Abstract
Observational studies that have reported an association between aspirin use in chronic obstructive pulmonary disease (COPD) with reductions in mortality and COPD exacerbations were shown to be affected by time-related biases. We assessed this association using a prevalent new-user study design that avoids these biases. We used the United Kingdom’s Clinical Practice Research Datalink (CPRD) to form a cohort of patients with COPD. Aspirin initiators were matched on time and propensity score with nonusers during 2002–2018. The outcomes were all-cause mortality and COPD exacerbation within a one-year follow-up. Hazard ratios (HR) and 95% confidence interval (CI) of each outcome associated with aspirin use compared to nonuse were estimated using an as-treated approach. The study cohort included 10,287 initiators of aspirin and 10,287 matched nonusers. The cumulative incidence of all-cause mortality at one year was 11.5% for aspirin users and 9.2% for nonusers. The HR of all-cause mortality associated with aspirin initiation was 1.22 (95% CI: 1.08–1.37), while for severe exacerbation it was 1.21 (95% CI 1.08–1.37), compared with nonuse. The HR of a first moderate or severe exacerbation with aspirin use was 0.90 (95% CI 0.85–0.95). These estimates did not vary by platelet count. This large population-based study, designed to emulate a trial, found aspirin use in patients with COPD associated with a higher risk of all-cause mortality and severe exacerbation, but a lower risk of moderate or severe exacerbation. Further research is warranted to assess this reduction in moderate or severe exacerbations, particularly in patients with cardiovascular risk factors.
Introduction
Chronic obstructive pulmonary disease (COPD) is a leading cause of death and disability worldwide [Citation1,Citation2]. Cardiovascular disease is a major co-morbidity in COPD and is a predictor of all-cause mortality [Citation3]. As such, there has been increasing interest in studying the role of cardiovascular medications, including statins and beta-blockers, to reduce morbidity and mortality in patients with COPD. Several observational studies have reported important reductions in COPD exacerbations and mortality with the use of these drugs [Citation4,Citation5]; reductions that later could not be confirmed in large randomized controlled trials [Citation6,Citation7]. Reviews into the studies’ methods and results revealed various biases that tended to skew the results toward beneficial effects [Citation8–10].
Aspirin, or acetylsalicylic acid (ASA), has also been the subject of several observational studies that reported decreased incidence of death and exacerbations with its use [Citation11–18]. A meta-analysis of five of these observational studies reported a relative risk of all-cause mortality of 0.81 (95% CI; 0.75–1.13) with aspirin/anti-platelet use in patients with COPD [Citation19]. Indeed, aspirin is hypothesized to decrease platelet activation and inflammation, which have been associated with COPD exacerbation and mortality [Citation16,Citation20]. Yet, a review of their methods found bias in these studies, casting doubt over the reliability of their results [Citation21].
We therefore conducted an observational study to assess the effectiveness of aspirin use in reducing all-cause mortality and exacerbations in patients with COPD. We used a study design that emulated a trial, thus avoiding the sources of bias and methodological limitations in the previous studies.
Methods
Data sources
We used the UK’s Clinical Practice Research Datalink (CPRD), a primary care database that contains electronic medical records for 50 million people enrolled in more than 1,800 general practices. Trained participating general practitioners record medical information, including demographic data, lifestyle factors, including smoking and alcohol use, and medical information, such as demographic data, and medical diagnoses, using the Read and SNOMED coding classification. Prescriptions written by general practitioners are automatically coded based on the UK Prescription Pricing Authority Dictionary. The Hospital Episodes Statistics (HES) database records data on all hospital admissions, primary and secondary diagnoses using ICD-10 codes (International Classification of Diseases, Tenth Revision). The Office for National Statistics (ONS) database collects electronic death certificates, including the cause of death using ICD-10 codes. Data from the HES and the ONS can only be linked to general practices in England which represent 85% of all practices. CPRD was linked with both HES and ONS for this study.
Study design
We used a prevalent new-user design, a cohort approach designed to emulate a trial, to assess the effectiveness of aspirin use. First, we formed a base cohort of patients with a diagnosis of COPD with at least one prescription for long-acting β2 agonists (LABAs), long-acting muscarinic antagonists (LAMAs) or LABAs with inhaled corticosteroids (LABA-ICS) from January 1, 2002, until November 2018. To increase the accuracy of the diagnosis of COPD, we included only patients older than 50 years at the first such prescription (base cohort entry). Subjects with less than one year of baseline data prior to base cohort entry, with an asthma diagnosis (to exclude patients with only asthma), use of LABA, LAMA or aspirin prior to base cohort entry were excluded. Subjects were followed up from base cohort entry until November 30, 2018.
The prevalent new-user design first involved identifying all subjects initiating aspirin use during base cohort follow-up. Each new user of aspirin generated an exposure set of all subjects in the base cohort with the same year of base cohort entry and who had a physician visit on the date of the new user’s aspirin initiation (± 7 d). The date of aspirin initiation and of the corresponding physician visit in the exposure set defined study cohort entry. Subjects with a diagnosis of myocardial infarction in the year prior to study cohort entry were excluded. Time-conditional propensity scores were computed, conditional on the time of the exposure set, as a function of all covariates measured at that time (see below). Each new user of aspirin was matched with the nonuser within the exposure set with the closest value of the propensity score. Subjects were followed up to 1 year after study cohort entry. Follow-up ended on the date of death, on November 30, 2018 (availability of HES data), at the end of coverage in the practice, or one year after cohort entry whichever came first.
Outcomes
The primary outcome was all-cause mortality. Secondary outcomes were time to first moderate or severe exacerbation and to first severe exacerbation to occur after cohort entry. A moderate exacerbation was defined as a new prescription for prednisolone while a severe exacerbation was defined as an admission to hospital for COPD as the primary cause (ICD-10 codes J41-J44). This definition for severe exacerbation has been previously validated with a positive predictive value of 86% and a specificity of 99% [Citation22].
Covariates
Measured covariates included age, sex, body-mass index (BMI), smoking status, and alcohol abuse (i.e. alcoholism, alcoholic fatty liver, alcoholic cirrhosis, alcohol induced polyneuropathy, mental & behave disease due to use alcohol: psychotic disorder, schizophrenia-like, nondependent alcohol abuse NOS). COPD factors included the number of moderate and severe exacerbations, of hospitalizations for pneumonia, an asthma diagnosis, the use of respiratory drugs, and antibiotics, all measured during the 1-year baseline period prior to study cohort entry, as well as a severe exacerbation in the 30 d prior to study cohort entry. The most recent value of percent predicted FEV1, and the degree of dyspnea as measured by mMRC, CAT score or dyspnea symptoms were also included. The most recent platelet count was also identified. We used diagnoses and prescriptions during the 1-year baseline period prior to study cohort entry to identify various co-morbid conditions observed in COPD. In particular, cardiovascular disease risk factors in the year before study cohort entry included diagnoses of stroke and heart failure, the use of ACE-inhibitors, Angiotensin II receptor blockers, β-blockers, calcium-channel blockers, thiazides diuretics, anti-platelets, and statins. In addition, diagnoses of cancer, renal disease and thyroid disease were identified during this baseline period.
Statistical analyses
Descriptive statistics were used to summarize the characteristics of the patients in the matched study arms, with the accuracy of matching assessed using standardized differences. Kaplan-Meier curves were used to estimate the survival function of time to death and time to exacerbations from cohort entry. The Cox proportional hazards model was used to estimate the hazard ratio (HR) and the 95% confidence interval (CI) of each outcome associated with aspirin use compared to no use. We considered an as-treated approach, with continuing exposure to aspirin measured by successive prescriptions, assuming a 45-day exposure period after each prescription.
We assessed the influence of platelet levels given their potential implication in prognostic worsening in COPD patients[Citation16]. We categorized platelet levels as fewer than 200 cells per µL, 200–300 cells per µL, and more than 300 cells per µL. We also used cubic splines to model the hazard ratio (HR) of death and severe and moderate exacerbations comparing aspirin use with no use, as a function of continuous platelet concentration.
Sensitivity analyses included varying the prescription durations to 30 and 60 d as well as using a modified intention-to-treat (ITT) approach where nonusers who received aspirin during follow-up were censored. We also used an alternative validated definition of moderate or severe exacerbation defined by prescriptions for antibiotics and oral corticosteroids, diagnostic codes for lower respiratory tract infection or acute exacerbation of COPD, as well as a record of two or more treated respiratory symptoms of AECOPD, namely increase in breathlessness or cough or sputum volume and/or purulence[Citation23]. This definition of exacerbations yielded a positive predictive value of over 85% [Citation24]. Finally, we carried out subgroup analyses. We stratified by prior cardiovascular disease risk factors in the year before study cohort entry, as well as by other antiplatelet medications used in this baseline year. As suggested by a review, we performed a post-hoc sensitivity analysis in excluding all patient with FEV1/FVC above the lower limit of normal (LLN) at baseline and missing values [Citation25].
We performed all analyses using SAS version 9.4. The study protocol was approved by the Independent Scientific Advisory Committee of the CPRD (protocol 20_135) and by the Research Ethics Board of the Jewish General Hospital, Montreal, Canada.
Results
The base cohort included 98,069 eligible patients after applying the exclusion criteria, of which 10,539 initiated aspirin after cohort entry (). The baseline characteristics of these patients before matching are given in Supplementary Table 1. After propensity score matching and applying the positivity assumption, the study cohort included 10,287 initiators of aspirin and 10,287 matched nonusers. The baseline characteristics of these matched patients were well balanced as indicated by the standardized differences <0.10 (). Only 466 (4.5%) of the nonusers had initiated aspirin during follow-up.
Figure 1. Flowchart of cohort selection.
Table 1. Baseline characteristics of the study cohort of 10,287 initiators of aspirin and their time-conditional propensity score matched nonusers, among patients with COPD, with the corresponding standardized mean differences.
There were 1345 deaths from any cause, 1,271 severe exacerbations requiring hospitalizations and 5,828 severe or moderate exacerbations in the study cohort. The Kaplan-Meier analyses find a cumulative incidence of all-cause mortality at one year of 11.5% for aspirin users and 9.2% for nonusers (). For a severe exacerbation, the one-year incidence was 10.5% for aspirin users and 8.7% for nonusers (), while for a moderate or severe exacerbation, it was 37.0% for aspirin users and 40.6% for nonusers ().
Figure 2. Kaplan–Meir curves and cumulative incidences comparing acetylsalicylic acid (ASA) use (red) with nonuse (blue), in patients with COPD in the first year after treatment initiation on: (A) All-cause mortality; (B) Severe exacerbation; and (C) Moderate or severe exacerbation.
The HR of all-cause mortality associated with aspirin initiation was 1.22 (95% CI: 1.08–1.37), while for severe exacerbation it was 1.21 (95% CI 1.08-1.37), compared with nonuse (). The HR of a first moderate or severe exacerbation with aspirin use was 0.90 (95% CI 0.85–0.95).
Table 2. Adjusted HRs of all-cause mortality and COPD exacerbation, comparing ASA use with nonuse in patients with COPD in the first year after treatment initiation, from the as-treated analysis.
As a function of platelet levels, the HR for all-cause mortality was 0.98 (95% CI 0.74–1.29) in patients with counts <200 × 103 cells/µL, while it was 1.32 (95% CI 1.10–1.58) for counts of 200-300 × 103 cells/µL, and 1.12 (95% CI 0.91–1.38) for counts >300 × 103 cells/µL (). The HR of a moderate or severe exacerbation did not vary with platelet count (). However, the HR of a severe exacerbation decreased gradually from 1.57 (95% CI 1.16–2.14) in patients with counts <200 × 103 cells/µL, to 1.09 (95% CI 0.88–1.35) for counts >300 × 103 cells/µL ().
Table 3. Adjusted HRs of all-cause mortality and COPD exacerbation, comparing ASA use with nonuse in patients with COPD in the first year after treatment initiation, stratified by platelet count at baseline (among subjects with platelet count), from the as-treated analysis.
In the sensitivity analyses for the primary mortality outcome, the results were consistent when prescription durations were varied to 30 and 60 d, when excluding patients with FEV1/FVC > LLN or missing value at baseline, and when using the intention-to-treat analysis (). The HR of mortality with aspirin use varied somewhat when stratified by CVD risk factors and other antiplatelet medication use prior to cohort entry (). Indeed, the HR was lower in patients with prior CVD risk factors and was null in patients with prior use of other antiplatelet medications (HR 0.81; 95% CI: 0.58–1.13). The results for severe and moderate or severe exacerbations of COPD were also consistent in sensitivity analyses with no evidence of effect-modification by prior CVD risk factor, other antiplatelet use or when excluding patients with FEV1/FVC > LLN or missing value at baseline, for these two outcomes (Supplementary Tables 2 and 3). Finally, the effect of aspirin on moderate and severe exacerbation (HR 0.90; 95% CI: 0.85–0.95) was similar when the alternative definition of exacerbation was used (HR 0.93; 95% CI: 0.88–0.98).
Table 4. Adjusted HRs of all cause-mortality, from sensitivity analyses and stratified by prior cardiovascular disease (CVD), by other antiplatelet, and restricting to patients with FEV1/FVC < lower limit of normal (LLN), comparing ASA use with nonuse in patients with COPD in the first year after treatment initiation.
Discussion
This population-based cohort study finds that aspirin use was associated with a slightly higher risk of death and of severe exacerbation of COPD during the first year after aspirin initiation, compared with nonuse. On the other hand, aspirin use was associated with a lower incidence of moderate or severe exacerbation. Overall, these findings remained robust in several sensitivity analyses, though the effects on mortality varied according to the presence of CVD risk factors and the use of other antiplatelet medication prior to aspirin initiation.
These results contrast with most observational studies and meta-analyses published to date. This is the first study suggesting a small increase in the risk of death and of severe exacerbation of COPD in aspirin users compared to nonusers [Citation19,Citation21]. These discrepancies can possibly be explained by the more rigorous methods applied in designing our study, which aimed to minimize and account for all potential biases affecting the previous observational studies [Citation21]. Indeed, our prevalent new-user design avoided immortal time bias by defining cohort entry at the first prescription of aspirin and using a time-matched nonuser comparator [Citation26]. This also allowed us to avoid selection bias from collider stratification by properly timing the covariates used in the propensity score [Citation27]. We also avoided outcome misclassification by using specific and validated codes for mortality and COPD exacerbations [Citation22]. Lastly, we took into account a broad range of known confounding factors in estimating the time-conditional propensity score for matching aspirin users and nonusers.
However, the increase in mortality with aspirin use, despite the design and analysis techniques used to minimize biases, was not uniform across all patients. Indeed, we found that, among patients who were using other antiplatelet medications prior to initiating aspirin, there was no increased risk of death with aspirin initiation (HR 0.81; 95% CI 0.58–1.13). As well, among patients with prior CVD risk factors the mortality increase with aspirin was lower (HR 1.16; 95% CI 1.02–1.32) than among those with no such risk factors. These findings suggest a role for CVD in this association, while the higher deleterious effect of aspirin in COPD patients with low cardiovascular risk could potentially be related to hemorrhagic events [Citation28]. However, we did not have the cause of death in this study to further explore this hypothesis.
On the other hand, we found that aspirin use was associated with a 10% lower incidence of moderate or severe exacerbation of COPD. This finding was confirmed when we used an alternative definition for this combined outcome that has been previously validated [Citation24]. While this was not our primary outcome of interest, it is consistent with the SPIROMICS study that reported a 22% reduction in the rate of moderate or severe exacerbation[Citation14]. While this latter study was limited by potential misclassification of aspirin exposure, because it was not updated during follow-up, the corresponding intent-to-treat analysis of our study, which does not update aspirin exposure during follow-up, also found a reduction in the risk of moderate or severe exacerbation. These two consistent findings warrant further studies of this question, including small, randomized trials for this outcome.
Our study has limitations. Firstly, while aspirin is mainly prescribed by general practitioners in the UK, exposure misclassification is possible, given that CPRD does not record prescriptions written by specialists and that aspirin can be obtained over the counter. Moreover, the CPRD only record drug prescription but not patients’ adherence to the treatment regimens. Secondly, some outcome misclassification may have been present notably for moderate exacerbation, which was only defined by a new prescription of prednisolone in COPD patients, though the results were similar using a previously validated alternative definition for this outcome. Thirdly, residual confounding, notably on CV risk factors, is possible given the observational nature of the study and the absence of an active comparator. However, we performed several sensitivity analyses to explore the impact of prior CVD and platelet use, and we used time-conditional propensity scores that included several potential confounders, which likely reduced the possibility of major unmeasured confounding.
Thrombocytosis and systemic inflammation have been shown to be linked to morbidity and mortality in COPD patients [Citation16,Citation20,Citation29]. By reducing circulating inflammatory markers and activated platelets, aspirin is hypothesized to improve clinical outcomes [Citation30]. A crossover pilot trial aiming at determining the optimal dose of aspirin that blocks platelet activation in COPD patients, before conducting larger clinical trials on clinical outcomes, is ongoing [Citation31]. In our study we were unable to assess the impact of the dose of aspirin and a subgroup of COPD patients presenting systemic inflammatory markers. However, the lack of a clear link between platelet counts and aspirin effectiveness in our study lends some support to the absence of a robust beneficial pharmacological action of aspirin in COPD exacerbation and pathophysiology.
In summary, in this large population-based study, the use of aspirin in COPD patients was associated with a slightly higher risk of all-cause mortality and severe exacerbations among patients with chronic obstructive pulmonary disease, but a lower risk of moderate or severe exacerbation. Further research will be needed to investigate this potential impact of aspirin use on reducing moderate or severe exacerbations, as well as on aspirin dose, particularly in subgroups of COPD patients with systemic inflammation and cardiovascular risk factors.
Authors contributions
Dr. Kouri participated in the study design, data interpretation, and wrote the initial draft of the manuscript. Ms Dell’Aniello participated in the study design, data analysis and writing of the manuscript. Dr. Ernst participated in the study design, data interpretation, and writing of the manuscript. Dr. Suissa participated in data acquisition, study design, data interpretation, writing of the manuscript, and acts as guarantor of this entire manuscript.
Supplemental Material
Download PDF (607.4 KB)Disclosure statement
S.S. attended scientific advisory committee meetings or received speaking fees from AstraZeneca, Atara, Boehringer-Ingelheim, Bristol-Myers-Squibb, Merck, Novartis, Panalgo, Pfizer and Seqirus. Role of the sponsors: The funding sources had no role in the design, analysis, or interpretation of the results.
Additional information
Funding
References
- Christenson SA, Smith BM, Bafadhel M, et al. Chronic obstructive pulmonary disease. Lancet. 2022;399(10342):1–9. doi: 10.1016/S0140-6736(22)00470-6.
- Adeloye D, Song P, Zhu Y, et al. Global, regional, and national prevalence of, and risk factors for, chronic obstructive pulmonary disease (COPD) in 2019: a systematic review and modelling analysis. Lancet Respir Med. 2022;10(5):447–458. doi: 10.1016/S2213-2600(21)00511-7.
- Huiart L, Ernst P, Suissa S. Cardiovascular morbidity and mortality in COPD. Chest. 2005;128(4):2640–2646. doi: 10.1378/chest.128.4.2640.
- Mancini GBJ, Etminan M, Zhang B, et al. Reduction of morbidity and mortality by statins, angiotensin-converting enzyme inhibitors, and angiotensin receptor blockers in patients with chronic obstructive pulmonary disease. J Am Coll Cardiol. 2006;47(12):2554–2560. doi: 10.1016/j.jacc.2006.04.039.
- Etminan M, Jafari S, Carleton B, et al. Beta-blocker use and COPD mortality: a systematic review and meta-analysis. BMC Pulm Med. 2012;12(1):48. doi: 10.1186/1471-2466-12-48.
- Criner GJ, Connett JE, Aaron SD, et al. Simvastatin for the prevention of exacerbations in moderate-to-severe COPD. N Engl J Med. 2014;370(23):2201–2210. doi: 10.1056/NEJMoa1403086.
- Dransfield MT, Voelker H, Bhatt SP, et al. Metoprolol for the prevention of acute exacerbations of COPD. N Engl J Med. 2019;381(24):2304–2314. doi: 10.1056/NEJMoa1908142.
- Suissa S. Co-morbidity in COPD: the effects of cardiovascular drug therapies. Respiration. 2010;80(1):3–7. doi: 10.1159/000315387.
- Suissa S, Ernst P. Beta-blockers in COPD: a methodological review of the observational studies. COPD. 2018;15(5):520–525. doi: 10.1080/15412555.2018.1554637.
- Sule NO, Suissa S. Statins and mortality in COPD: a methodological review of observational studies. COPD J. 2023;20(1):284–291. doi: 10.1080/15412555.2023.2242489.
- Campo G, Pavasini R, Malagù M, et al. Relationship between troponin elevation, cardiovascular history and adverse events in patients with acute exacerbation of COPD. COPD. 2015;12(5):560–567. doi: 10.3109/15412555.2014.995293.
- Ekström MP, Hermansson AB, Ström KE. Effects of cardiovascular drugs on mortality in severe chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 2013;187(7):715–720. doi: 10.1164/rccm.201208-1565OC.
- Ellingsen J, Johansson G, Larsson K, et al. Impact of comorbidities and commonly used drugs on mortality in COPD – Real-world data from a primary care setting. Int J Chron Obstruct Pulmon Dis. 2020;15:235–245. doi: 10.2147/COPD.S231296.
- Fawzy A, Putcha N, Aaron CP, et al. Aspirin use and respiratory morbidity in COPD: a propensity score-matched analysis in subpopulations and intermediate outcome measures in COPD study. Chest. 2019;155(3):519–527. doi: 10.1016/j.chest.2018.11.028.
- Goto T, Faridi MK, Camargo CA, et al. The association of aspirin use with severity of acute exacerbation of chronic obstructive pulmonary disease: a retrospective cohort study. NPJ Prim Care Respir Med. 2018;28(1):7. doi: 10.1038/s41533-018-0074-x.
- Harrison MT, Short P, Williamson PA, et al. Thrombocytosis is associated with increased short and long term mortality after exacerbation of chronic obstructive pulmonary disease: a role for antiplatelet therapy? Thorax. 2014;69(7):609–615. doi: 10.1136/thoraxjnl-2013-203996.
- Short PM, Lipworth SIW, Elder DHJ, et al. Effect of beta blockers in treatment of chronic obstructive pulmonary disease: a retrospective cohort study. BMJ. 2011;342(2):d2549–d2549. doi: 10.1136/bmj.d2549.
- Søyseth V, Brekke PH, Smith P, et al. Statin use is associated with reduced mortality in COPD. Eur Respir J. 2007;29(2):279–283. doi: 10.1183/09031936.00106406.
- Pavasini R, Biscaglia S, d’Ascenzo F, et al. Antiplatelet treatment reduces all-cause mortality in COPD patients: a systematic review and meta-analysis. COPD. 2016;13(4):509–514. doi: 10.3109/15412555.2015.1099620.
- Sin DD. The devastating power of platelets in COPD exacerbations: can aspirin save lives in COPD? Thorax BMJ Publishing Group Ltd. 2014;69(7):603–604. doi: 10.1136/thoraxjnl-2014-205329.
- Bakshi A, Suissa S. Effectiveness of aspirin in COPD: biases in the observational studies. COPD. 2021;18(4):449–455. doi: 10.1080/15412555.2021.1963696.
- Stein BD, Bautista A, Schumock GT, et al. The validity of international classification of diseases, ninth revision, clinical modification diagnosis codes for identifying patients hospitalized for COPD exacerbations. Chest. 2012;141(1):87–93. doi: 10.1378/chest.11-0024.
- Rothnie KJ, Müllerová H, Smeeth L, et al. Natural history of chronic obstructive pulmonary disease exacerbations in a general practice-based population with chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 2018;198(4):464–471. doi: 10.1164/rccm.201710-2029OC.
- Rothnie KJ, Müllerová H, Hurst JR, et al. Validation of the recording of acute exacerbations of COPD in UK primary care electronic healthcare records. PloS One. 2016;11(3):e0151357. doi: 10.1371/journal.pone.0151357.
- Coates AL, Wong SL, Tremblay C, et al. Reference equations for spirometry in the Canadian population. Ann Am Thorac Soc. 2016;13(6):833–841. doi: 10.1513/AnnalsATS.201508-569OC.
- Suissa S. Immortal time bias in pharmaco-epidemiology. Am J Epidemiol. 2008;167(4):492–499. doi: 10.1093/aje/kwm324.
- Cole SR, Platt RW, Schisterman EF, et al. Illustrating bias due to conditioning on a collider. Int J Epidemiol. 2010;39(2):417–420. doi: 10.1093/ije/dyp334.
- McNeil JJ, Wolfe R, Woods RL, et al. Effect of aspirin on cardiovascular events and bleeding in the healthy elderly. N Engl J Med. 2018;379(16):1509–1518. doi: 10.1056/NEJMoa1805819.
- Agustí A, Edwards LD, Rennard SI, et al. Persistent systemic inflammation is associated with poor clinical outcomes in COPD: a novel phenotype. PLOS One. 2012;7(5):e37483. doi: 10.1371/journal.pone.0037483.
- Ikonomidis I, Andreotti F, Economou E, et al. Increased proinflammatory cytokines in patients with chronic stable angina and their reduction by aspirin. Circulation. 1999;100(8):793–798. doi: 10.1161/01.cir.100.8.793.
- Johns Hopkins University. Randomized trial to determine effective aspirin dose in COPD [Internet]. clinicaltrials.gov; 2023 MayReport No.: NCT05265299. https://clinicaltrials.gov/study/NCT05265299.