Efficacy and Safety of an Aclidinium Bromide Treatment for 12 Weeks or Longer in Patients with Moderate-To-Severe COPD: A Meta-Analysis

Abstract

Background: Bronchodilators are commonly used to manage patients with stable chronic obstructive pulmonary disease (COPD). This meta-analysis assessed the efficacy of aclidinium bromide with respect to clinical events, health-related quality of life and symptom scales, pulmonary function, and safety among patients with stable COPD. Methods: A comprehensive search of MEDLINE, EMBASE, CINAHL and the Cochrane Library databases was undertaken to identify randomized controlled trials (RCTs) that compared aclidinium with placebo, treatment over at least 12 weeks. Trials were measured using either odds ratios (OR) or mean differences (MD) with 95% confidence intervals (CI). Results: Seven studies, containing 7001 patients, were included in this meta-analysis. Compared with placebo, aclidinium bromide reduced the incidence of exacerbation-related hospitalizations (OR 0.64; 95% CI 0.47 to 0.89) and improved quality of life as measured by a lower total George's Respiratory Questionnaire [SGRQ] score (MD −2.34; 95% CI −3.18 to −1.51) and attenuated dyspnea symptom as assessed by changes in the Transitional Dyspnea Index [TDI] (MD 0.76; 95% CI 0.43 to 1.10). Similar changes were observed with regard to trough FEV₁ and FVC and peak FEV₁ and FVC. No significant differences were observed with regard to all-cause mortality, COPD exacerbations, non-fatal serious adverse events or cardiac adverse events. Conclusions: Aclidinium reduced the incidence of exacerbation-related hospitalizations and improved quality of life, COPD symptoms and pulmonary function. In addition, aclidinium did not increase the incidence of non-fatal serious adverse events, cardiac adverse events, or COPD exacerbations and was not associated with increased mortality.

Keywords:

• aclidinium bromide
• COPD
• efficacy
• meta-analysis
• safety

Introduction

Chronic obstructive pulmonary disease (COPD) is characterized by persistent and irreversible airflow limitations usually caused by a mixture of chronic bronchitis and emphysema (1). World Health Organization (WHO) estimates that COPD will become the third-leading cause of death by 2030 worldwide because of the increasing elderly population. This demographic will also be associated with global increases in morbidity, mortality and health-care costs worldwide (2,3). The main goals for treating stable COPD are controlling symptoms, improving lung function, preventing and controlling exacerbations, improving quality of life and decreasing mortality (4).

At present, the Global Initiative for Chronic Obstructive Lung Disease (GOLD) guidelines recommend using inhaled long-acting bronchodilators as the first-line therapy for patients with stable, moderate-to-severe COPD (4). Based on the results of recent studies, long-acting muscarinic antagonists (LAMAs) are considered as a more favorable type of bronchodilator in the management of COPD (5–7), because an important underlying mechanism of COPD is an increase in cholinergic tone, which might be the sole reversible element of the disease (5). However, traditional muscarinic receptor antagonists, including ipratropium bromide and tiotropium bromide, are associated with several adverse side effects, particularly adverse cardiac events (8,9). Therefore, novel LAMAs have been developed and confirmed as effective with regard to controlling symptoms in patients with COPD, without increasing the incidence of adverse cardiac events (10,11).

Aclidinium bromide is a novel LAMA, approved by the U.S. Food and Drug Administration for use among patients with stable, moderate-to-severe COPD (12). Its primary mechanism of action involves the blockade of muscarinic acetylcholine receptors, particularly M3 receptors with an affinity approximately 6-fold that involving the M2 receptor subtype in terms of kinetic selectivity (13,14). This action is the primary reason for the aforementioned decreased incidence of cardiac side effects. Compared with tiotropium, aclidinium has a faster onset but a shorter duration (t_1/2 = 29 hours). Therefore, aclidinium bromide may be prescribed either once or twice daily. Because butyrylcholinesterase quickly hydrolyzed aclidinium into either alcohol metabolites or carboxylic acid, it has a short plasma half-life and fewer drug-drug interactions, the primary reasons for the decreased incidence of the systemic side effects associated with aclidinium compared with LAMAs, such as tiotropium (13,15,16).

Several randomized controlled trials (RCTs) have described both the efficacy and safety of aclidinium in the treatment of patients with stable COPD; however, controversies exist regarding these patients' clinical outcomes. To perform a high-quality meta-analysis, we did not consider either crossover studies or studies with brief follow-up period. The current meta-analysis might provide clinicians with a clearer picture of the true effects of aclidinium bromide treatment compared with placebo.

Methods

Data sources

We comprehensively searched the MEDLINE, EMBASE, CINAHL and Cochrane Libraryn databases to identify published studies using the following medical subject headings and keyword terms: “COPD” AND “Aclidinium” OR “Aclidinium bromide” OR “LAS34273” OR “Eklira” OR “Genuair.” We also performed searches using drug manufacturer's database and Clinical Trials.gov; furthermore, we manually searched the respiratory journals without restrictions pertaining to language. All trials in this meta-analysis were published prior to March 1, 2015. Moreover, we did not included clinical studies published only in abstract form because of the unavailability of the data related to outcomes and methodological quality.

Study selection

The clinical trials were required to meet the following criteria for inclusion in our analysis: (1) Target population: patients with stable, moderate-to-severe COPD consistent with the diagnostic criteria of the GOLD (4), American Thoracic Society (ATS) and European Respiratory Society (ERS) (17); (2) Intervention: clinical trials comparing aclidinium bromide with placebo; (3) Study design: RCTs employing a parallel-group, placebo-controlled, double-blinded design that followed patients for at least 12 weeks after randomization.

Two primary reviewers (Yong Zou and Jian Xiao) independently identified trials after reading the titles, abstracts and full texts when necessary using the above inclusion criteria. All disagreements were resolved via consensus.

Data extraction and methodological quality assessment

Two reviewers (Yong Zou and Jun Li) independently extracted data using a standardized data extraction form and agreed on all of the items. The primary outcomes included COPD exacerbations, hospitalizations for COPD exacerbations, all-cause mortality and a quality of life scale (St George's Respiratory Questionnaire, SGRQ) (18). The secondary outcomes included symptom scores (Transitional Dyspnea Index, TDI) (19), trough FEV₁ and FVC values, peak FEV₁ and FVC values, non-fatal serious adverse events and cardiac adverse events. Based on the Cochrane Collaboration guidelines, we assessed the methodological quality of all of the included trials using a risk of bias graph and summary.^{Figure 1. Table 1. Figure 2.}

Statistical analyses

RevMan 5.3 was used to perform the meta-analysis. Odds ratios (OR) or mean differences (MD) with 95% confidence intervals (CI) were calculated for each trial to assess the dichotomous variables and continuous outcomes, respectively. The Cochran Q test and I² test were used to determine data heterogeneity, which was considered significant when the Cochran Q test p-value ≤ 0.10 and the I² value ≥ 50%. A random effects model was used when heterogeneity was observed; otherwise a fixed-effects model was employed for the data synthesis.

Trials were combined into categories related to the frequency of the aclidinium bromide administration (once or twice daily) for each outcome. Potential publication biases would have been examined using a funnel plot if the outcomes had included enough studies. A sensitivity analysis would have been used if any of the subgroups had included a low-quality trial.

Figure 1.   PRISMA flow diagram.

Table 1.   Main characteristic of seven trials included in this meta-analysis.

Study	Design type	Duration (weeks)	Patients, n(% male)	Age, ME(SD), years	FEV1, % predicted (SD)	FEV1/FVC ratio, ME(SD)%	Smoking history, ME(SD) pack-years	Treatments	Inclusion/exclusion criteria	Concomitant medications
ACCLAIM/COPD I^20	DB/PC/PG	52 weeks	AC:488(77.8)PL:175(81.0)	62.6(8.2) 61.9(8.3)	54.2(15.1) 52.9(15.2)	49.2(11.4) 48.9(11.1)	40.4(21.0) 38.4(18.3)	AC: 200 μg OD PL: matched QD	Inclusion: COPD, FEV1< 80% of predicted and ratio ≤0.7, age ≥40 years, smoking history ≥10 pack- years.Exclusion: asthma, blood eosinophil count > 600 cell/mm^3, respiratory Tract infection or COPD exacerbation.use of long-term oxygen therapy.	SABA, ICS, theophyllines,oral or parenteral corticosteroids ≤10 mg/day.
ACCLAIM/COPD II^20	DB/PC/PG	52 weeks	AC:383(63.8)PL:124(60.8)	65.1(8.6) 65.2(8.6)	50.6(15.6) 49.4(15.1)	48.0(11.7) 47.1(11.9)	57.8(29.9) 58.2(28.4)	AC: 200 μg OD PL: matched QD	Inclusion: COPD, FEV1< 80% of predicted and ratio ≤0.7,age ≥ 40 years, smoking history ≥ 10 pack-years.Exclusion: asthma, blood eosinophil count > 600 cell/mm^3, respiratory tract infection or COPD exacerbation.use of long-term oxygen therapy.	SABA, ICS, theophyllines, oral or parenteral corticosteroids ≤ 10 mg/day.
ACCORD/ COPD I ^21	DB/PC/PG	12 weeks	AC:201(53.7)PL:96(51.6)	64.0(9.5) 65.1(9.2)	53.3(13.3) 54.6(13.5)	51.2(10.4) 52.7(10.5)	55.1(25.9) 52.7(28.1)	AC: 200 μg TW 400 μg TW PL: matched TW	Inclusion: COPD, FEV1< 80% and ≥ 30% of predicted and ratio <0.7, age ≥ 40years, smoking history ≥ 10 pack-years. Exclusion: asthma, respiratory tract infection or COPD exacerbation; myocardial infarction; unstable arrhythmia.	ICS, theophyllines, oral or parenteral corticosteroids ≤ 10 mg/day. Anticholinergics And LABAs were prohibited throughout the study
ACCORD/ COPD II^22	DB/PC/PG	12 weeks	AC:188(52.2)PL:100(54.9)	63.3(8.7) 61.7(9.3)	51.1(13.4) 55.2(13.2)	50.1(10.5) 53.3(11.2)	54.5(29.5) 52.6(28.4)	AC: 200 μg TW 400 μg TW PL: matched TW	Inclusion: COPD, FEV1< 80% and ≥ 30% of predicted and ratio < 0.7, age ≥ 40 years, smoking history ≥10 pack-years.Exclusion: asthma, respiratory tract infection or COPD exacerbation requiring hospitalisation; myocardial infarction; unstable arrhythmia; hypersensitivity	ICS, theophyllines, oral or parenteral corticosteroids ≤ 10 mg/day. Anticholinergics And LABAs were prohibited throughout the study
ACLIFOR^23	DB/PC/PG	24 weeks	AC:256(66.5)PL:138(71.1)	63.1(8.2) 64.2(8.0)	53.6(13.0) 55.0(13.4)	NA NA	NA NA	AC: 400 μg TW PL: matched TW	Inclusion: COPD,age ≥ 40 years, smoking history ≥10 pack-years. Exclusion: asthma, respiratory tract infection or COPD exacerbation requiring hospitalisation; significant respiratory conditions; significant cardiovascular conditions	NA
ATTAIN^24	DB/PC/PG	24 weeks	AC:263(48.2)PL:189(69.2)	62.6(8.1) 62.0(8.0)	56.9(12.7) 56.6(12.8)	NA NA	40.8(20.4) 38.9(18.3)	AC: 200 μg TW 400 μg TW PL: matched TW	Inclusion: COPD, FEV1< 80% of predicted and ratio ≤ 0.7, age ≥ 40 years, smoking history ≥10 pack-years. Exclusion: asthma respiratory tract infection or COPD exacerbation;unstable cardiac conditions	ICS, theophyllines, oral or parenteral corticosteroids ≤ 10 mg/day.
AUGMENT COPD^25	DB/PC/PG	24 weeks	AC:188(55.8)PL:175(52.7)	64.4(8.7) 63.5(8.9)	53.0(13.3) 52.6(13.3)	NA NA	52.0(26.1) 53.3(28.5)	AC: 400 μg TW PL: matched TW	Inclusion: COPD, age ≥ 40 years, smoking history ≥ 10 pack-years. Exclusion: asthma, respiratory tract infection or COPD exacerbation requiring hospitalisations; significant cardiovascular conditions; use of oxygen therapy	NA

Note: DB = Double-blind; PC = Placebo-controlled; PG = Parallel-group; AC = Aclidinium; PL = Placebo; ME = Mean; SD = standard deviation; NA = Not available; OD = Once daily; TD = Twice daily; ICS = Inhaled corticosteroid; SABA = Short acting β2-agonist; LABA = Long-acting beta2-agonist).

Figure 2.   (A) Risk of bias graph: A review of each risk of bias item as a percentage among the trials; (B) A summary of the risk of bias assessment.

Results

Of the 251 potentially relevant studies identified (after duplicates removed), 194 records were excluded based on the title or abstract and 50 records were excluded based on the full text. For detailed search results, please see Figure 1. Thus, 7 RCTs (20–25), including 7001 participants, were kept in the meta-analysis. The largest study (23) included 1729 participants, whereas the smallest trial (22) contained 544 patients. The main characteristics of the seven included trials are listed in Table 1.

Each of the seven trials compared aclidinium bromide with placebo; in two trials (20), aclidinium was prescribed once daily, whereas the remaining five trials prescribed it twice daily (21–25). Each RCT employed a double-blinded, placebo-controlled, parallel design, and lasted for at least 12 weeks. Each study was of acceptable methodological quality; detailed assessments of each trial are presented as percentage bar plots for each quality item and as a risk of bias summary (Figure 2). The baseline characteristics, including gender, age, smoking history, disease severity were generally balanced between the two groups. The inclusion criteria and concomitant medications were also similar between groups.

Clinical events

COPD exacerbations

Seven articles reported the number of patients who suffered from at least one COPD exacerbation and were treated with antibiotics, short course of oral steroids, or both (20–25). The pooled results demonstrated that the cumulative rates of COPD exacerbation were 15.4% and 15.3% in the aclidinium and placebo groups, respectively. Fewer COPD exacerbations were observed in the aclidinium group than the placebo group, although the difference was small and not significant (OR 0.90; 95% CI 0.75 to 1.07; p = 0.22). A subgroup analysis did not reveal significant differences (p = 0.57) between once daily (OR 0.95; 95% CI 0.73 to 1.22) and twice daily (OR 0.85; 95% CI 0.67 to 1.09) administrations of aclidinium Figure 3A.

Figure 3.   A summary of the effects of aclidinium on (A) COPD exacerbations (B) exacerbation-related hospitalizations and (C) all-cause mortality.

Hospital admission due to exacerbation

Seven trials reported data regarding hospitalizations due to exacerbations; aclidinium was administered once daily in 2 trials (20) and twice daily in 5 trials (21–24). The cumulative rates of exacerbation-related hospitalizations between the aclidinium and placebo group were 3.2% and 4.1%, respectively. Compared with placebo, aclidinium significantly reduced the rate of hospital admission due to exacerbation (OR 0.64; 95% CI 0.47 to 0.89; p = 0.008). A subgroup analysis did not reveal significant differences (p = 0.7) between once daily (OR 0.68; 95% CI 0.45 to 1.02) and twice daily (OR 0.59; 95% CI 0.35 to 1.01; Figure 3B).

All-cause mortality

Seven trials reported the number of deaths in the two groups (20–25). In general, no significant between-group differences were observed with regard to death (OR 0.92; 95% CI 0.43 to 1.94; p = 0.82). No data heterogeneity was noted among the studies (I² = 0%). A subgroup analysis between once daily (OR 0.63; 95% CI 0.25 to 1.60) and twice daily (OR 1.69; 95% CI 0.46 to 6.21) aclidinium administrations did not indicate any significant difference (p = 0.23; Figure 3C).

Health-related quality of life and symptom scales

SGRQ

SGRQ was used to evaluate health-related quality of life in seven trials (20–25). Compared with placebo, treatment with aclidinium significantly reduced the total SGRQ score (Figure 4A). More patients achieved the threshold of 4 units in the aclidinium group than did those in the placebo group (Figure 4B). However, significant heterogeneity was noted among the studies (I² = 58%). A subgroup analysis revealed that both once and twice daily aclidinium administration groups exhibited significantly reduced total SGRQ scores and increased numbers of patients who achieved ≥ 4 units of improvement in their total SGRQ score despite the absence of a significant difference between group differences (p = 0.56 and p = 0.25, respectively).

Figure 4.   A summary of the effects of aclidinium on the (A) changes in total SGRQ scores and (B) number of patients who achieved ≥ 4 units of improvement in their total SGRQ scores.

Figure 5.   A summary of the effects of aclidinium on the (A) change in the TDI focal score and the (B) patients who achieved ≥ 1 unit of improvement in their TDI focal score.

Figure 6.   A summary of the effects of aclidinium on the (A) change in trough FEV₁ (ml) and (B) change in the peak FEV₁ (ml).

Figure 7.   A summary of the effects of aclidinium on the (A) change in trough FVC (ml) and the (B) change in peak FVC (ml).

TDI

Both change in the TDI and the number of patients who meet the threshold of at least 1 unit of improvement on the TDI were suitable evaluations of dyspnea. The pooled results demonstrated that the change in the TDI was greater among patients treated with aclidinium compared with those receiving placebo (Figure 5A). A random effects model was employed because of the significant heterogeneity (I² = 66%). To determine the source of the heterogeneity, we performed sensitivity analyses by excluding one study (23) in which the baseline FEV₁ was lower among patients treated with aclidinium than those receiving placebo and observed no heterogeneity (MD 0.95; 95% CI 0.72 to 1.19; I² = 0%). A similar result was found regarding the number of patients who experienced a clinically significant change in their TDI (Figure 5B). A subgroup analysis between the once and twice daily administration groups did not reveal significant differences in the above indices (p = 0.67 and p = 0.90, respectively).^{Figure 6.}

Spirometric indices

Change in trough FEV₁ and FVC (ml)

The pooled results from the seven included trials (20–25) showed that the mean improvement in the trough FEV₁ related to aclidinium was greater than that associated with placebo from baseline to the end of the study (Figure 6A). Some data heterogeneity was observed among the studies (I² = 39%). A subgroup analysis determined that twice daily aclidinium treatments (MD 101.52; 95% CI 83.37 to 119.68) elicited greater improvement than once daily treatments (MD 63.48; 95% CI 38.33 to 88.62) (p = 0.02). A similar result was observed with regard to the trough FVC between the aclidinium and placebo groups (Figure 6B). A small amount of data heterogeneity was noted among the trials (I² = 23%). A subgroup analysis did not reveal a significant difference between the once daily and twice daily groups (p = 0.14).

Change in peak FEV₁ and FVC (ml)

Pooling the data of all of the included trials indicated a greater improvement in peak FEV₁ (Figure 7A) and FVC (Figure 7B) in the aclidinium group versus the placebo group from baseline to the end of the trial. No data heterogeneity was noted in the above two indices. A subgroup analysis did not indicate any significant difference between the once and twice daily groups with respect to peak FEV₁ and FVC (p = 0.38 and p = 0.37, respectively).

Safety

Non-fatal serious adverse events

Seven trials included detailed information regarding the rates of non-fatal serious adverse events and cardiac adverse events during the trials. The overall cumulative incidence of non-fatal adverse events was approximately the same for both groups (6.3%). We did not observe significant between-group differences in the rates of non-fatal serious adverse events (OR 0.88; 95% CI 0.69 to 1.14; p = 0.34; Figure 8A). No statistical heterogeneity was observed across the studies. An additional subgroup analysis of the two aclidinium treatment frequencies did not uncover significant differences (p = 0.68).

Figure 8.   A summary of the effects of aclidinium on (A) non-fatal serious adverse events and (B) adverse cardiac events.

Cardiac adverse events

All seven of the included trials reported the rates of cardiac adverse events. Aclidinium was taken once daily in two trials (20) and twice daily in five trials (21–24). The cumulative incidence of cardiac events was lower in the aclidinium group than the placebo group although this difference was not significant (OR 0.94; 95% CI 0.68 to 1.30; p = 0.71; Figure 8B). No statistical heterogeneity was noted among the studies. An additional subgroup analysis did not reveal significant differences between the once daily and twice daily groups (p = 0.27).

Publication bias

The funnel plot of COPD exacerbations did not demonstrate significant publication biases based on a comparison between the aclidinium and placebo groups (Figure 9).

Figure 9.   A funnel plot of COPD exacerbations based on a comparison the aclidinium and placebo groups.

Discussion

This systematic review of seven double-blinded, placebo-controlled, parallel, long-term RCTs demonstrated that aclidinium, compared with placebo, reduced the incidence of exacerbation-related hospitalizations and improved quality of life among patients with moderate to severe, stable COPD. Decreases were noted regarding the number of COPD exacerbations and the number of total deaths among participants treated with aclidinium compared with those receiving a placebo, although no significant differences were observed. In term of secondary outcomes, clinically significant improvements in both symptom scales and spirometric indices were noted among the patients treated with aclidinium bromide, compared with those receiving a placebo. Compared with the placebo, both the change in the TDI focal score and the number of patients who achieved the threshold of 1 unit were larger among patients treated with aclidinium. Improvements in both trough and peak FEV₁ as well as trough and peak FVC were significantly greater in the aclidinium group than the placebo group. In terms of safety, decreases were observed in the rates of non-fatal serious adverse events and adverse cardiac events among participants treated with aclidinium compared with those receiving a placebo, although the decreases were not significant.

The results of this meta-analysis were somewhat inconsistent with those of previously published reviews. The combined results of two RCTs (21,24) inlcuded in a previous review (26) shown significant improvements in both spirometric indices (trough FEV₁ and peak FEV₁) and quality of life among patients treated with aclidinium compared with those receiving a placebo, which is consistent with our findings. However, aclidinium significantly reduced the rates of COPD exacerbations according to the pooled analysis, which differs from the findings of our review. Results similar to those found in our review was reported with regard to both trough and peak FEV₁ in a meeting abstract (27) including the results of three RCTs (21, 22, 24) in the comparison aclidinium versus placebo.

Ten trials included in another review (28) involving both parallel-group and cross-over studies reported the data regarding FEV₁, exercise capacity and hyperinflation, rescue medication use, health status and symptom relief. However, this review did not perform a meta-analysis, and its conclusions were only based on the strengths of the individual studies with respect to particular outcomes. Therefore, the results reported by that review were not comparable with the current results. Two reviews (29,30) analyzed the rates cardiovascular death, non-fatal myocardial infarction and stroke; no significant between-group differences were observed, which supported our results.

Seven trials were of excellent methodological quality and met all of the inclusion criteria. The clinical characteristics and disease severity values of the samples recruited in these studies were similar across all but one RCT (22) in which the participants treated with aclidinium exhibited lower FEV₁ values than did those receiving a placebo. Therefore, we repeated our analysis of all outcomes by excluding this trial to detect any changes. However, the corresponding results indicated that alterations were not present in the heterogeneity observed among the outcome parameters with the exception of the mean change in the TDI focal score.

Another potential restriction of this meta-analysis was the double counting of participants included in multiple publications and we successfully averted this bias by extracting only the data pertaining to the first period of the primary study (25) rather than the results at the end of the extension study (31) because the extension period was not a randomized comparison and patient withdrawals affected the results. We undertook systematic searches to determine all relevant studies, emailed the authors of the trials that included incomplete data and independently evaluated the included trials by two reviewers, avoiding selection bias adequately. A funnel plot of the COPD exacerbation data did not exhibit any publication bias.

The current meta-analysis did not consider RCTs with a crossover design and restricted the duration to at least 12 weeks in order to increase the quality of the included trials. Moreover, this meta-analysis was the first to evaluate the benefits and risks of aclidinium treatment among patients with moderate-to-severe COPD using sub-group analysis that compared the results of once and twice daily administrations. The results indicated that aclidinium dosing did not significantly affect the outcome measures with the exception of the mean change in the trough FEV₁ value; specifically two daily dose elicited greater improvements than one daily dose.

Our meta-analysis has several limitations and shortcomings. First, only seven trials were selected for this meta-analysis and a ongoing study (32) was not included because its data was not available. Second, we did not compare aclidinium with either LABAs or LAMAs because they lack the trials needed to fulfill the review's inclusion criteria. Finally, we did not perform a sub-group analysis of aclidinium dosages between 200 μg and 400 μg twice daily because of the lack of eligible studies.

Taken together, this meta-analysis demonstrates that inhaled aclidinium bromide reduces the incidence of exacerbation-related hospitalizations and improves quality of life over placebo. Aclidinium also significantly improves symptoms and pulmonary function but did not significantly decrease the rates of COPD ­exacerbations or mortality. A decreased rather than increased trend toward cardiac adverse events, non-fatal serious adverse events was noted among participants treated with aclidinium compared to placebo in spite of no statistical difference. Consequently, aclidinium was of good efficacy and safety profile and recommended as a reasonable first-line choice of managing the patients with moderate-to-severe stable COPD. Because of the limited number of eligible trials included in this paper, clinicians should be cautious when applying these results to clinical decision-making; additional long-term trials should be performed to evaluate the risks and benefits of aclidinium bromide compared with other LAMAs or LABAs.

Declaration of interest statement

The authors have no conflicts of interest to disclose. The authors alone are responsible for the content and writing of the paper.

Acknowledgments

We express our gratitude to the authors of the included studies and we thank Almirall for providing additional data and information.