Chronic obstructive pulmonary disease is a respiratory disease characterized by persistent airflow limitation and progressive decline in lung function. The World Health Organization (WHO) predicts that COPD will become the third leading cause of death worldwide by 2030, with its disease burden being particularly pronounced in older populations (1). A large-scale epidemiological study in China reported a COPD prevalence of 13.7% among adults aged 40 years and older (2). In older COPD patients, reduced lung elasticity and accelerated alveolar structural damage lead to an accelerated decline in forced expiratory volume in 1 s (FEV₁), with an annual decline rate approximately 2.5-fold higher than that in healthy individuals (3). This significantly exacerbates the decline in exercise endurance and deterioration of quality of life.

Current clinical management of COPD centers on medications such as bronchodilators (4). However, these drugs have limited efficacy in improving lung function (FEV₁/FEV₁%) and cannot reverse respiratory muscle weakness (e.g., maximal inspiratory pressure (PImax) < 60 cmH₂O) or declining exercise endurance (5). Pulmonary rehabilitation (PR), a non-pharmacological intervention, improves functional status through physical activities such as aerobic exercise and resistance training (Report). Multiple studies demonstrate that exercise-based PR enhances exercise capacity, quality of life, and reduces dyspnea more effectively than non-exercise programs in COPD patients (6–9). Consequently, PR is recognized as the most cost-effective therapeutic strategy (10). Inspiratory muscle training (IMT), a key component of PR, significantly improves respiratory function and exercise capacity in COPD patients (11–13). A report by the American Thoracic Society (ATS) suggests that IMT, used as an independent intervention or added to PR for patients with respiratory muscle weakness, may offer benefits (10). However, existing systematic reviews are inconsistent regarding the efficacy of PR combined with IMT: Cochrane reviews indicate that the superiority of PR combined with IMT versus PR alone in improving dyspnea, functional exercise capacity, and quality of life remains unclear (14–16).

This study aims to systematically compare the intervention effects of PR combined with IMT versus PR on quality of life [St. George’s Respiratory Questionnaire (SGRQ)], 6-min walk distance (6MWD), respiratory muscle strength (PImax), and lung function (forced expiratory volume in 1 s (FEV₁), forced expiratory volume in 1 s% predicted (FEV₁%)) in older COPD patients through meta-analysis. Subgroup analyses will explore how intervention parameters (e.g., frequency, duration) influence treatment efficacy, providing evidence-based insights for developing personalized non-pharmacological intervention protocols for elderly COPD patients.

This study adhered to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA 2020) guidelines (17) and was registered on PROSPERO (CRD420251010168).

This meta-analysis included 9 randomized controlled trials (582 older COPD patients) to investigate the effects of PR combined with IMT versus PR alone on lung function and exercise endurance. Primary outcomes showed no significant differences between PR combined with IMT and PR alone in improving 6MWD (SMD = 0.15, 95% CI: −0.11–0.42) or SGRQ scores (SMD = −0.19, 95% CI: −0.38–0.01). Although PR combined with IMT demonstrated a moderate effect size for maximal inspiratory pressure (PImax; SMD = 0.78, 95% CI: 0.44–1.13), the evidence was downgraded due to moderate heterogeneity (I² = 48.7%). For lung function, FEV₁ and FEV₁% predicted values trended positively (SMD = 0.50) but did not reach statistical significance, attributable to high heterogeneity (I² = 72.9%) and a CI crossing the null effect (95% CI: −0.37–1.36).

While prior studies reported PR combined with IMT improved FEV₁ in COPD patients, our meta-analysis found FEV₁ trended positively (SMD = 0.50) but did not reach statistical significance (95% CI: −0.37–1.36, I² = 72.9%) (15, 16). This highlights potential heterogeneity in patient populations, notably driven by this study’s unique inclusion of patients with GOLD stage II–IV and older age. Bodduluri et al. analyzed airway trees in 7,641 participants from the COPD Gene cohort and found that T-Slope–a quantitative measure of airway lumen narrowing–decreased progressively with GOLD stage severity (Jonckheere-Terpstra p = 0.04) (35). Critically, T-Slope was independently associated with both FEV₁ [β = 0.13 (95% CI: 0.10–0.15) L; p < 0.001] and annual FEV₁ decline rate [β = −4.50 (95% CI: −7.32 to −1.67) mL⋅year^–1; p = 0.001] (35), confirming irreversible structural damage (e.g., airway fibrosis, alveolar destruction) as the core driver of airway obstruction (36). This structural resistance may dissociate gains in respiratory muscle strength (PImax) from functional outcomes (6MWD/SGRQ) through dual physiological constraints. In moderate-severe COPD (GOLD II–IV), irreversible parenchymal destruction such as airway fibrosis and alveolar loss reduces pulmonary elastic recoil, compromising compensatory capacity for increased physiological dead space and dynamic hyperinflation despite improved respiratory muscle efficiency (37, 38). Concurrently in older patients, elevated respiratory muscle workload heightens oxygen consumption, diverting metabolic resources from locomotor muscles via the “respiratory metaboreflex” phenomenon–an effect exacerbated in advanced GOLD stages due to elevated baseline ventilatory demands (39, 40). This constraint may be partially mitigated by targeted inspiratory muscle warm-up prior to training, which has been shown to significantly enhance inspiratory strength and efficiency, thereby reducing respiratory oxygen demand and improving functional exercise capacity in moderate-to-severe COPD patients with inspiratory muscle weakness (41). This aligns with the pathological characteristics of older patients and those with high GOLD stages in our study: their profound loss of pulmonary elastic recoil–exacerbated by age-related respiratory muscle degeneration–limits the structural benefits of PR combined with IMT (e.g., airway remodeling reversal). This finding aligns with the pathological characteristics of elderly patients and those with advanced GOLD stages: their profound loss of pulmonary elastic recoil–exacerbated by age-related respiratory muscle degeneration–limits the structural benefits of PR combined with IMT (e.g., airway remodeling). PR and IMT predominantly promote functional adaptations (e.g., enhanced muscle efficiency), which cannot reverse established parenchymal damage (42). Notably, our meta-analysis included fewer studies reporting FEV₁/FEV₁% outcomes, and substantial methodological heterogeneity across trials (e.g., varying training protocols) likely contributed to inconsistent cumulative effects of interventions. Thus, while PR combined with IMT showed moderate PImax benefits (SMD = 0.78), the lack of significant FEV₁ gains and high heterogeneity (I² = 72.9%) hinder definitive conclusions on respiratory mechanics optimization.

Current evidence suggests that standalone respiratory training improves PImax in COPD patients (43–45). However, PR combined with IMT does not confer additional PImax benefits over PR alone (15). Ammous et al. reported that PR combined with IMT increased PImax by 11.46 cmH₂O (95% CI: 7.42–15.50) versus PR alone, but this fell short of the minimal clinical important difference (MCID) of 17.2 cmH₂O (16). This discrepancy with our findings may stem from our study’s strict inclusion of patients aged ≥ 55 years, a population vulnerable to age-related respiratory muscle decline and sarcopenia (46, 47). Ammous’ subgroup analysis found no differences between PR combined with IMT and PR alone across duration, frequency, session duration, and weekly cumulative duration [short-term (< 4 weeks), mid-term (4–7 weeks), long-term (≥8 weeks)] (16). In contrast, our PImax subgroup analysis showed that PR combined with IMT significantly and robustly improved PImax versus PR alone when intervention duration was ≥ 12 weeks (SMD = 0.866, 95% CI: 0.579–1.153), session duration ≥ 60 min (SMD = 0.922, 95% CI: 0.666–1.177), or weekly cumulative duration ≥ 180 min (SMD = 0.922, 95% CI: 0.666–1.177). Notably, we found no significant associations between PImax and duration, frequency, weekly cumulative duration, or total intervention time (p > 0.05). These divergent meta-analytic results likely reflect heterogeneity in subgroup definitions based on duration, frequency, session duration, and weekly cumulative duration. Given the substantial variability in IMT protocols across included studies–particularly regarding session duration and weekly cumulative duration–no definitive recommendations for COPD patients based on these parameters can be made at this time. Future research, including a priori stratification by baseline PImax to address its influence on treatment response, is needed to determine optimal PR combined with IMT parameters for PImax improvement.

Although our study observed a positive trend for PR combined with IMT on FEV₁ and FEV₁%, the small number of included studies and high heterogeneity undermined the reliability of these results. Standalone PR and IMT both improve exercise capacity and lung function in COPD patients (9, 16). Notably, clinical significance for such improvements requires meeting MCID thresholds (e.g., 25–30 m for 6MWT; 4 units for SGRQ).This observation may be explained by the intrinsic mechanisms of PR. Güneş et al. demonstrated that 6 weeks of PR (3 sessions/week) significantly increased diaphragm thickness in COPD patients (48). The endurance training components of PR (e.g., walking, cycling) indirectly activate respiratory muscles through increased ventilatory demand, potentially blunting the specific benefits of IMT. Notably, the meta-regression revealed no significant associations between intervention parameters (duration, frequency, session duration, weekly cumulative duration) and outcomes (6MWD/PImax)–contrasting with subgroup analyses indicating that these factors heterogeneously influenced treatment efficacy. This apparent discrepancy likely arises because subgroup analyses identified threshold-dependent effects (e.g., significant PImax improvements only at intervention durations ≥ 12 weeks or weekly cumulative durations ≥ 180 min), while meta-regression modeled parameters as continuous variables, potentially obscuring non-linear relationships. Furthermore, substantial heterogeneity in intervention protocols (evident in the I² = 86.1% for 6MWD in < 6-week subgroups) may have compromised statistical power to detect meta-regression associations (49). Lötters et al. (50) recommended IMT as an adjunct for patients with severe baseline inspiratory muscle weakness (PImax < 60% predicted), whereas the lack of baseline stratification in our study may have diluted treatment effects due to heterogeneous patient populations. This underscores the need for future studies to perform stratified analyses based on baseline respiratory muscle function to establish optimal IMT eligibility criteria.

This study has several limitations that warrant attention. The incomplete reporting of baseline FEV₁ across studies prevented robust stratification by COPD severity (GOLD stages), contributing to unmeasured heterogeneity in subgroup analyses. The absence of baseline PImax stratification likely obscured differential effects in clinically distinct subgroups (e.g., severe vs. moderate respiratory muscle weakness). Predominant short-term follow-up durations (median ≤ 12 weeks in most trials) preclude conclusions about long-term sustainability (e.g., ≥1 year follow-up data). Heterogeneity in IMT protocols–session duration, frequency, cumulative weekly duration–complicated synthesis of optimal intervention parameters and dose-response relationships. Furthermore, exclusion of non-English publications and limited non-Western representation may restrict applicability to diverse healthcare contexts, as non-English studies (e.g., Chinese) may report distinct intervention efficacies. Finally, the exclusive focus on RCT designs may overlook implementation evidence from real-world settings.

Pulmonary rehabilitation combined with IMT improved PImax in older COPD patients compared to PR alone, but no advantages of PR combined with IMT were observed for exercise capacity (e.g., 6MWD) or lung function (e.g., 6MWD). For clinicians aiming to enhance respiratory muscle strength in older COPD patients, this meta-analysis suggests that PR combined with IMT lasting ≥ 12 weeks, with individual sessions exceeding 60 min and total weekly intervention time reaching 180 min, may be beneficial. However, these recommendations are based on low-to-very-low quality evidence, and the optimal protocol requires further validation through large-scale, standardized clinical trials.