Exercise capacity and physical activity in COPD patients treated with a LAMA/LABA combination: a systematic review and meta-analysis

Background Persistent airflow limitation and dyspnoea may reduce chronic obstructive pulmonary disease (COPD) patients exercise capacity and physical activity, undermining their physical status and quality of life. Long-acting muscarinic antagonists and long-acting beta-2 agonists (LAMA/LABA) combinations are amongst moderate-to-severe COPD recommended treatments. This article analyses LAMA/LABA combinations effect on COPD patients exercise capacity and physical activity outcomes. Methods A systematic review and meta-analysis of double-blind randomized controlled trials comparing LAMA/LABA combinations against monotherapy or placebo was conducted. Results Seventeen articles were identified (N = 4041 patients). In endurance shuttle walk test and constant work rate cycle ergometry, LAMA/LABA combinations obtained better results than placebo, but not monotherapy, whereas in 6-min walking test, results favoured LAMA/LABA over monotherapy (four studies), but not over placebo (one study). Moreover, LAMA/LABA combinations obtained better results than placebo in number of steps per day, reduction in percentage of inactive patients and daily activity-related energy expenditure, and better than monotherapy when measuring time spent on ≥ 1.0–1.5, ≥ 2.0 and ≥ 3.0 metabolic equivalents of task activities. Conclusions LAMA/LABA combinations in COPD patients provided better results than monotherapy or placebo in most exercise capacity and physical activity outcomes. Supplementary Information The online version contains supplementary material available at 10.1186/s12931-022-02268-3.

Chronic Obstructive Lung Disease (GOLD) recommends regular physical activity for COPD patients [1]. The documented reduction in daily activity in COPD patients results from the respiratory and non-respiratory clinical conditions of each patient. Particularly, the limitation on exercise capacity is mainly due to dynamic pulmonary hyperinflation, although other factors also contribute, such as comorbidities or an imbalance between respiratory and locomotive muscles due to limited energy supply [5]. Moreover, it has been shown that the exercise capacity and the limitation in daily activities are closely related to life expectancy and, therefore, pulmonary and systemic manifestations would be improved by improving exercise capacity and physical activity [6].
The recommended treatment for patients with moderate-to-severe COPD and for symptomatic patients or those with exercise limitation, is inhaled long-acting beta-2 agonists (LABA) and/or long-acting muscarinic antagonists (LAMA) [1,6,7]. Bronchodilators increase lung emptying by reducing airway resistance, enabling COPD patients to achieve better alveolar ventilation with a lower operating pulmonary volume, both at rest and during exercise. As a result, patients using bronchodilators are able to exercise for longer before reaching the critical limit of their inspiratory reserve [8].
Due to the relevance of exercise capacity and physical activity on the quality of life of COPD patients, we have conducted a systematic literature review (SLR) and metaanalysis of randomized clinical trials aimed to evaluate the effect of the combination of LAMA/LABA bronchodilators compared with placebo or LAMA or LABA monotherapy on the exercise capacity and physical activity outcomes of COPD patients.

Methods
This SLR was carried out according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses Statement (PRISMA) and the QUORUM Statement [9]. The protocol was registered with PROSPERO (CRD42020191639).

Inclusion and exclusion criteria
We included randomized clinical trials in patients aged ≥ 40 years diagnosed with COPD, with a post-bronchodilator forced expiratory volume at 1 s (FEV 1 )/forced vital capacity (FVC) < 0.7 and treated with a combination of LAMA/LABA inhaled bronchodilators compared with placebo or monotherapy with LAMA or LABA. To be included, the trials had to evaluate at least one variable related to exercise capacity or physical activity.

Search strategy
A search strategy was designed for MEDLINE (through PubMed), CENTRAL and EMBASE using appropriate controlled terms related to COPD, LAMA, LABA, exercise capacity, physical activity and lung function in articles published between the 1st January 2012 and the 31st December 2021 as the first LAMA/LABA combination inhaler was approved in 2013 and prior 2012 there was no evidence about double bronchodilators (Additional file 2: Table A1). There were no limitations regarding language. Additionally, references to selected articles were also reviewed to identify other articles that met the inclusion and exclusion criteria.

Study selection and data extraction
The titles and abstracts resulting from the search were evaluated by two reviewers. The studies that didn't meet inclusion and exclusion criteria were ruled out, collecting the reasons for exclusion. The articles selected were read independently in full by the same two reviewers, who recorded the reasons for non-selection. In the event of discrepancies between the reviewers, the criterion of a third reviewer was used.
Data from the selected articles were tabulated by one reviewer and validated by a second reviewer in a detailed extraction form. From each article we extracted the study characteristics (type of study, design, countries), patient characteristics (mean age, sex, disease severity), and interventions and comparators (LAMA/LABA, LAMA, LABA or placebo inhalers used, dose, treatment duration), and the results of variables related to the exercise capacity and physical activity.

Assessed outcomes
The identified outcome variables are defined as:  [10].

Assessment of risk of bias
The risk of bias assessment was carried out according to the Cochrane Manual for Systematic Reviews and Meta-Analysis of Interventions criteria [11] and evaluated the generation of the randomization sequence, concealment of the assignment, blinding of patients and researchers, blinding of the results of the variables to be evaluated, data on incomplete results, bias of scientific information, and other biases. The risk of bias was assessed by one reviewer and validated by a second on a detailed form. Review Manager 5.4 was used for the risk of bias assessment.

Data analysis
The analysis was based on the change from baseline in the above-mentioned outcome variables and assessed using dichotomous and continuous outcomes. Dichotomous data were analysed by calculating the estimate for the odds ratio (OR) and their corresponding 95% confidence intervals (CI). Continuous data were analysed by calculating weighted mean differences (WMD) and standardized mean differences (SMD), both with the corresponding 95% CI. When useful, forest plots were created, in order to graphically assess the variability of sample estimates and the weight of sample sizes in the calculation of estimates (weighted averages). In addition, to facilitate interpretation of the results from studies that were not included in the forest plots, the mean and standard deviation were shown. A significance level of α = 0.05 was considered.
For data synthesis among studies, statistical heterogeneity was evaluated using I 2 , with I 2 > 50% considered to be significant heterogeneity. In those comparisons with no statistical evidence of heterogeneity, a fixed effects model was used; otherwise, a random effects model was employed.
A sensitivity analysis stratified by study design (parallel and cross-over) was performed for results that showed heterogeneity (I 2 > 0%).
The analysis considered the results of two treatment arms compared in each study. For studies with more than 2 treatment arms, comparisons were made separately, dividing the sample size of the study by the number of comparisons to avoid overestimation of results. The analysis was made using Review Manager 5.4.

Risk of bias
The risk of bias was considered low for all domains evaluated except for blinding of the results and concealment of assignment domains, where the risk of bias was unclear for the majority of studies analysed (Additional file 1: Fig.  A1). Table 2 shows a summary of meta-analysis comparisons for those variables of interest that estimated change from baseline.
Regarding 6MWT, the mean difference of 11.87 mts. observed between LABA/LAMA and placebo in a single study (n = 125) did not reach statistical significance (Table 2); however, the meta-analysis of results of the 4 studies comparing LAMA/LABA with monotherapies    (n = 634) showed significant differences in 6MWT in favour of LAMA/LABA combination ( Table 2; Fig. 2c).

Effectiveness of the intervention in physical activity
When measured in steps per day, LAMA/LABA combinations were significantly superior to both placebo and monotherapy (Fig. 3a). Regarding daily duration activity, patients treated with LAMA/LABA combination reduced the duration of ≥ 1.0-1.5 METs activity than patients treated with monotherapy. On the other hand, for moderate physical activity, the results favoured LAMA/LABA therapy by increasing the duration of ≥ 2.0 METs activities. For vigorous physical activity (≥ 3.0 METs), LAMA/ LABA therapy was superior to both monotherapy and placebo, although the latter results were not statistically significant when standardized under a random effects model (Fig. 3b). Daily activity-related energy expenditure was higher in the LAMA/LABA group than in the placebo group (Fig. 3c). Finally, more inactive patients (< 6000 steps/day) were observed in the placebo group than in the LAMA/LABA combination group (OR [95% CI]: 0.27 [0.14-0.51]; 1 study, N = 267) [15]. In Troosters et al. [12] walking intensity and walking time per day were also evaluated at week 12, results for average daily walking time mirrored those of steps per day and there was a small but significant increase in average daily walking  [12] similar results were observed for LAMA/LABA combination and LAMA monotherapy vs. baseline at week 9 for difficulty and amount domain, and at week 12 for difficulty domain. No statistically improvements were shown when comparing LAMA/LABA vs. placebo, although numerically the combination showed better results in both domains at week 9 and 12 [12].

Sensitivity analysis
Sensitivity analysis, stratified by study design in cases where heterogeneity was present, confirmed the LAMA/LABA combinations favorable results compared to monotherapy Table 2 Summary of meta-analysis comparisons and main results in weighted mean differences (WMD) and standardised mean differences (SMD) * Statistically significant differences, both in WMD and SMD. **Statistically significant differences in SMD. ***Statistically significant differences in WMD CI confidence interval, CT clinical trial, CWRCE constant work rate cycle ergometry, ESWT endurance shuttle walk test, kcal kilocalories, LABA long-acting beta-2 agonists, LAMA long-acting muscarinic antagonists, m meters, MD mean difference, MET metabolic equivalent of task, min minutes, N number of patients, s seconds, SMD standardized mean difference, T lim CWRT tolerance limit in constant work rate test, WMD weighted mean difference, 6MWT 6-min walking test

LAMA/LABA comparator Characteristics Weighted results
Standardized results  [29], as the combination of the two mechanisms of action effectively reduce the dynamic hyperinflation process characteristic in Fig. 3 Steps/day, duration of activity and energy expenditure; mean change from baseline, LAMA/LABA vs placebo and vs monotherapy. Random effects analysis model (against placebo, ≥ 3.0 METs); fixed effects analysis model for other comparisons. Aclid aclidinium, CI confidence interval, Formo formoterol, Glicopi glycopyrronium, Inda indacaterol, IV inverse variance, kcal kilocalories, LABA long-acting beta-2 agonists, LAMA long-acting muscarinic antagonists, MET metabolic equivalent task, Olo olodaterol, Plcb placebo, SE standard error, Std. standardized, Tio Tiotropium, µm microgram COPD patients, that usually limits their ability to exercise [6,7,30]. Several studies on COPD patients have associated low levels of physical activity and sedentary time with an increased frequency of exacerbations, hospitalizations, worse quality of life, and also an increased risk of death as a result of progressive ventilatory limitation, cardiac impairment, peripheral muscle, and psychological factors [3,31,32]. Increasing physical activity and its intensity in those patients may improve quality of life and reduce the loss of pulmonary function [33,34].

Number
Moreover, increasing the duration of low-intensity activity, instead of high-intensity activity, contributes to a lower risk of hospitalization in patients with moderate to severe COPD, which can be achieved with combined LAMA/LABA therapies [16,25]. However, reaching better exercise capacity is no guarantee of physical activity improvements [12,35]. Regarding this topic, in this metaanalysis, LAMA/LABA therapy significantly reduced the duration of 1.0-1.5 METs (sedentary time) and increased the durations of ≥ 2.0 METs (standing position or walking less than 55 m/min), and ≥ 3.0 MET (walking faster than 55 m/min). In general, our results provide proof of a significant reduction in sedentary time in patients with COPD who are administered LAMA/LABA compared to monotherapy.
The results observed in sedentary time were paralleled with significant improvements in daily walking time and in the intensity of walking in the D-PPAC questionnaire score where superiority of LAMA/LABA combinations over placebo was observed, as it was already noticed in the PHYSACTO and ACTIVATE studies. In the PHYS-ACTO study, significant differences in the questionnaire score between tiotropium monotherapy and the tiotropium/olodaterol combination were found [12], and in the ACTIVATE study between placebo and aclidinium/formoterol combination [15], indicating that LAMA/LABA combination improves the amount and level of intensity of physical activity in COPD patients.
In some observational studies [36][37][38][39][40][41] the use of tiotropium/olodaterol showed improvements in patient selfreported physical condition. Therapeutic success in the physical functioning score varied from 48.9% to 67.8%, with improved patient general condition as indicated by an improvement in Physician's Global Evaluation scores between visits in these studies [36][37][38][39][40][41] and increased absolute physical functioning scores [36]. These results are consistent with those obtained in our meta-analysis, where tiotropium/olodaterol was the most frequent LAMA/LABA analysed versus monotherapy, used in five different studies [13,14,16,25]. Also, tiotropium/olodaterol was compared to placebo in the study by Maltais et al. [20]. In general, LAMA/LABA combinations were superior to LAMA or LABA monotherapies. Differences were not significant when comparing LAMA/LABA versus monotherapy in ESWT or CWRCE tests, probably because there could be a threshold for bronchodilation to immediately translate into better exercise tolerance. It may be unrealistic to expect the same exercise benefit when adding a second bronchodilator to an existing one than when adding a bronchodilator to placebo [13]. These results agree with recent meta-analysis, which also concluded that LAMA/LABA combinations were more effective than LABA or LAMA monotherapy in terms of exercise capacity and symptoms [6,42]. The metaanalysis by Di Marco et al. [6] showed weighted mean increase in endurance time of 78.4 s with LAMA/LABA, 72.6 s with LAMA monotherapy and 51 s with LABA monotherapy compared to placebo, and improvements in BORG scale score of -0.25 units with LAMA/LABA versus − 0.51 and − 0.45 with LABA and LAMA monotherapies respectively. The relative effect results of the meta-analysis by Calzetta et al. [42] also pointed LABA/ LAMA as the combination significantly (P < 0.05) more effective than the LABA or LAMA alone and placebo in terms of improvement in endurance time (+ 43, + 22 and + 60 s, respectively) and increase in inspiratory capacity as measure of reduction in lung hyperinflation (+ 107 ml, + 87 ml and + 229 ml, respectively), although these improvements were slightly lower than the ones observed by Di Marco et al. [6] as Calzzeta et al. point out [42]. The results of both meta-analyses are in line with our results, as in our analysis differences between LAMA/LABA versus placebo or monotherapy were also significant (LAMA/LABA vs placebo + 31.75-72.45 s, vs. monotherapy + 11.36% and + 24.23 s).
Besides pharmacological treatment, the ATS, the European Respiratory Society (ERS) and the Spanish guidelines for COPD agree on using non-pharmacological treatment as part of the comprehensive COPD patient care as increasing physical activity and reducing discomfort during physical activity requires a more integrated approach than only providing adequate bronchodilation and it should consider all aspects of the disease, including mental, physical and emotional health [43][44][45][46][47]. Besides, as hyperinflation is the main driver of the reduced physical activity in COPD patients, by combining effective bronchodilators with pulmonary rehabilitation pulmonary function will be optimized and gas trapping reduced, increasing patient's exercise capacity [48][49][50]. Pulmonary rehabilitation includes exercise training, education and behavior change, aimed to improve the physical and psychological condition of COPD patients and to promote the long-term adherence to health-enhancing behaviours [47]. Before any actions are undertaken it is important to assess the initial level of physical activity in daily life as physical activity can be improved with the appropriate strategies in most COPD patients, and during all this process counselling or psychological programmes help supporting the change in behaviour that is needed for patients to be more active. Accordingly, the implementation of physical performance or muscle function/mass tests that correlate with objectively measured physical activity in clinical practice can be a good implementation to assess COPD patients' level of daily physical activity, to identify those with severely reduced levels of physical activity (such the 6MWT, or the 30-s chair stand test), and establish an exercise plan taking into account personal needs, preferences and personal goals to go along with the pharmacological treatment [47,51,52]. The ESWT, CWRCE and 6MWT are the commonest test used to assess COPD patients' level of physical activity; these are reliable tests to which patients respond and are familiarized with, they can be used in a multicentre trial setting, as they have good reproducibility and repeatability, and have an important intra class (IC) correlation and are significant predictors of mortality in COPD [14,22,53]. Particularly, ESWT has been reported to be more sensitive than other tests to therapeutic intervention in a systematic review, where protocol variations significantly affected performance in several studies [53].
This SRL and meta-analysis has some limitations, the main one is the existing differences between the studies on variables used to measure physical activity which, in some cases, makes comparison difficult. Furthermore, it should be taken into account that in some analyses different LAMA/LABA combinations were compared with different LAMA or LABA monotherapies, and also outcomes evaluation times were different between studies, ranging from 3 to 12 weeks. Another limitation is that statistical heterogeneity was high in some comparison, limiting the validity and the generalizability of these results. Despite these limitations, the use of LAMA/ LABA consistently improves exercise capacity and physical activity compared with placebo or monotherapy in most outcomes and combinations analysed. On the other hand, our study has the following strengths: a reasonable number of studies and patients available and their rigorous methodological quality, as none of the studies included showed high risk of bias in any item.

Conclusion
In conclusion, our review showed that LAMA/LABA combination therapy was superior to placebo and monotherapy in terms of evaluating exercise capacity and physical activity in patients with COPD in almost every comparison. Enhancing physical activity and exercise capacity in COPD patients might lead to improve their quality of life and minimize the burden of the disease.