Abstract
Background
Real-world evidence concerning the cardiovascular (CV) risk of umeclidinium (UMEC) monotherapy or UMEC/vilanterol (VI) is scarce. This real-world study investigated CV risk in new users of UMEC or UMEC/VI versus new users of tiotropium (TIO) in patients with chronic obstructive pulmonary disease (COPD).
Methods
This prospective, observational, multinational, cohort study enrolled patients ≥18 years with COPD who initiated UMEC, UMEC/VI or TIO between 2 February 2016 and 31 January 2023. Noninferiority (95% confidence interval upper bound <2.0) of UMEC and UMEC/VI to TIO was compared for the time-to-first event (hazard ratio (HR) over 24 months) of a composite CV end-point of myocardial infarction, stroke, heart failure or sudden cardiac death. Stabilised inverse probability of treatment weighting adjusted for differences in baseline covariate balance between groups. Incidence rates for the composite CV end-point were calculated.
Results
In total, 6606 patients were enrolled; 6165 were included in the analysis. Both UMEC (n=1246) and UMEC/VI (n=2448) were noninferior to TIO (n=2471) for the risk of the composite CV end-point (adjusted HR (95% CI): UMEC versus TIO: 1.254 (0.830–1.896); UMEC/VI versus TIO: 1.352 (0.952–1.922)). Unadjusted composite CV incidence rates were low across cohorts (incidence rate (95% CI) per 100 person-years: UMEC: 1.157 (0.814–1.594); UMEC/VI: 1.287 (1.034–1.584); TIO: 0.924 (0.716–1.174)).
Conclusions
Both UMEC and UMEC/VI were noninferior to TIO for composite CV risk, suggesting that physicians may consider escalating patients to dual bronchodilator therapy if COPD symptoms are not effectively managed with monotherapy.
Shareable abstract
The prospective, observational study provides further evidence on the safety profile and tolerability of bronchodilator therapy for the treatment of patients with COPD. UMEC and UMEC/VI were noninferior to TIO for CV risk using a composite end-point. https://bit.ly/4oZZbG3
Introduction
The Global Initiative for Chronic Obstructive Lung Disease (GOLD) report recommends the use of inhaled long-acting bronchodilators for the maintenance treatment of COPD; either as long-acting muscarinic antagonist (LAMA) or long-acting β2-agonist (LABA) monotherapy or in combination as LAMA/LABA dual therapy, based on symptoms and exacerbation history [1].
Tiotropium (TIO) is an effective, well-tolerated LAMA used to treat COPD since 2002 [2–4]. Umeclidinium (UMEC), another LAMA, approved for clinical practice in 2013, is available alone or as a fixed-dose combination with the LABA vilanterol (VI) [5]. There are important pharmacologic differences between LAMAs, including muscarinic receptor affinities, which may influence efficacy and safety [6]. Although the safety and efficacy of UMEC and VI have been reported, most evidence comes from randomised controlled trials (RCTs) [7–9] with restrictive inclusion criteria. There is a need to further investigate the efficacy and safety of inhaled treatments in real-world studies [10].
Studies of the cardiovascular (CV) risk associated with long-acting bronchodilators in COPD have produced diverse results. One meta-analysis of RCTs reported no increased risk of CV events associated with LAMA versus placebo [11], although patients with high baseline CV risk may have been excluded from these studies [12]. In contrast, another meta-analysis of RCTs found LAMA/LABA combinations to be associated with an increased risk of CV events versus inhaled corticosteroid (ICS)/LABA therapy in COPD, although it is unclear whether this is due to a negative effect of the LAMA or a protective effect of ICS [13]. Given the high prevalence of CV comorbidities among patients with COPD [14, 15] and the mixed evidence for CV risk associated with long-acting bronchodilators [12], further real-world research is warranted. The aim of this prospective, real-world study was to examine CV risk among new users of UMEC monotherapy or UMEC/VI dual therapy compared with new users of TIO in patients with COPD. TIO, the first LAMA introduced into clinical practice [16], is the most widely used LAMA for treatment of COPD [17] and thus an appropriate comparator, as agreed by the European Medicines Agency (EMA) at the protocol stage.
Methods
Study design
This was a prospective, observational, multinational, real-world cohort study of patients with COPD who initiated treatment with UMEC, UMEC/VI or TIO (dry powder inhaler or soft mist inhaler) between 2 February 2016 and 31 January 2023 (figure 1). The study was carried out in nine European countries (Belgium, Czech Republic, Germany, Hungary, Italy, Netherlands, Poland, Spain and the United Kingdom) and the United States.
FIGURE 1.
Study design and follow-up schedule. #: The follow-up period was defined as the period between the prescription index date and the earliest of: 2 years after the date of the planned number of events reached, 14 days following the date of discontinuation of index COPD medication (if this was the reason for study discontinuation), withdrawal from the study, conclusion of study follow-up, or death. HCP: healthcare provider; TIO: tiotropium; UMEC: umeclidinium; VI: vilanterol.
Patients
Eligible patients had a clinical diagnosis of COPD in accordance with GOLD recommendations at the time (2019 [18] and 2023 [19]) and verified by spirometry (forced expiratory volume in 1 s (FEV1)/forced vital capacity (FVC) fixed ratio <70%), initiated UMEC, UMEC/VI or TIO treatment, were ≥18 years of age, willing and able to provide written informed consent, and had ≥12 months of medical records prior to enrolment. Patients were excluded if currently participating in any interventional clinical trials in which treatment regimen and/or monitoring was dictated by a protocol or if they had hypersensitivity to UMEC, VI, TIO or excipients. No restrictions were placed on concomitant medication; however, patients were excluded if they received maintenance treatment (≥60 days continuous use) with LAMA-containing medication during the 12 months prior to enrolment.
Procedures
The index date was defined as the date of a new UMEC, UMEC/VI or TIO prescription (either at the enrolment visit or up to 7 days prior), prescribed independently by the treating physician. The baseline period covered the 12 months prior to and including the index date. The exposure period ran from the index date until 14 days post-discontinuation of study medication or until switch to another COPD medication, study discontinuation, death, site closure or study last participant last visit date, whichever came first. Patients were observed from the index date until the earliest of ≥24 months of follow-up, end of the exposure period, withdrawal from the study or death. Data were collected during routine (twice-yearly) or unscheduled visits to the treating physician and from patients’ pre-enrolment medical history. Patients without physician contact for a 6-month period were contacted directly to collect minimal safety information, provided this was considered standard care by the treating physician.
Outcomes
The primary outcomes of the study were 1) noninferiority of UMEC and UMEC/VI to TIO for the risk of a composite end-point including myocardial infarction (MI), stroke, heart failure or sudden cardiac death, measured by the hazard ratio (HR) over 24 months (time-to-first event analysis), and 2) incidence rate of the composite end-point (number of first events per 100 person-years (PY)). The secondary outcomes included 1) UMEC and UMEC/VI versus TIO for the risk of an individual component of the composite (MI, stroke or heart failure), measured by the HR over 24 months, 2) incidence rate of each MI, stroke and heart failure end-point (number of first events per 100 PY), and 3) rates of overall mortality, CV mortality and non-CV mortality (number of deaths per 100 PY). Data for these outcomes were collected via electronic case report forms and regular monitoring.
Statistical analysis
Sample size calculations were applied separately to each treatment comparison for the primary end-point (time-to-first event analysis) to provide adequate power to demonstrate noninferiority of UMEC/VI or UMEC, relative to TIO [20]. Assuming a one-sided alpha of 2.5%, 90% power, a noninferiority margin of 2.0 and 1:1 ratio of treatment groups compared, it was estimated that 98 patients with adjudicated events were required per comparison. Patients were followed until the required number with adjudicated events for the primary end-point was observed. Assuming an event rate of 0.89 per 100 PY for the composite end-point and a mean follow-up time of 2.5 years, an estimated 2233 patients per treatment cohort (approximately 6700 total) was required for the primary end-point analysis.
Demographic and clinical characteristics collected during the baseline period are reported descriptively. Details of initiated COPD treatment with UMEC, UMEC/VI or TIO including duration of prescription and dose were collected at baseline. Factors associated with treatment choice (UMEC, UMEC/VI or TIO) or with any of the study outcomes of interest (including completeness of information) were measured at enrolment and balanced through propensity score (PS) methods. These factors include indication of use (treatment choice associated with severity of disease), prescribing pattern differences across practices and countries and reimbursement differences.
Covariate balance between cohorts was assessed using average standardised absolute mean difference. To increase the reliability of the treatment effect variance estimate and minimise the influence of high weights, stabilised inverse probability of treatment weighting was applied, with weights truncated at 10 to limit the impact of extreme values, as previously described [21].
Time-to-first event for the composite end-point (primary) and individual end-points (secondary) were analysed using weighted Cox proportional hazards models, comparing UMEC and UMEC/VI initiators with TIO initiators. Adjusted HRs and 95% confidence intervals were calculated for each treatment comparison. As agreed with the EMA's Pharmacovigilance Risk Assessment Committee, if the upper bound of the 95% confidence interval for the HR exceeded 2.0, the noninferiority assumption was rejected; if the lower bound of the 95% confidence interval exceeded 1.0, noninferiority was not assumed. Kaplan–Meier survival curves comparing UMEC and UMEC/VI initiators with TIO initiators for the composite CV end-point were also generated. As this study was powered only for the primary end-point only, analyses of the secondary end-points were considered descriptive.
Only MI, stroke, heart failure and mortality events confirmed by independent adjudication using pre-defined, standardised criteria during follow-up were included.
PS for each treatment comparison were estimated using multivariable logistic regression models. Confounders and effect modifiers of primary outcomes included in the PS model included site characteristics, physician specialty, patient demographics, clinical assessments, COPD severity, smoking and alcohol history CV and cerebrovascular diagnosis history, other comorbidities, family history, and concomitant medications. In general, missing data were not imputed and data were analysed as recorded in the study electronic case report forms. Further detail on statistical analysis is provided in the supplementary materials.
Results
Study population
A total of 6606 patients were enrolled in this study, of whom 6165 were included in the full analysis set (FAS) population. Of the FAS population, 1246 patients were included in the UMEC cohort, 2448 in the UMEC/VI cohort and 2471 in the TIO cohort (supplementary table 1).
Baseline demographic and clinical characteristics
Baseline demographic and clinical characteristics, treatment patterns and medical histories for the FAS population are reported in table 1 and supplementary tables 2, 3 and 4, respectively. The mean age (66.0–66.6 years) and proportion of males (56.7–62.4%) was similar across cohorts. The total proportion of patients from Eastern European countries (Czech Republic, Hungary and Poland) was greater in the UMEC/VI cohort (48.2%) than UMEC (38.7%) and TIO (23.5%) cohorts. In Spain, the UK and US, more patients initiated TIO than UMEC or UMEC/VI. A higher proportion of patients in the UMEC (74.2%) and UMEC/VI (75.9%) cohorts were cared for by pulmonologists than in the TIO cohort (50.8%). The mean (sd) time from spirometry to enrolment was 84.7 (322.56) days in the UMEC cohort, 74.4 (277.72) days in the UMEC/VI cohort and 190.5 (552.69) days in the TIO cohort. Using GOLD 2019 classifications, most patients were classed as Grade 2 and Group B across cohorts, and this was consistent with GOLD 2023 classification. The rate per PY of moderate/severe COPD exacerbations in the prior 12 months was similar across cohorts (0.220–0.263). Mean modified Medical Research Council Dyspnoea Scale (1.5–1.6) and COPD Assessment Test (CAT) (15.9–17.4) scores were also similar across cohorts. In total, 23.4% of patients in the UMEC cohort, 10.0% in the UMEC/VI cohort and 25.3% in the TIO cohort were taking ICS/LABA fixed combination therapies in the 12 months pre-index.
TABLE 1.
Baseline demographics and clinical characteristics in the full analysis set (FAS) population
| UMEC (N=1246) | UMEC/VI (N=2448) | TIO (N=2471) | Total (N=6165) | Standard difference# UMEC versus TIO | Standard difference# UMEC/VI versus TIO | |
|---|---|---|---|---|---|---|
| Country, n | 1246 | 2448 | 2471 | 6165 | 0.16 | 0.19 |
| Belgium, n (%) | 22 (1.8) | 27 (1.1) | 11 (0.4) | 60 (1.0) | ||
| Czech Republic, n (%) | 155 (12.4) | 359 (14.7) | 232 (9.4) | 746 (12.1) | ||
| Germany, n (%) | 206 (16.5) | 448 (18.3) | 309 (12.5) | 963 (15.6) | ||
| Hungary, n (%) | 161 (12.9) | 126 (5.1) | 114 (4.6) | 401 (6.5) | ||
| Italy, n (%) | 71 (5.7) | 82 (3.3) | 52 (2.1) | 205 (3.3) | ||
| Netherlands, n (%) | 32 (2.6) | 82 (3.3) | 82 (3.3) | 196 (3.2) | ||
| Poland, n (%) | 166 (13.3) | 694 (28.3) | 234 (9.5) | 1094 (17.7) | ||
| Spain, n (%) | 86 (6.9) | 130 (5.3) | 213 (8.6) | 429 (7.0) | ||
| UK, n (%) | 217 (17.4) | 156 (6.4) | 531 (21.5) | 904 (14.7) | ||
| USA, n (%) | 130 (10.4) | 344 (14.1) | 693 (28.0) | 1167 (18.9) | ||
| Age (years), n | 1246 | 2448 | 2471 | 6165 | 0.05 | 0.05 |
| Mean (sd) | 66.0 (9.91) | 66.0 (9.38) | 66.6 (10.08) | 66.2 (9.77) | ||
| Gender, n | 1246 | 2448 | 2471 | 6165 | 0.03 | 0.08 |
| Male, n (%) | 706 (56.7) | 1527 (62.4) | 1439 (58.2) | 3672 (59.6) | ||
| Physician specialty, n | 1246 | 2448 | 2471 | 6165 | 0.36 | 0.42 |
| Pulmonologist, n (%) | 925 (74.2) | 1857 (75.9) | 1256 (50.8) | 4038 (65.5) | ||
| Primary care physician, n (%) | 302 (24.2) | 527 (21.5) | 1197 (48.4) | 2026 (32.9) | ||
| Other, n (%) | 19 (1.5) | 64 (2.6) | 18 (0.7) | 101 (1.6) | ||
| BMI, n | 1171 | 2387 | 2396 | 5954 | 0.04 | <0.01 |
| Mean (sd) | 27.6 (5.95) | 27.8 (6.11) | 27.8 (5.91) | 27.8 (6.00) | ||
| Systolic blood pressure (mmHg), n | 1043 | 2173 | 2260 | 5476 | 0.09 | 0.14 |
| Mean (sd) | 132.8 (15.29) | 133.7 (15.01) | 131.5 (15.52) | 132.6 (15.30) | ||
| Diastolic blood pressure (mmHg), n | 1043 | 2173 | 2260 | 5476 | 0.08 | 0.07 |
| Mean (sd) | 78.8 (9.11) | 78.7 (9.48) | 78.0 (9.42) | 78.5 (9.39) | ||
| NYHA heart failure class, n | 1246 | 2448 | 2471 | 6165 | 0.17 | 0.12 |
| I, n (%) | 493 (35.2) | 844 (34.5) | 1059 (42.9) | 2342 (38.0) | ||
| II, n (%) | 138 (11.1) | 390 (15.9) | 466 (18.9) | 994 (16.1) | ||
| III, n (%) | 49 (3.9) | 108 (4.4) | 144 (5.8) | 301 (4.9) | ||
| IV, n (%) | 4 (0.3) | 6 (0.2) | 11 (0.4) | 21 (0.3) | ||
| Not assessed, n (%) | 616 (49.4) | 1100 (44.9) | 791 (32.0) | 2507 (40.7) | ||
| Pre-bronchodilator | ||||||
| FEV1 % pred, n | 282 | 392 | 439 | 1113 | 0.24 | 0.05 |
| Mean±(sd) | 62.0 (16.54) | 57.0 (17.26) | 57.9 (16.87) | 58.6 (17.03) | ||
| FEV1/FVC, n | 283 | 398 | 435 | 1116 | 0.27 | 0.02 |
| Mean (sd) | 0.61 (0.099) | 0.58 (0.113) | 0.58 (0.108) | 0.58 (0.108) | ||
| Post bronchodilator | ||||||
| FEV1 % pred, n | 819 | 1879 | 1846 | 4544 | 0.32 | 0.09 |
| Mean (sd) | 64.4 (17.68) | 57.6 (15.09) | 59.1 (16.57) | 59.4 (16.36) | ||
| FEV1/FVC, n | 877 | 1949 | 1905 | 4731 | 0.13 | 0.13 |
| Mean (sd) | 0.59 (0.096) | 0.56 (0.104) | 0.57 (0.107) | 0.57 (0.104) | ||
| Time from spirometry to enrolment (days),¶ n | 1240 | 2430 | 2469 | 6139 | 0.19 | 0.21 |
| Mean (sd) | 84.7 (322.56) | 74.4 (277.72) | 190.5 (552.69) | 123.3 (421.20) | ||
| Years since COPD diagnosis, n | 1240 | 2427 | 2465 | 6132 | 0.13 | 0.08 |
| Mean (sd) | 4.1 (6.22) | 4.4 (6.41) | 4.9 (6.28) | 4.5 (6.33) | ||
| Age at COPD diagnosis, n | 1240 | 2427 | 2465 | 6132 | 0.02 | <0.01 |
| Mean (sd) | 61.9 (10.78) | 61.7 (10.59) | 61.7 (10.77) | 61.7 (10.70) | ||
| GOLD 2019 classification, n | 554 | 982 | 990 | 2526 | ||
| Grade 1, n (%) | 108 (19.5) | 102 (10.4) | 150 (15.2) | 360 (14.3) | 0.08 | 0.08 |
| Grade 2, n (%) | 428 (77.3) | 836 (85.1) | 810 (81.8) | 2074 (82.1) | ||
| Grade 3, n (%) | 13 (2.3) | 40 (4.1) | 27 (2.7) | 80 (3.2) | ||
| Grade 4, n (%) | 5 (0.9) | 4 (0.4) | 3 (0.3) | 12 (0.5) | ||
| Grade A, n (%) | 182 (32.9) | 233 (23.7) | 291 (29.4) | 706 (27.9) | 0.05 | 0.09 |
| Grade B, n (%) | 354 (63.9) | 705 (71.8) | 669 (67.6) | 1728 (68.4) | ||
| Grade C, n (%) | 1 (0.2) | 7 (0.7) | 1 (0.1) | 9 (0.4) | ||
| Grade D, n (%) | 17 (3.1) | 37 (3.8) | 29 (2.9) | 83 (3.3) | ||
| GOLD 2023 classification, n | 554 | 982 | 990 | 2526 | ||
| Grade A, n (%) | 182 (32.9) | 233 (23.7) | 291 (29.4) | 706 (27.9) | 0.05 | 0.10 |
| Grade B, n (%) | 354 (63.9) | 705 (71.8) | 669 (67.6) | 1728 (68.4) | ||
| Grade E, n (%) | 18 (3.2) | 44 (4.5) | 30 (3.0) | 92 (3.6) | ||
| Number of moderate/severe COPD exacerbations,+ prior 12 months, n | 1246 | 2448 | 2471 | 6165 | 0.05 | 0.04 |
| 0 events, n (%) | 982 (78.8) | 2007 (82.0) | 2002 (81.0) | 4991 (81.0) | ||
| ≥1 event | 264 (21.2) | 441 (18.0) | 469 (19.0) | 1174 (19.0) | ||
| Total number of events | 328 | 539 | 626 | 1493 | ||
| Rate per person-year (95% CI) | 0.263 (0.236–0.293) | 0.220 (0.202–0.240) | 0.253 (0.234–0.274) | 0.242 (0.230–0.255) | ||
| Number of COPD exacerbation-related hospitalisations, prior 12 months, n | 1246 | 2448 | 2471 | 6165 | 0.03 | 0.04 |
| 0, n (%) | 1190 (95.5) | 2323 (94.9) | 2381 (96.4) | 5894 (95.6) | ||
| ≥1, n (%) | 56 (4.5) | 125 (5.1) | 90 (3.6) | 271 (4.4) | ||
| Total number of hospitalisations | 60 | 139 | 101 | 300 | ||
| Rate per person-year (95% CI) | 0.048 (0.037–0.062) | 0.057 (0.048–0.067) | 0.041 (0.034–0.050) | 0.049 (0.044–0.055) | ||
| Smoking status at enrolment, n | 1246 | 2447 | 2470 | 6163 | 0.12 | 0.05 |
| Never, n (%) | 72 (5.8) | 224 (9.2) | 301 (12.2) | 597 (9.7) | ||
| Current, n (%) | 608 (48.8) | 1136 (46.4) | 1112 (45.0) | 2856 (46.3) | ||
| Former, n (%) | 566 (45.4) | 1087 (44.4) | 1057 (42.8) | 2710 (44.0) | ||
| mMRC, n | 1235 | 2436 | 2464 | 6135 | 0.04 | 0.05 |
| Mean (sd) | 1.5 (0.91) | 1.6 (0.88) | 1.5 (0.89) | 1.6 (0.89) | ||
| CAT, n | 1236 | 2430 | 2459 | 6125 | 0.20 | <0.01 |
| Mean (sd) | 15.9 (7.42) | 17.4 (7.57) | 17.4 (7.61) | 17.1 (7.58) |
BMI: body mass index; CAT: COPD Assessment Test; FEV1: forced expiratory volume in 1 s; FVC: forced vital capacity; GOLD: Global Initiative for Chronic Obstructive Lung Disease; mMRC: modified Medical Research Council; NYHA: New York Heart Association; TIO: tiotropium; UMEC: umeclidinium; VI: vilanterol. #: The standardised difference between those exposed to UMEC versus TIO or those exposed to UMEC/VI versus TIO. This was calculated as raw difference between groups/sd of the specific measure in the TIO group. ¶: Not all participants with known time from spirometry to enrolment had all spirometry results reported; +: COPD exacerbations requiring treatment with antibiotics, systemic steroids and/or hospitalisation.
Exposure to study medication
Study medication was discontinued by almost half of patients in the UMEC cohort (49.2%) and by over a third of patients in the UMEC/VI (36.9%) and TIO cohorts (35.0%) (supplementary table 5). The mean±sd duration of exposure to study medication was 950.5±615.85 days in the UMEC cohort, 1049.3±593.91 days in the UMEC/VI cohort and 1084.9±596.93 days in the TIO cohort (supplementary table 5). In the UMEC/VI cohort, the most common switch was to ICS/LABA (single or multiple inhalers; 7.9%). UMEC/VI was the most common switch for the UMEC cohort (13.0%) and LAMA/LABA (single or multiple inhalers; other than UMEC/VI) was most common for the TIO cohort (7.2%) (supplementary table 6).
Primary outcomes
For the primary end-point of risk of a composite event during the exposure period in the FAS population, the adjusted HR (95% CI) was 1.254 (0.830–1.896) for UMEC versus TIO and 1.352 (0.952–1.922) for UMEC/VI versus TIO (figure 2a); as the upper bound of the 95% confidence interval was below 2 in both comparisons, both UMEC and UMEC/VI demonstrated noninferiority versus TIO. Using a Kaplan–Meier survival function to assess time-to-first event for the composite end-point, a similar number of patients were at risk at all time-points for both UMEC versus TIO and UMEC/VI versus TIO (figure 3). The proportion of patients who had ≥1 composite event during the exposure period was low across all cohorts (UMEC: 3.0%; UMEC/VI: 3.6%; TIO: 2.7%). The corresponding unadjusted incidence rates (95% CI) per 100 PY for a composite event were also low across cohorts (UMEC: 1.157 (0.814–1.594); UMEC/VI: 1.287 (1.034–1.584); TIO: 0.924 (0.716–1.174)) (table 2).
FIGURE 2.
Risk of a) the composite end-point and b) confirmed myocardial infarction (MI), stroke and heart failure with umeclidinium (UMEC) or UMEC/vilanterol (VI) versus tiotropium (TIO) during the exposure period in the full analysis set population. Composite end-point includes MI, stroke, heart failure or sudden cardiac death. HR: hazard ratio.
FIGURE 3.
Survival function for the composite end-point in those initiated on a) umeclidinium (UMEC)/vilanterol (VI) or b) UMEC versus tiotropium (TIO) during the exposure period in the full analysis set population. #: Number of patients at risk/number of first events.
TABLE 2.
Incidence rate of the composite end-point during the exposure period in the full analysis set (FAS) population
| Composite outcome | UMEC (N=1246) | UMEC/VI (N=2448) | TIO (N=2471) | Total (N=6165) |
|---|---|---|---|---|
| Incidence rate #, ¶ | ||||
| Participants, n (%) | 37 (3.0) | 89 (3.6) | 67 (2.7) | 193 (3.1) |
| Rate per 100 person-years (95% CI) | 1.157 (0.814–1.594) | 1.287 (1.034–1.584) | 0.924 (0.716–1.174) | 1.112 (0.960–1.280) |
| Event rate#, + | ||||
| Number of events | 44 | 105 | 85 | 234 |
| Rate per 100 person-years (95% CI) | 1.357 (0.986–1.822) | 1.493 (1.221–1.807) | 1.158 (0.925–1.432) | 1.328 (1.164–1.510) |
TIO: tiotropium; UMEC: umeclidinium; VI: vilanterol. #: The composite outcome comprised any confirmed myocardial infarction, stroke, heart failure or sudden cardiac death. ¶: Only the first event of a given outcome was included in the counts and time after the first event was not included in person-time totals. +: Multiple events of each outcome except death were included in counts.
Secondary outcomes
The risk (adjusted HRs over 24 months) for the individual components of the composite are presented in figure 2b. The adjusted HR (95% CI) for MI was 1.754 (0.748–4.115) for UMEC versus TIO and 2.195 (1.053–4.575) for UMEC/VI versus TIO; the adjusted HR (95% CI) for stroke was 1.096 (0.458–2.621) for UMEC versus TIO and 1.018 (0.470–2.207) for UMEC/VI versus TIO; and the adjusted HR (95% CI) for heart failure was 1.287 (0.654–2.532) for UMEC versus TIO and 0.832 (0.459–1.509) for UMEC/VI versus TIO (figure 2b). Overall, low frequencies and incidences of MI, stroke and heart failure were observed across all cohorts during the exposure period (unadjusted data) (table 3); no cases of MI in the UMEC/VI cohort were considered related to study treatment by the investigator.
TABLE 3.
Incidence and event rates for confirmed myocardial infarction (MI), stroke and heart failure for umeclidinium (UMEC), UMEC/vilanterol (VI) and tiotropium (TIO) during the exposure period in the full analysis set population
| UMEC (N=1246) | UMEC/VI (N=2448) | TIO (N=2471) | |
|---|---|---|---|
| MI | |||
| Participants, n (%) | |||
| No events | 1237 (99.3) | 2422 (98.9) | 2456 (99.4) |
| 1 event | 9 (0.7) | 26 (1.1) | 14 (0.6) |
| ≥2 events | 0 (0.0) | 0 (0.0) | 1 (<0.1) |
| Incidence rate | |||
| Number of first events | 9 (0.7) | 26 (1.1) | 15 (0.6) |
| Incidence rate per 100 person-years (95% CI)# | 0.28 (0.127–0.529) | 0.37 (0.243–0.546) | 0.21 (0.115–0.338) |
| Incidence rate ratio (95% CI), unadjusted | 1.36 (0.594–3.103) | 1.82 (0.962–3.427) | Reference |
| Event rate | |||
| Total number of events | 9 | 26 | 18 |
| Event rate per 100 person-years (95% CI)¶ | 0.28 (0.127–0.527) | 0.37 (0.242–0.542) | 0.25 (0.145–0.388) |
| Event rate ratio (95% CI), unadjusted | 1.13 (0.508–2.519) | 1.51 (0.827–2.749) | Reference |
| Stroke | |||
| Participants, n (%) | |||
| No events | 1239 (99.4) | 2431 (99.3) | 2453 (99.3) |
| 1 event | 7 (0.6) | 16 (0.7) | 17 (0.7) |
| ≥2 events | 0 (0.0) | 1 (<0.1) | 1 (<0.1) |
| Incidence rate | |||
| Number of first events | 7 (0.6) | 17 (0.7) | 18 (0.7) |
| Incidence rate per 100 person-years (95% CI) # | 0.22 (0.087–0.446) | 0.24 (0.141–0.388) | 0.25 (0.146–0.389) |
| Incidence rate ratio (95% CI), unadjusted | 0.88 (0.367–2.105) | 0.99 (0.508–1.911) | Reference |
| Event rate | |||
| Total number of events | 7 | 18 | 19 |
| Event rate per 100 person-years (95% CI)¶ | 0.22 (0.087–0.445) | 0.26 (0.152–0.405) | 0.26 (0.156–0.404) |
| Event rate ratio (95% CI), unadjusted | 0.83 (0.351–1.984) | 0.99 (0.519–1.884) | Reference |
| Heart failure | |||
| Participants, n (%) | |||
| No events | 1228 (98.6) | 2424 (99.0) | 2445 (98.9) |
| 1 event | 14 (1.1) | 18 (0.7) | 18 (0.7) |
| ≥2 events | 4 (0.3) | 6 (0.2) | 8 (0.3) |
| Incidence rate | |||
| Number of first events | 18 (1.4) | 24 (1.0) | 26 (1.1) |
| Incidence rate per 100 person-years (95% CI)# | 0.56 (0.332–0.884) | 0.34 (0.220–0.511) | 0.36 (0.233–0.522) |
| Incidence rate ratio (95% CI), unadjusted | 1.57 (0.861–2.866) | 0.97 (0.554–1.681) | Reference |
| Event rate | |||
| Total number of events | 23 | 36 | 37 |
| Event rate per 100 person-years (95% CI)¶ | 0.71 (0.450–1.064) | 0.51 (0.359–0.709) | 0.50 (0.355–0.695) |
| Event rate ratio (95% CI), unadjusted | 1.41 (0.836–2.368) | 1.02 (0.642–1.607) | Reference |
#: Calculated as number of first events per 100 person-years. ¶: Calculated as total number of events per 100 person-years.
A total of 291 deaths occurred during the exposure period (UMEC: n=41 (3.3%); UMEC/VI: n=143 (5.8%); TIO: n=107 (4.3%)); unadjusted overall mortality rates (95% CI) per 100 PY were 1.264 (0.907–1.715) in the UMEC cohort, 2.033 (1.714–2.395) in the UMEC/VI cohort and 1.458 (1.195–1.762) in the TIO cohort (supplementary table 7). Unadjusted sudden CV-related mortality rates (95% CI) per 100 PY were 0.154 (0.050–0.360) in the UMEC cohort, 0.355 (0.230–0.525) in the UMEC/VI cohort and 0.150 (0.075–0.268) in the TIO cohort. Unadjusted nonsudden CV-related mortality rates (95% CI) per 100 PY were 0.216 (0.087–0.445) in the UMEC cohort, 0.569 (0.406–0.775) in the UMEC/VI cohort and 0.395 (0.265–0.567) in the TIO cohort. Additionally, unadjusted non-CV-related mortality rates were 0.894 (0.599–1.284) in the UMEC cohort, 1.109 (0.877–1.384) in the UMEC/VI cohort and 0.913 (0.707–1.159) in the TIO cohort (supplementary table 7). Similar results were seen during the observation period (supplementary table 8). A similar proportion of patients experienced CV and cerebrovascular adverse events of special interest across all cohorts during the exposure period (supplementary table 9).
Discussion
This prospective, observational, multinational, real-world cohort study demonstrated that both UMEC and UMEC/VI were noninferior to TIO for the risk of a composite end-point of MI, stroke, heart failure or sudden cardiac death, in patients with COPD after initiating treatment. Overall, incidence and event rates for the composite end-point, individual CV end-points and mortality were low across all cohorts.
There are mechanisms by which dual long-acting bronchodilator therapy can improve CV function, for example, by reducing gas trapping, which can decrease ventricular compression [22, 23]. Furthermore, as treatment with LAMA/LABA reduces COPD exacerbations, which are in turn associated with an enhanced risk of CV events, this may reduce CV risk in patients with COPD [24]. However, the systemic exposure of long-acting bronchodilators may increase the risk of arrythmias, which may increase CV risk with LAMA/LABA compared with other COPD treatments, including monotherapy [13, 25]. Some studies have shown either no increased CV risk or reduced CV risk with LAMA/LABA versus monotherapy [26–30], varying between molecules. Meta-analyses have not found a significant difference between the risk of CV adverse events with UMEC/VI versus UMEC monotherapy [29, 30], though a more recent database analysis did identify significantly fewer CV adverse events with UMEC/VI compared with UMEC alone [28]. The current noninferiority study did not suggest a protective effect of UMEC/VI on CV events versus UMEC, although these treatments were not directly compared. Though the adjusted HR values suggest greater risk of CV events with UMEC-containing therapies versus TIO, aligning with the findings of the recent database study [28], noninferiority was demonstrated and statistically significant differences in CV risk were not identified.
The secondary end-point analyses showed conflicting results. The risk of stroke was similar across all cohorts; however, in the UMEC/VI cohort, the risk of heart failure was lower but the risk of MI was greater compared with the TIO cohort. Given that no cases of MI in the UMEC/VI cohort were deemed to be related to study treatment, and the previously reported relationship between COPD severity and CV events [31], these data may reflect that patients in this cohort had more severe COPD compared with the UMEC and TIO cohorts. In previously published comparisons of the safety profiles of UMEC/VI and TIO, CV risk was similar between treatments [32, 33], including incidences of stroke and heart failure [34]. However, as CV events reported often differ between clinical trials, direct comparisons between studies must be interpreted with caution. The use of a standard composite CV risk end-point, as presented here, may help improve the comparability of safety outcomes in future long-term studies. The overall number of events was generally low for all cohorts and the study was only powered to assess the primary composite end-point.
The baseline characteristics of the UMEC/VI cohort suggest that these patients had more severe disease than patients in the TIO cohort. Patients in the UMEC/VI cohort had a substantially lower mean time from spirometry to enrolment compared with the TIO cohort, suggesting a more recent contact with healthcare providers and potentially a greater need for regular monitoring. However, as almost double the proportion of TIO users were US-based versus UMEC/VI users (28% versus 14%), this difference may be influenced by the underutilisation of spirometry in the US [35]. Additionally, over 75% of patients in the UMEC/VI cohort had pulmonologists as their primary prescribing physician, whereas in the TIO cohort, the treating physician speciality was split between pulmonologists (50.8%) and primary care practitioners (48.4%). The proportion of current and former smokers in the UMEC/VI cohort was numerically greater than in the TIO cohort, and a greater proportion of patients initiating UMEC/VI were classified at higher GOLD grades than TIO initiators. This further suggests that patients initiating UMEC/VI had more severe COPD at baseline than patients initiating TIO, potentially influencing CV and mortality risk. We also found greater proportions of UMEC/VI initiators than TIO initiators in the Eastern European countries of the Czech Republic, Hungary and Poland; countries that are known to have high rates of CV disease [36]. The geographic profiles of these cohorts may therefore be a contributing factor to the increased risk of MI and CV-related mortality observed in the UMEC/VI versus TIO cohort reflecting regional differences in CV burden and other environmental factors [37].
The duration of exposure to study medication was lowest in the UMEC cohort, which also had a higher proportion discontinuing or switching treatment than the UMEC/VI or TIO cohorts. Although reasons for study discontinuation varied, adverse events or a lack of efficacy were more common in both the UMEC and UMEC/VI cohorts than the TIO cohort, likely reflecting greater baseline disease burden. The most common first switch among UMEC initiators was to UMEC/VI, suggesting a need to escalate treatment [38]. In the UMEC/VI cohort, the most common first switch was to ICS/LABA, possibly reflecting physicians’ efforts to reduce exacerbations risk [39, 40], consistent with GOLD 2019 recommendations to consider ICS/LABA for patients with exacerbations not controlled with long-acting bronchodilators [18].
There were several limitations to this study. This study used data from patients’ medical charts and although measures were taken to limit bias and confounding, including using PS methods and other multivariate analyses, inherent limitations of observational studies remain. Centres were expected to enrol prospectively all eligible patients, but healthcare settings (primary versus secondary care) and country of origin may have influenced prescribing patterns. Due to the pragmatic real-world study design, factors affecting treatment choice, such as COPD severity or device type, were not controlled, though partially addressed through PS weighting. Patients with more severe COPD were more likely to be initiated on LAMA/LABA than LAMA alone, per GOLD treatment recommendations [1]. Patients lost to follow-up may differ from those completing the study, despite efforts to follow-up directly with patients even if they did not return to the enrolling centre. Missing data limited the number of variables in the PS model and, despite standardised instructions, baseline information may not have been collected uniformly. Finally, although centres and country selection aimed to reflect patients initiating COPD treatment, these patients may differ from the broader population of patients already receiving long-term treatment.
Conclusion
This prospective, observational, real-world study provides further evidence to support the safety profile and tolerability of UMEC/VI dual bronchodilator therapy for the treatment of patients with COPD. Both UMEC and UMEC/VI were noninferior to TIO for CV risk using a composite end-point including MI, stroke, heart failure and sudden cardiac death. These real-world results support escalation to dual bronchodilator therapy if COPD symptoms are not effectively managed with LAMA alone.
Acknowledgements
D. Singh is supported by the National Institute for Health Research (NIHR) Manchester Biomedical Research Centre (BRC). The authors wish to acknowledge Ewa Nowicka (CEP safety (SERM), GSK, London, UK) for their contributions to data acquisition and manuscript development. Editorial support, in the form of preparation of the first draft based on input from all authors, and collation and incorporation of author feedback to develop subsequent drafts, was provided by Jen Rouine of Fishawack Indicia Ltd, UK, part of Avalere Health, and was funded by GSK.
Footnotes
Provenance: Submitted article, peer reviewed.
Ethics statement: This study was conducted under the Guidelines for Good Pharmacovigilance Practices and Good Pharmacoepidemiology Practices (International Society for Pharmacoepidemiology), the Declaration of Helsinki, and applicable national guidelines. Written informed consent was obtained from all patients prior to participation and all patient data were anonymised.
Author contributions: All authors had full access to the data and participated in data analysis/interpretation and writing of the report. The corresponding author had final responsibility to submit for publication. D. Singh and C. Compton contributed to the study concept or design and data interpretation. A.R. Sousa and G. Requena contributed to the study concept or design, data acquisition, analysis and interpretation. D.E. Newby, M. Jonsson Funk, M. Roman-Rodriguez and P. Kardos contributed to data interpretation. A. Donald contributed to data analysis and interpretation. D. Slade contributed to data acquisition, analysis and interpretation. All authors were involved in preparation and review of the manuscript and approved the final version to be submitted. All authors take complete responsibility for the integrity of the data and accuracy of the data analysis.
Conflict of interest: D. Singh declares consulting fees from Aerogen, AstraZeneca, Boehringer Ingelheim, Chiesi, Cipla, CSL Behring, Epiendo, Genentech, Glenmark, GSK, Gossamerbio, Kinaset, Menarini, Novartis, Pulmatrix, Sanofi, Synairgen, Teva, Theravance and Verona. A.R. Sousa, A. Donald and D. Slade are employed by GSK and hold financial equities in GSK. D.E. Newby declares consulting fees from GSK. M. Jonsson Funk receives salary support as Director of the Center for Pharmacoepidemiology at UNC, which has cooperative agreements with GSK, AbbVie, Astellas, Boehringer Ingelheim, Sarepta, Takeda and UCB. M. Roman-Rodriguez declares consulting fees from AstraZeneca, Boehringer Ingelheim and GSK. P. Kardos declares consulting fees from AstraZeneca, Chiesi, GSK, Menarini, Novartis and Sanofi-Genentech. G. Requena was employed by GSK at the time of the study. C. Compton was employed by GSK at the time of the study and holds financial equities in GSK.
Support statement: This study was funded by GSK (GSK Study 201038). The funder of the study had a role in study design, data collection, data analysis, data interpretation and writing of the report. Funding information for this article has been deposited with the Open Funder Registry.
Supplementary material
Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.
Supplementary material
01691-2025.SUPPLEMENT
Data availability
Please refer to GSK weblink to access GSK's data sharing policies and as applicable seek anonymised subject level data via the link https://www.gsk-studyregister.com/en/.
References
- 1.Global Initiative for Chronic Obstructive Lung Disease . Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease. Date last accessed: 6 February 2025. Date last updated: 15 November 2024. https://goldcopd.org/wp-content/uploads/2024/11/GOLD-2025-Report-v1.0-15Nov2024_WMV.pdf
- 2.Bateman E, Singh D, Smith D, et al. Efficacy and safety of tiotropium Respimat SMI in COPD in two 1-year randomized studies. Int J Chron Obstruct Pulmon Dis 2010; 5: 197–208. [PMC free article] [PubMed] [Google Scholar]
- 3.Kesten S, Celli B, Decramer M, et al. Tiotropium HandiHaler in the treatment of COPD: a safety review. Int J Chron Obstruct Pulmon Dis 2009; 4: 397–409. doi: 10.2147/COPD.S4802 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Boehringer Ingelheim . Spiriva RESPIMAT prescribing information. Date last accessed: 28 October 2024. Date last updated: January 2025. https://content.boehringer-ingelheim.com/DAM/68a8a6b5-4e9a-4508-85d3-af1e01205009/spiriva%20respimat-us-pi.pdf
- 5.GSK . Incruse ELLIPTA prescribing information. Date last accessed: 28 October 2024. Date last updated: December 2023. https://gskpro.com/content/dam/global/hcpportal/en_US/Prescribing_Information/Incruse_Ellipta/pdf/INCRUSE-ELLIPTA-PI-PIL-IFU.PDF
- 6.Babu KS, Morjaria JB. Umeclidinium in chronic obstructive pulmonary disease: latest evidence and place in therapy. Ther Adv Chronic Dis 2017; 8: 81–91. doi: 10.1177/2040622317700822 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Trivedi R, Richard N, Mehta R, et al. Umeclidinium in patients with COPD: a randomised, placebo-controlled study. Eur Respir J 2014; 43: 72–81. doi: 10.1183/09031936.00033213 [DOI] [PubMed] [Google Scholar]
- 8.Yamagata E, Soutome T, Hashimoto K, et al. Long-term (52 weeks) safety and tolerability of umeclidinium in Japanese patients with chronic obstructive pulmonary disease. Curr Med Res Opin 2016; 32: 967–973. doi: 10.1185/03007995.2016.1140029 [DOI] [PubMed] [Google Scholar]
- 9.Bjermer LH, Boucot IH, Vogelmeier CF, et al. Efficacy and safety of umeclidinium/vilanterol in current and former smokers with COPD: a prespecified analysis of the EMAX trial. Adv Ther 2021; 38: 4815–4835. doi: 10.1007/s12325-021-01855-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Pelaia C, Ferrante Bannera A, Rotundo FL, et al. Clinical and functional effects of inhaled dual therapy umeclidinium/vilanterol in patients with chronic obstructive pulmonary disease: a real-life study. Int J Chron Obstruct Pulmon Dis 2023; 18: 995–1002. doi: 10.2147/COPD.S407238 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Zhang C, Zhang M, Wang Y, et al. Efficacy and cardiovascular safety of LAMA in patients with COPD: a systematic review and meta-analysis. J Investig Med 2021; 69: 1391–1398. doi: 10.1136/jim-2021-001931 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Regard L, Burgel PR, Roche N. Inhaled therapy, cardiovascular risk and benefit–risk considerations in COPD: innocent until proven guilty, or vice versa? Eur Respir J 2023; 61: 2202135. doi: 10.1183/13993003.02135-2022 [DOI] [PubMed] [Google Scholar]
- 13.Yang M, Li Y, Jiang Y, et al. Combination therapy with long-acting bronchodilators and the risk of major adverse cardiovascular events in patients with COPD: a systematic review and meta-analysis. Eur Respir J 2023; 61: 2200302. doi: 10.1183/13993003.00302-2022 [DOI] [PubMed] [Google Scholar]
- 14.Battaglia S, Basile M, Scichilone N, et al. Prevalence of co-morbidities and severity of COPD. COPD 2015; 12: 390–394. doi: 10.3109/15412555.2014.974734 [DOI] [PubMed] [Google Scholar]
- 15.Müllerova H, Agusti A, Erqou S, et al. Cardiovascular comorbidity in COPD: systematic literature review. Chest 2013; 144: 1163–1178. doi: 10.1378/chest.12-2847 [DOI] [PubMed] [Google Scholar]
- 16.Anzueto A, Miravitlles M. Tiotropium in chronic obstructive pulmonary disease – a review of clinical development. Respir Res 2020; 21: 199. doi: 10.1186/s12931-020-01407-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Kerstjens HA, Engel M, Dahl R, et al. Tiotropium in asthma poorly controlled with standard combination therapy. N Engl J Med 2012; 367: 1198–1207. doi: 10.1056/NEJMoa1208606 [DOI] [PubMed] [Google Scholar]
- 18.Global Initiative for Chronic Obstructive Lung Disease . Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease. Date last accessed: 7 February 2025. Date last updated: 14 November 2018. https://goldcopd.org/wp-content/uploads/2018/11/GOLD-2019-v1.7-FINAL-14Nov2018-WMS.pdf
- 19.Global Initiative for Chronic Obstructive Lung Disease . Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease. Date last accessed: 7 February 2025. Date last updated 17 February 2023. https://goldcopd.org/wp-content/uploads/2023/03/GOLD-2023-ver-1.3-17Feb2023_WMV.pdf
- 20.Schoenfeld DA. Sample-size formula for the proportional-hazards regression model. Biometrics 1983; 39: 499–503. doi: 10.2307/2531021 [DOI] [PubMed] [Google Scholar]
- 21.Cole SR, Hernán MA. Constructing inverse probability weights for marginal structural models. Am J Epidemiol 2008; 168: 656–664. doi: 10.1093/aje/kwn164 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Watz H. On trapped air and trapped blood in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2019; 199: 1047–1048. doi: 10.1164/rccm.201901-0061ED [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Hohlfeld JM, Vogel-Claussen J, Biller H, et al. Effect of lung deflation with indacaterol plus glycopyrronium on ventricular filling in patients with hyperinflation and COPD (CLAIM): a double-blind, randomised, crossover, placebo-controlled, single-centre trial. Lancet Respir Med 2018; 6: 368–378. doi: 10.1016/S2213-2600(18)30054-7 [DOI] [PubMed] [Google Scholar]
- 24.Singh D, Han MK, Hawkins NM, et al. Implications of cardiopulmonary risk for the management of COPD: a narrative review. Adv Ther 2024; 41: 2151–2167. doi: 10.1007/s12325-024-02855-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Parkin L, Williams S, Barson D, et al. Is the use of two versus one long-acting bronchodilator by patients with COPD associated with a higher risk of acute coronary syndrome in real-world clinical practice? BMJ Open Respir Res 2021; 8: e000840. doi: 10.1136/bmjresp-2020-000840 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Andreas S, McGarvey L, Bothner U, et al. Absence of adverse effects of tiotropium/olodaterol compared with the monocomponents on long-term heart rate and blood pressure in patients with moderate-to-very-severe COPD. Int J Chron Obstruct Pulmon Dis 2020; 15: 1935–1944. doi: 10.2147/COPD.S246348 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Andreas S, Bothner U, de la Hoz A, et al. No influence on cardiac arrhythmia or heart rate from long-term treatment with tiotropium/olodaterol versus monocomponents by Holter ECG analysis in patients with moderate-to-very-severe COPD. Int J Chron Obstruct Pulmon Dis 2020; 15: 1945–1953. doi: 10.2147/COPD.S246350 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Matera MG, Calzetta L, Rogliani P, et al. Cardiovascular events with the use of long-acting muscarinic receptor antagonists: an analysis of the FAERS database 2020–2023. Lung 2024; 202: 119–125. doi: 10.1007/s00408-024-00677-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Calzetta L, Rogliani P, Matera MG, et al. a systematic review with meta-analysis of dual bronchodilation with LAMA/LABA for the treatment of stable COPD. Chest 2016; 149: 1181–1196. doi: 10.1016/j.chest.2016.02.646 [DOI] [PubMed] [Google Scholar]
- 30.Rogliani P, Matera MG, Ora J, et al. The impact of dual bronchodilation on cardiovascular serious adverse events and mortality in COPD: a quantitative synthesis. Int J Chron Obstruct Pulmon Dis 2017; 12: 3469–3485. doi: 10.2147/COPD.S146338 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Kunisaki KM, Dransfield MT, Anderson JA, et al. Exacerbations of chronic obstructive pulmonary disease and cardiac events. A post hoc cohort analysis from the SUMMIT randomized clinical trial. Am J Respir Crit Care Med 2018; 198: 51–57. doi: 10.1164/rccm.201711-2239OC [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Maleki-Yazdi MR, Singh D, Anzueto A, et al. Assessing short-term deterioration in maintenance-naïve patients with COPD receiving umeclidinium/vilanterol and tiotropium: a pooled analysis of three randomized trials. Adv Ther 2016; 33: 2188–2199. doi: 10.1007/s12325-016-0430-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Kerwin EM, Kalberg CJ, Galkin DV, et al. Umeclidinium/vilanterol as step-up therapy from tiotropium in patients with moderate COPD: a randomized, parallel-group, 12-week study. Int J Chron Obstruct Pulmon Dis 2017; 12: 745–755. doi: 10.2147/COPD.S119032 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Maleki-Yazdi MR, Kaelin T, Richard N, et al. Efficacy and safety of umeclidinium/vilanterol 62.5/25 mcg and tiotropium 18 mcg in chronic obstructive pulmonary disease: results of a 24-week, randomized, controlled trial. Respir Med 2014; 108: 1752–1760. doi: 10.1016/j.rmed.2014.10.002 [DOI] [PubMed] [Google Scholar]
- 35.Baldomero AK, Kunisaki KM, Bangerter A, et al. Beyond access: factors associated with spirometry underutilization among patients with a diagnosis of COPD in urban tertiary care centers. Chronic Obstr Pulm Dis 2022; 9: 538–548. doi: 10.15326/jcopdf.2022.0303 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Ezzati M, Obermeyer Z, Tzoulaki I, et al. Contributions of risk factors and medical care to cardiovascular mortality trends. Nat Rev Cardiol 2015; 12: 508–530. doi: 10.1038/nrcardio.2015.82 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Movsisyan NK, Vinciguerra M, Medina-Inojosa JR, et al. Cardiovascular diseases in central and eastern Europe: a call for more surveillance and evidence-based health promotion. Ann Glob Health 2020; 86: 21. doi: 10.5334/aogh.2713 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Maltais F, Bjermer L, Kerwin EM, et al. Efficacy of umeclidinium/vilanterol versus umeclidinium and salmeterol monotherapies in symptomatic patients with COPD not receiving inhaled corticosteroids: the EMAX randomised trial. Respir Res 2019; 20: 238. doi: 10.1186/s12931-019-1193-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Kardos P, Wencker M, Glaab T, et al. Impact of salmeterol/fluticasone propionate versus salmeterol on exacerbations in severe chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2007; 175: 144–149. doi: 10.1164/rccm.200602-244OC [DOI] [PubMed] [Google Scholar]
- 40.Dransfield MT, Bourbeau J, Jones PW, et al. Once-daily inhaled fluticasone furoate and vilanterol versus vilanterol only for prevention of exacerbations of COPD: two replicate double-blind, parallel-group, randomised controlled trials. Lancet Respir Med 2013; 1: 210–223. doi: 10.1016/S2213-2600(13)70040-7 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.
Supplementary material
01691-2025.SUPPLEMENT
Data Availability Statement
Please refer to GSK weblink to access GSK's data sharing policies and as applicable seek anonymised subject level data via the link https://www.gsk-studyregister.com/en/.