Asthma; Breathing exercises; Exercise tolerance; Respiratory function tests; Diaphragm; Respiratory muscles; Respiratory therapy

Asthma is a long-term airway disorder in which inflammation, narrowing of the bronchi, and excess mucus act together to restrict airflow and produce the familiar symptoms of wheezing, breathlessness, and cough [1,2]. It is among the most common chronic lung conditions and has a heavy clinical and economic burden, with an estimated 262 million people affected worldwide [3,4]. Although drug therapy—mainly bronchodilators and inhaled corticosteroids—remains the cornerstone of management, a large proportion of patients continue to have inadequately controlled disease. That shortfall has directed growing interest toward nonpharmacological adjuncts intended to bolster lung mechanics, increase functional capacity and ease day-to-day living for those with the disease [5,6,7].

In particular, diaphragmatic stretching and respiratory training through nonpharmacological approaches have shown promise as interventions that can improve asthma patients' pulmonary function and exercise tolerance [8]. Patients with asthma often develop dysfunction of their diaphragm as well as altered breathing patterns, both of which lead to ineffective ventilation and worsened dyspnea [9]. Techniques that release tension in the diaphragm and the surrounding fascia are intended to increase diaphragmatic strength, flexibility, and excursion and thereby to improve ventilation by allowing deeper and more controlled breathing [10]. This can decrease airway resistance, improve gas exchange, and relieve symptoms such as wheezing. It also helps move patients away from shallow, chest-based breathing to diaphragmatic, abdominal breathing, improving overall respiratory mechanics [11,12].

In contrast, respiratory training aims to strengthen respiratory muscles, including the diaphragm and intercostal muscles, and to optimize breathing patterns. Maneuvers of this kind, such as pursed-lip and diaphragmatic breathing, lower airway resistance, ease dyspnea and extend exercise tolerance by building strength and coordination in the respiratory musculature [13,14]. By engaging in slower, deeper breaths, respiratory training decreases the perception of breathlessness, especially with exertion, and enhances aerobic work. This results in increased endurance, improved adherence, and reduced exercise-induced bronchoconstriction [15,16].

Asthma imposes a psychological and physical burden in addition to impairing respiratory function; interventions such as diaphragmatic stretching and respiratory training are therefore of increasing interest. Stress is a recognized trigger of asthma attacks, and many patients are also highly anxious because of their reduced capacity for physical labor or exercise [17]. By promoting relaxation and greater control over breathing, diaphragmatic stretching may reduce the stress and anxiety that can precipitate an asthma attack when a patient is emotionally or physically strained. Respiratory training is an additional aid because it makes patients more comfortable and confident in managing symptoms while they are engaged in physical exertion, which may reduce episodes of panic and support better mental health [18].

Although diaphragmatic stretching and respiratory training have each shown promise for improving pulmonary function and exercise tolerance in patients with asthma, direct head-to-head evaluation within a single population remains scarce, and the relative benefit of a diaphragm-focused technique over general respiratory-muscle training remains unclear. The mechanisms by which these interventions influence day-to-day asthma symptoms are also not fully understood, and evidence on their effects on exercise tolerance under real-life conditions is limited. To help address this gap, the present study aimed to compare the effects of diaphragmatic stretching and respiratory training on pulmonary function, exercise tolerance, asthma control, and shortness of breath in adults with mild-to-moderate asthma. The specific objectives were to quantify the change in each outcome within each group after a 14-week program and to determine whether the two approaches differ in terms of the magnitude of improvement. We hypothesized that both interventions would improve the measured outcomes and that diaphragmatic stretching would result in greater gains than respiratory training would.

2.1. Study design

This was a two-arm comparative quasi-experimental study with pre- and postintervention assessments. Participants were enrolled consecutively and assigned alternately to one of two intervention groups according to their order of presentation.

2.2. Ethics approval

The Ethics Review Committee of Hussain Memorial Hospital approved the protocol (No. HMH-23-ERC-51), and institutional permission to recruit participants was granted by the same hospital, at which all participants were enrolled. Written informed consent preceded enrollment in every case, and the work adhered throughout the Declaration of Helsinki. Data were collected over eight months between January and August 2024.

2.3. Study setting

The study was carried out in Lahore, the provincial capital of Punjab, the most populous province of Pakistan. Participants were recruited from the outpatient department of Hussain Memorial Hospital, a private hospital in the city that provides outpatient, inpatient, and intensive care services together with round-the-clock diagnostic and emergency coverage. Patients were approached at the time of routine outpatient visits.

2.4. Participant recruitment

Adults between 18 and 65 years of age were eligible for whom a physician had confirmed asthma under the 2023 Global Initiative for Asthma (GINA) recommendations [19]. The diagnosis rested on a consistent history of fluctuating respiratory symptoms alongside documented variability in expiratory airflow, taken as present where postbronchodilator FEV1 increased by ≥ 12% and ≥ 200 mL above baseline, which is consistent with the GINA diagnostic criteria. Participants were required to have clinically stable mild-to-moderate asthma, with no recent acute exacerbation and a stable prescribed asthma treatment regimen at enrollment. Asthma severity was classified during clinical assessment according to the GINA treatment step required to maintain symptom control.

Participants were excluded if they had severe asthma, chronic obstructive pulmonary disease (COPD), or other significant respiratory or cardiovascular comorbidities that could independently affect pulmonary function or exercise tolerance. Pregnant or lactating women and individuals unable or unwilling to participate in the prescribed intervention and follow-up assessments were also excluded.

Although the predefined eligibility age range was 18–65 years, the observed age range of the enrolled participants was 30–65 years. This was not an additional exclusion criterion; rather, it reflected the age distribution of patients who met the eligibility criteria in the study setting during the recruitment period.

Usually, prescribed asthma treatment was to continue unchanged across the 14-week intervention, and the protocol introduced no deliberate alteration to routine medication. However, detailed baseline medication types and inhaled corticosteroid doses were not systematically recorded in the study dataset. Consequently, potential differences in inhaled corticosteroid exposure between groups could not be formally assessed or adjusted for in the analysis. This issue was therefore considered a potential source of residual confounding and has been acknowledged as a limitation of the study.

2.5. Sampling technique and sample size

Participants were recruited consecutively from the outpatient department of the study hospital as they presented and met the inclusion and exclusion criteria. They were then allocated alternately, in order of presentation, to two treatment groups of 50 patients each: one group received the diaphragmatic stretching technique, and the other received respiratory training. Because allocation was not random, baseline characteristics were compared between the groups and are reported below. A priori calculations in G*Power 3.1.9.7 (Heinrich Heine University, Düsseldorf, Germany) fixed the target at 100 or 50 per arm. Lacking directly comparable data, we posited a moderate difference between arms in the change in FEV1 (Cohen's d = 0.60); 45 per arm satisfied 80% power at a two-sided alpha of 0.05, rounded up to 50 to absorb attrition.

2.6. Study instruments

Study instruments were selected to assess pulmonary function, exercise tolerance, asthma control, and shortness of breath. Baseline sociodemographic and clinical characteristics were recorded using a structured data-collection form.

Spirometry followed American Thoracic Society/European Respiratory Society (ATS/ERS) technical standards on an MIR Spirobank II Smart portable spirometer (MIR Medical International Research, Rome, Italy), calibrated to the manufacturer’s specifications ahead of each testing session. Testing was performed seated with a nose clip in place and standardized verbal prompts guiding every maximal forced expiratory maneuver. Three or more maneuvers were obtained per participant and screened for acceptability and repeatability against the ATS/ERS criteria. The highest technically acceptable FEV1 and FVC values were retained for analysis. FEV1 and FVC were recorded as absolute values. Spirometry was performed without routine bronchodilator administration so that the FEV1 and FVC reflected prebronchodilator function at both assessment points; bronchodilator responsiveness testing was used only to support the diagnosis at screening. Participants were instructed to avoid short-acting bronchodilator use for at least 4 hours before testing. These procedures were based on the ATS/ERS 2019 technical standard for spirometry [20].

Exercise tolerance was gauged by a six-minute walk test (6MWT) under the standardized American Thoracic Society protocol along a straight, 30-meter-long indoor corridor. Each participant walked without pause for six minutes at the quickest pace they could sustain, receiving the encouragement scripted in that protocol. The distance covered was logged in meters. Greater 6MWT distances indicate better functional exercise capacity [21].

Asthma control was assessed using the five-item Asthma Control Test (ACT). The total ACT score ranged from 5 to 25, with higher scores indicating better asthma control. The ACT was administered in its validated form as described by Nathan et al. [22].

Breathlessness was quantified with the University of California, San Diego Shortness of Breath Questionnaire (UCSD SOBQ): 24 items, each rated 0 to 5, for a range of 0–120. Larger totals indicate more severe dyspnea, so a decrease in score signals relief. Approximately 5 points was taken as the minimum clinically important difference (MCID) [23], a threshold established in patients with fibrotic interstitial lung disease and chronic obstructive pulmonary disease rather than asthma and adopted here for an asthma-specific value.

Participants also maintained a study-specific diary to document adherence to the prescribed intervention and any adverse effects experienced during the study period. The diary and interview schedule were pilot tested before the main study commenced to assess the clarity, feasibility, and ease of completion. Adherence was expressed as the share of prescribed sessions attended, and adverse events were logged by occurrence, type and severity and by whether they prompted interruption or withdrawal. These diary records were not collated into the study analysis dataset and could not be retrieved for the present analysis; therefore, adherence and adverse events are not reported.

2.7. Data collection

The screening was performed jointly by a researcher and a physician, who reviewed each patient’s medical history, carried out a physical examination, and reviewed the relevant investigations to confirm eligibility against the inclusion and exclusion criteria. Candidates first received a brief Urdu-language information sheet, setting out the purpose of the work, its likely benefits and risks, ethics approval and the safeguards protecting confidentiality. Once written informed consent had been obtained, a brief face-to-face interview lasting 10 to 15 minutes was conducted in the patient’s local language. Assessment was performed at baseline and again at the completion of the 14-week program so that the two interventions could be weighed against each other before and after treatment.

2.8. Interventional protocol

Diaphragmatic stretching intervention was performed by a trained physiotherapist and was designed to promote diaphragmatic excursion, improve lung expansion, and enhance respiratory muscle function. Every 30-minute session comprised 10 sets of 5 repetitions separated by 2-minute rests. The intensity of the stretching progressively increased weekly, as tolerated by the participants. The intervention was administered three times per week for 14 weeks, with the physiotherapist supervising and delivering the stretching technique during each session. Exercise intensity progressed according to the participants’ tolerance and response [24,25,26].

In the respiratory training group, the intervention combined pursed-lip breathing with inspiratory muscle training. Each session lasted 30 minutes, matching the session duration of the diaphragmatic stretching group. Sessions were performed three times per week for 14 weeks under the supervision of the same trained physiotherapist, who monitored technique, recorded adherence, and ensured appropriate progression.

A threshold loading device (Threshold IMT) delivered the inspiratory muscle training, starting resistance being fixed at 30% of each participant’s maximal inspiratory pressure (MIP). MIPs were assessed using a calibrated respiratory pressure meter, and the highest technically acceptable value was used to determine the initial training load. The resistance was reassessed weekly and progressively increased by 5% of the MIP, as tolerated by the participant, to provide progressive respiratory muscle loading. Each session ran to 10 sets of 6 repetitions, again with 2 minutes of rest. Pursed-lip breathing was incorporated to improve breathing efficiency and reduce shortness of breath.

The inspiratory muscle training load was individualized and progressively adjusted according to participants' tolerance and performance during follow-up sessions [15,27,28,29].

Both interventions included educational sessions to reinforce asthma management techniques and encourage adherence. Throughout the 14 weeks, progress in pulmonary function, exercise tolerance, and overall asthma control was monitored at the supervision visits, and the outcome measures reported below were formally assessed at baseline and at 14 weeks [30].

2.9. Statistical analysis

Descriptive statistics are summarized as frequencies and percentages for categorical variables and as the means with standard deviations for continuous variables. The distribution of continuous variables was checked with the Shapiro–Wilk test before analysis; the within-person differences were normally distributed for every group and outcome except for asthma control in the respiratory training group (Shapiro–Wilk p = 0.022) and FEV1 in the diaphragmatic stretching group (Shapiro–Wilk p = 0.013), which were analyzed by the Wilcoxon signed-rank test. Group comparisons of categorical data were performed using the Pearson chi-square test without continuity correction, with Fisher's exact test given alongside 2 × 2 tables. Changes within each arm were tested by paired t tests or Wilcoxon signed-rank tests, and differences between arms were tested by independent-samples t tests or Mann–Whitney U tests, according to distribution. Because the within-person changes in the ACT score in the respiratory training group and in the FEV1 in the diaphragmatic stretching group were not normally distributed, the corresponding between-arm comparisons were rerun nonparametrically as a sensitivity check. Baseline comparability of the outcome variables was tested before the between-group analyses. The primary between-group comparison was of the within-person change scores; because allocation was not randomized and the groups differed at baseline, posttreatment values were additionally compared by analysis of covariance with the corresponding baseline value as a covariate for sensitivity analysis. Thresholds for the minimal clinically important difference (MCID) were set in advance: a gain of ≥ 0.20 L in FEV1 and in FVC, ≥ 30 m in six-minute walk distance, ≥ 3 points on the ACT and a fall of ≥ 5 points on the UCSD SOBQ. Effect sizes appear as Cohen’s dz within arms and as Cohen’s d for the between-arm difference in change scores. Responder proportions are given for every outcome. FEV1 served as the primary outcome and anchored the sample size calculation, with the other four being secondary. Significance was two-sided at p < 0.05. Analysis was carried out in IBM SPSS Statistics, version 26 (IBM Corp., Armonk, NY, USA). This study is reported in line with the TREND statement for nonrandomized interventional studies.

3.1. Patients’ sociodemographic and baseline clinical characteristics

The mean age in the diaphragmatic stretching arm was 48.68 years (SD = 10.80), spanning 30 to 65 years (Table 1); the respiratory training arm averaged 45.02 years (SD = 9.46) across a marginally closer 30 to 63 years. All 100 participants had complete data, with 50 valid cases in each group and no missing values. Age did not separate the arms (mean difference 3.66 years, 95% CI −0.37 to 7.69; Welch’s t = 1.80; p = 0.075).

Table 1 shows the sociodemographic characteristics of the diaphragmatic stretching (DS) and respiratory training (RT) groups (n = 100). Sex was balanced across the arms (Pearson chi-square p = 0.181; Fisher's exact p = 0.265): the DS group was 66% female and 34% male, whereas the RT group was 78% female and 22% male. Geographic location differed between the groups (Pearson chi-square p = 0.045; Fisher's exact p = 0.071), with 58% of the DS group being from rural areas and 62% of the RT group being from urban areas. Because the study was not randomized, this imbalance in geographic location was a confounder and was considered when the between-group comparisons were interpreted; it was not entered as a covariate because the adjusted analyses were specified to adjust for baseline outcome values only.

Neither employment status (p = 0.915) nor marital status (p = 0.264) separated the arms. Compared with the diaphragmatic stretching group, the respiratory training group included a greater proportion of married participants (40% vs 26%), although the gap was not significant. Disease duration was likewise comparable (p = 0.915), with 30% of participants in the diaphragmatic stretching group presenting with a disease duration of 1–5 years compared with 24% in the respiratory training group. Longer disease durations were similarly distributed in both groups (6–10 years, 40% vs 44%; 11–15 years, 20% vs 20%; >15 years, 10% vs 12%).