, 3339 patients (mean age 66.8 years, 27.9% women) were included. During follow-up, ischaemic events occurred in 202 patients (6.0%) and bleeding events in 565 patients (16.9%). After adjustment for demographic and clinical characteristics, ischaemic events remained independently associated with lower QoL regardless of metric used. Bleeding was also independently associated with lower EQ-5D VAS and PCS scores, but not with a lower MCS score. The QoL decrement associated with ischaemic events was numerically larger than the decrement associated with bleeding.
Conclusions Ischaemic and bleeding events remain prevalent and are independently associated with lower QoL at 12 months follow-up in patients previously admitted for ACS. The incidence and impact of these adverse events should be considered when balancing individual ischaemic and bleeding risks.
- Acute Coronary Syndrome
- Myocardial Infarction
- Coronary Artery Disease
Data availability statement
Data are available upon reasonable request.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Ischaemic and bleeding events in patients previously admitted for acute coronary syndrome are associated with lower quality of life in the short term.
WHAT THIS STUDY ADDS
Both ischaemic and bleeding events are associated with lower quality of life at 12 months, although the quality of life decrement associated with ischaemic events is more profound compared with the decrement associated with bleeding.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Quantifying the quality of life decrement associated with ischaemic and bleeding events is important to facilitate clinical decision making and interpretation of these events in future trials evaluating antithrombotic therapy.
Introduction
Patients presenting with acute coronary syndrome (ACS) are routinely treated with aspirin and a P2Y12-inhibitor (dual antiplatelet therapy (DAPT)) to reduce the risk of recurrent ischaemic events.1 2 However, the reduction in ischaemic events is partly counterbalanced by an increase in bleeding. Previous studies have demonstrated that major bleeding is associated with a similar increase in mortality compared with recurrent ischaemic events.3 Still, both recurrent ischaemic and bleeding events are not always fatal. Physicians should therefore also consider the impact of these adverse events on patient’s quality of life (QoL) when deciding on the optimal antithrombotic strategy. Importantly, the impact on QoL can be markedly different depending on the type and severity of the event. For example, bleeding events range from nuisance bleeding to disabling intracranial haemorrhage and recurrent ischaemic events range from periprocedural myocardial infarction (MI) with limited prognostic relevance to severe events resulting in heart failure or cardiac arrest.4 By quantifying the QoL decrement associated with adverse events, clinical decision making and interpretation of these events in future trials evaluating antithrombotic regimens can be facilitated. Previous studies have primarily focused on the short-term (1–6 months) impact of bleeding on QoL.5 6 Bleeding events seldom cause irreversible organ damage. Consequently, any decrement in QoL might restore over time, especially following minor bleeding events. Moreover, the impact of recurrent ischaemic events is unclear and therefore the impact of ischaemic events relative to bleeding events is currently unknown. Additionally, most data stem from post-hoc analyses of randomised controlled trials, which are not always representative of daily clinical practice. Therefore, using real-world data from the FORCE-ACS registry, our study aimed to quantify and compare the long-term impact of both recurrent ischaemic and bleeding events on QoL measured at 12 months in patients previously admitted for ACS.
Methods
Study design and patient population
The rationale and design of the FORCE-ACS registry have been described previously.7 In brief, the FORCE-ACS registry is an ongoing prospective registry of nine Dutch hospitals. The primary aim of the registry is to provide insight into different aspects of the diagnosis, management and follow-up of patients with ACS. Patient management, including the use of invasive and pharmacological therapies, in all participating hospitals is performed according to relevant guidelines.1 2 From 2015 onwards, all consecutive adult patients admitted for (suspected) ACS were eligible for participation. Written informed consent was obtained from each patient. The institutional review boards of all participating centres approved the protocol of the FORCE-ACS registry. The current study complies with the principles of the Declaration of Helsinki and reports according to the STrengthening the Reporting of OBservational studies in Epidemiology statement.8
Clinical endpoints
Ischaemic and bleeding events were reported via questionnaires at 1 and 12 months after hospital admission. If patients did not complete the questionnaire, they were contacted by phone. Additionally, the electronic health records of all patients were checked for events. Events were reviewed and adjudicated by the first three authors who had full access to the patient’s electronic health record. The primary ischaemic event of interest was hospital readmission due to unstable angina, non-ST-segment elevation myocardial infarction (NSTEMI) or ST-segment elevation myocardial infarction (STEMI). In case of multiple readmissions due to ACS, patients were assigned the most severe ACS subtype. Events were also classified as early (≤30 days) or late (>30 days) relative to the initial hospital admission. The primary bleeding event of interest was Bleeding Academic Research Consortium (BARC) type 2 or 3 bleeding.9 In short, BARC type 2 bleeding constitutes any overt actionable bleeding not fitting criteria for the more severe bleeding types, but that requires medical intervention, leads to (prolonged) hospitalisation or prompts medical evaluation. BARC type 3 bleeding is defined as bleeding that is associated with a significant drop in haemoglobin, requires transfusion, surgical intervention or intravenous vasoactive agents including cardiac tamponade and intracranial haemorrhage. BARC type 1 bleeding defined as bleeding that is not actionable and does not lead to medical evaluation was documented as an event of secondary interest. Coronary artery bypass grafting (CABG)-related bleeding (BARC type 4) and fatal bleeding (BARC type 5) were not included in the present study. Patients with more than one bleeding event were assigned the most severe BARC bleeding type. Similarly to ischaemic events, bleeding events were classified as early or late events. Bleeding was also classified as access-site related or spontaneous.
Quality of life
At 12 months after initial hospital admission, QoL was measured using the EQ-5D Visual Analogue Scale (VAS) and 12-item Short Form Survey version 2 (SF-12v2).10 11 For the EQ-5D VAS score, patients scored their overall health status on a 20 cm scale ranging from worst imaginable health (0 points) to best imaginable health (100 points).10 From the SF-12v2, two summary scores were derived, the Physical Component Summary (PCS) score and Mental Health Component Summary (MCS) score.11 The items from the physical functioning, role-physical, bodily pain and general health subdomains of the SF-12v2 are indicators for the PCS score while vitality, social functioning, role-emotional and mental health items are indicators for the MCS score. For the general population, the PCS and MCS scores have a mean of 50 and a SD of 10 with higher scores indicating a better health status.11 If patients answered some, but not all, QoL-related items on the questionnaire, the missing items were imputed through multiple imputation using chained equations with five imputations of the dataset. The effect of using these imputed items was assessed by conducting a sensitivity analysis excluding patients with an incomplete questionnaire. Both the EQ-5D VAS and SF-12v2-derived PCS and MCS scores have demonstrated to be reliable and are well-validated across different clinical contexts including in patients with a history of ACS.12 13
Statistical analysis methods
Continuous variables were reported as mean±SD and categorical variables were reported as frequencies and percentages. Patient characteristics were compared by event status using an independent t-test for continuous variables and χ2 test for categorical variables. To estimate the independent association between adverse events and QoL, multivariate linear regression models with EQ-5D VAS, PCS and MCS scores at 12 months as dependent variable were used. The models were adjusted for the following potential confounders: age, sex, initial diagnosis (ie, unstable angina, NSTEMI or STEMI), revascularisation during initial hospital admission (ie, percutaneous coronary intervention (PCI) or CABG) and presence of at least one concomitant chronic disease (ie, diabetes, atrial fibrillation, chronic kidney disease, chronic obstructive pulmonary disease or peripheral artery disease). Additionally, in all models examining the association of ischaemic events and QoL, bleeding was included as a potential confounder and vice versa. Using a similar modelling approach, the impact of different types of recurrent ischaemic event (unstable angina, NSTEMI or STEMI) and severity of bleeding (BARC type 1, 2 or 3) were assessed. Likewise, the influence of early compared with late events was assessed.
In order to place these results into clinical context, the magnitude of the difference in QoL was estimated using Cohen’s d.14 15 Cohen’s d is a dimensionless index that quantifies the magnitude of effect size.14 15 Cohen’s d was calculated by dividing the adjusted effect estimates of ischaemic and bleeding events on the different QoL metrics by the SD of the corresponding QoL metric in the pooled study population. Based on current convention, a Cohen’s d of 0.2 was considered small, 0.5 moderate and 0.8 large.14 Significance was set at a p value of <0.05. Statistical analyses were performed using SPSS V.28 (SPSS, Chicago, Illinois, USA) and illustrative graphics were composed using GraphPad Prism V.8.3 (GraphPad Software, San Diego, California, USA).
Patient and public involvement
Patients and the wider public were not involved in the design or conduct of this study.
Results
Patient characteristics
From January 2015 until December 2019, 5813 patients were included in the FORCE-ACS registry. In total, 5188 patients were diagnosed with ACS at discharge. Overall, 294 patients died before 12 months follow-up and 1555 patients did not complete the QoL-related section of the questionnaire at 12 months. Patients who did not complete the QoL-related section of the questionnaire were younger, but had more concomitant chronic diseases compared with patients who did complete the questionnaire (online supplemental table 1). Ultimately, 3339 patients were included in the present analysis. A detailed flowchart is provided in figure 1. The mean age at time of enrollment was 66.8±11.1 years and 27.9% of patients were women (table 1). Overall, 9.0%, 49.7% and 41.3% of patients were diagnosed with unstable angina, NSTEMI and STEMI, respectively. Almost all patients (96.1%) underwent coronary angiography during initial hospital admission and 69.6% and 13.5% of patients subsequently underwent PCI or CABG (table 2). Most patients (84.5%) were treated with DAPT, while 15.7% of patients received an oral anticoagulant at discharge.
Supplemental material
Flowchart. ACS, acute coronary syndrome; QoL, quality of life.
Baseline characteristics of patients with and without recurrent ischaemic and/or bleeding events during 12 months follow-up
In-hospital management of patients with and without recurrent ischaemic and/or bleeding events during 12 months follow-up
Ischaemic events during follow-up
During the first 12 months of follow-up, ischaemic events occurred in 202 (6.0%) of the 3339 included patients (table 3). Most patients with ischaemic events presented with NSTEMI followed by STEMI and unstable angina. Approximately one in three (37.1%) recurrent ischaemic events occurred within 30 days of the index hospital admission. Patients with recurrent ischaemic events were more likely to be older, current or former smokers, to have hypertension, diabetes, chronic kidney disease and atrial fibrillation and have a history of PCI, CABG, MI or stroke compared with patients without new ischaemic events (table 1). Patients with ischaemic events were less likely to undergo revascularisation during initial hospital admission and more likely to be discharged on an oral anticoagulant (table 2).
Clinical outcomes at 12 months in patients previously admitted for ACS
Bleeding during follow-up
BARC type 2 or 3 bleeding occurred in 565 (16.9%) of the 3339 included patients within 12 months follow-up (table 3). Almost half (44.2%) of all BARC type 2 or 3 bleeding events occurred within 30 days of the index hospitalisation and approximately three out of ten BARC type 2 or 3 bleeds were access-site related. Furthermore, 1049 patients (31.4%) who did not have a BARC type 2 or 3 bleeding did report at least one BARC type 1 bleeding within the first 12 months of follow-up. Patients with BARC type 2 or 3 bleeding were more likely to be older, female, have a lower body mass index and were more likely to have hypertension, atrial fibrillation and chronic kidney disease compared with those without bleeding (table 1). Patients with BARC type 2 or 3 bleeding were also more likely to be discharged on an oral anticoagulant (table 2).
Association of ischaemic events and QoL
At 12 months follow-up, patients with ischaemic events had a lower EQ-5D VAS score compared with patients without an ischaemic event (65.9±21.5 vs 72.3±18.6; p<0.01) (online supplemental figure 1). Similarly, PCS and MCS scores were lower at 12 months in patients with recurrent ischaemic events compared with those without (39.0±10.4 vs 44.0±10.2, p<0.01 and 46.2±10.9 vs 48.0±10.2, p=0.02). After adjustment for demographic and clinical characteristics, recurrent ischaemic events within 12 months remained independently associated with a lower EQ-5D VAS score of 5.06 (95% CI: −7.70 to −2.43; p<0.01; Cohen’s d: −0.27) (figure 2). Differences in EQ-5D VAS score were more pronounced in patients with unstable angina and NSTEMI compared with STEMI (online supplemental figure 2). Recurrent ischaemic events were also independently associated with lower PCS (−3.54; 95% CI: −4.90 to −2.18; p<0.01; Cohen’s d: −0.34) and MCS scores (−1.76; 95% CI: −3.22 to −0.30; p=0.02; Cohen’s d: −0.17). PCS score was lower in patients with unstable angina and NSTEMI, while MCS score was lower in patients with STEMI. Differences for all QoL metrics were more profound in patients with late ischaemic events compared with those with early events (online supplemental figure 3). A sensitivity analysis excluding patients with incomplete QoL questionnaires did not significantly affect the effect estimates (online supplemental figure 4).
Association between ischaemic versus bleeding events and QoL at 12 months in patients previously admitted for ACS. Association is expressed as effect estimates including 95% CI of ischaemic and bleeding events adjusted for age, sex, initial diagnosis, revascularisation during initial hospital admission and presence of at least one concomitant chronic disease. ACS, acute coronary syndrome; MCS, Mental Health Component Summary; PCS, Physical Ccomponent Summary; QoL, quality of life; VAS, Visual Analogue Scale.
Association of bleeding and QoL
Patients with BARC type 2 or 3 bleeding had a lower EQ-5D VAS score at 12 months compared with patients without bleeding (69.0±19.7 vs 72.5±18.6, p<0.01) (online supplemental figure 5). Similarly, PCS and MCS scores at 12 months were lower in patients with bleeding compared with those without bleeding (41.0±10.3 vs 44.2±10.2, p<0.01 and 46.7±10.5 vs 48.1±10.2, p<0.01). BARC type 2 or 3 bleeding within 12 months following hospital admission was independently associated with a lower EQ-5D VAS score of 2.84 (95% CI: −4.61 to −1.08; p<0.01; Cohen’s d: −0.15) (figure 2). As anticipated, BARC type 3 bleeding was associated with a greater difference in EQ-5D VAS score compared with BARC type 2 bleeding (online supplemental figure 6). BARC type 1 bleeding was not associated with a difference in EQ-5D VAS score. BARC type 2 or 3 bleeding was independently associated with a lower PCS score (−2.24; 95% CI: −3.15 to −1.33; p<0.01; Cohen’s d: −0.22). The difference in PCS score was again dependent on bleeding severity. Importantly, even BARC type 1 bleeding was associated with a lower PCS score. BARC type 2 or 3 bleeding was not associated with a difference in MCS score (−0.84 (95% CI: −1.82 to 0.13, p=0.09)). However, when patients with access-site related bleeding were excluded, BARC type 2 or 3 bleeding was associated with a lower MCS score (online supplemental figure 7). Exclusion of access-site related bleeding did not alter the association between BARC type 2 or 3 bleeding and other metrics of QoL. Differences were more profound in patients with late bleeding compared with patients with early bleeding regardless of QoL metric (online supplemental figure 8). The negative association between BARC type 2 or 3 bleeding and MCS became more apparent after exclusion of patients with incomplete QoL questionnaires, while the other QoL metrics were not significantly affected by exclusion of these patients (online supplemental figure 9).
Discussion
The most important findings of the study are as follows: (1) both recurrent ischaemic and bleeding events are associated with lower QoL at 12 months even after adjustment for demographic and clinical characteristics, (2) the degree of QoL decrement increases with bleeding severity (ie, BARC type 3 bleeding has the largest impact on QoL, whereas—contrary to previous studies—BARC type 1 bleeding has little or no impact on QoL) and (3) the impact of ischaemic events on QoL is numerically larger than the impact of bleeding (figure 3).
Negative impact of ischaemic and bleeding events on QoL at 12 months in patients previously admitted for ACS. Impact is expressed as Cohen’s d using the effect estimates including 95% CI of ischaemic and bleeding events adjusted for age, sex, initial diagnosis, revascularisation during initial hospital admission and presence of at least one concomitant chronic disease. ACS, acute coronary syndrome; MCS, Mental Health Component Summary; PCS, Physical Component Summary; QoL, quality of life; VAS, Visual Analogue Scale.
Observational studies have reported rates of recurrent ischaemic events up to 18% in the 12 months following hospital admission for ACS.16 Although these rates have declined in recent decades due to an uptake of coronary interventions and evidence-based medication, the residual ischaemic risk remains high.17 Previous studies have demonstrated that ischaemic events negatively affect different metrics of QoL.18–20 However, the specific impact of recurrent events, especially shortly after the index event has been less clear. Our results show that recurrent ischaemic events are negatively associated with all QoL metrics, although the impact on physical QoL, measured by the PCS score, was most profound. Importantly the QoL decrement associated with ischaemic events appears clinically meaningful with a small-to-moderate effect on QoL based on a Cohen’s d of −0.17 to −0.27 depending on the QoL metric used. Interestingly, ischaemic events occurring after 30 days of the index hospital admission were associated with a larger difference in QoL at 12 months compared with early events, insinuating that the impact of these events decreases over time. Still given the overall impact of recurrent ischaemic events, it remains important to implement strategies aimed at reducing the ischaemic risk. However, if these strategies increase bleeding risk (eg, DAPT prolongation), careful risk stratification is warranted to select patients who stand to benefit most for a given strategy.21
There is considerable variation in reported bleeding rates, partly due to the different bleeding definitions used.22 However most studies reiterate that bleeding is common, especially when considering minimal or nuisance bleeding (ie, BARC type 1 bleeding).22 In the present study, nearly half of patients had some form of bleeding during the 12 months following hospital admission. Although most bleeding was classified as BARC type 1 bleeding (eg, bruising or self-limiting epistaxis), almost one in five patients had a more severe BARC type 2 or 3 bleeding. Importantly, previous studies have shown that BARC type 2 or 3 bleeding is associated with lower QoL when measured within the first 6 months after ACS.5 6 Our study extends these observations by showing that the effect on quality of life of BARC type 2 and 3 bleeding persists beyond 6 months up until 12 months follow-up. However, effect estimates in our study were smaller compared with previous work measuring QoL at 1–6 months after admission, suggesting that the impact of bleeding on QoL attenuates over time.23 Furthermore, in the present study late bleeding events were associated with a larger difference in QoL at 12 months compared with bleeding that occurred within 30 days of the initial hospital admission. Thus, again, suggesting that the impact of bleeding on QoL decreases over time. Moreover, in our study BARC type 1 bleeding was not or minimally associated with lower QoL compared with a more pronounced and consistent association in studies with a 1–6 months timeframe.5 6 Importantly, previous work suggests that adequate counselling with regards to the risk of minimal or nuisance bleeding can limit the impact of these complications on QoL and therefore possible improve medication adherence.24 Although the impact of BARC type 2 or 3 bleeding on QoL appears smaller compared with the impact of ischaemic events based on the Cohen’s d ranging from −0.08 to −0.22, the high incidence of bleeding highlights the need to reduce bleeding risk. Similar to the present study, other observational studies have reported a threefold higher rate of BARC type 2 or 3 bleeding compared with the rate of recurrent ischaemic events.25
Most clinical trials focus on ‘hard’ outcomes, such as (cardiovascular) mortality, MI, stroke and major bleeding. Interpretation of results from these trials, especially those investigating antithrombotic therapies, is hampered by the use of composite endpoints, which combine ischaemic and bleeding outcomes. The individual components of these combined endpoints can have markedly different impact on mortality, morbidity and especially quality of life. Quantifying the impact of recurrent ischaemic and bleeding events therefore has important implications for future clinical trials. Until now, patient reported outcomes such as different QoL metrics have not been routinely collected during most clinical trials. By quantifying the QoL decrement associated with recurrent ischaemic and bleeding events, depending on type and severity, more nuanced balancing of ischaemic and bleeding risks is possible.
Limitations
This study has several important limitations. First, 31.8% of patients with ACS alive at 12 months follow-up did not complete the QoL-related section of the questionnaire, possibly introducing selection bias. Second, in the FORCE-ACS registry events were primarily self-reported by patients and may have been subject to recall bias. Although events were verified by reviewing the electronic health record and other relevant source documentation in most cases, this was not possible for BARC type 1 bleeding, which by definition is not actionable and does not lead to medical evaluation. Third, despite adjustment for potential confounders the possibility of unmeasured confounding remains. More specifically, future studies should include a QoL measurement at time of the index event to enable correction for QoL at baseline. Finally, given the observational nature of our study the association between ischaemic or bleeding events and lower QoL does not imply causality.
Conclusions
Ischaemic and bleeding events following hospital admission for ACS are common and associated with lower QoL even at 12 months follow-up, while the degree of QoL decrement depends on the type and severity of the adverse event. The incidence and impact of these adverse events should be considered when balancing ischaemic and bleeding risks.
Data availability statement
Data are available upon reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants and was approved by Ethics Committee: MEC-U Reference number: V.32279/W14.073/hs/cl. Participants gave informed consent to participate in the study before taking part.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
JMtB and WJK are joint senior authors.
NMRvdS and JA are joint first authors.
Twitter @NielsvdSangen
NMRvdS and JA contributed equally.
JMtB and WJK contributed equally.
Correction notice This article has been corrected since it was first published. Middle initials have been added to authors Jurriën M ten Berg and Wouter J Kikkert.
Contributors NMRvdS, JA and DCPY were responsible for data collection, statistical analysis and preparation of the original draft of the manuscript. WJK, JMtB and JPSH were responsible for the supervision of the project. All other authors were involved in data collection and critically reviewed the original draft of the manuscript. NMRvdS and JA are responsible for the overall content of the manuscript as a guarantor.
Funding The FORCE-ACS registry is supported by grants from ZonMw, the St. Antonius Research Fund and AstraZeneca. The authors are solely responsible for the design and conduct of this study, all study analyses, the drafting and editing of the manuscript and its final contents.
Competing interests WJK has received an institutional research grant from AstraZeneca. GJV has received institutional research grants from MicroPort and Ferrer and personal fees from Terumo and AstraZeneca. YA has received an institutional research from the Dutch Heart Foundation. JPSH has received institutional research grants from Abbott Vascular, AstraZeneca, B. Braun, Getinge, Ferrer, Infraredx and ZonMw. JMtB has received institutional research grants from AstraZeneca, Daiichi Sankyo and ZonMw and personal fees from AstraZeneca, Bayer, Boehringer Ingelheim, CeleCor Therapeutics, Daiichi Sankyo, Eli Lilly, Ferrer and Idorsia. All other authors have no relationships with industry to disclose.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.