Effect of manual diaphragmatic release on ventilatory function and functional capacity in elderly type 2 diabetic women: A randomised controlled trial

Abstract

Background:

Elderly individuals with type 2 diabetes mellitus (T2DM) often experience reduced ventilatory function and functional capacity, which negatively impact their quality of life (QoL), yet effective non-pharmacological interventions remain limited.

Objective:

To determine the effect of manual diaphragmatic release (MDR) added to aerobic training (AT) on ventilatory function, functional capacity, and QoL, compared to AT alone in elderly women with T2DM.

Methods:

Sixty elderly type 2 diabetic women were randomly assigned in equal numbers to either the MDR or control group. For 8 weeks, the MDR group received MDR in addition to AT, while the control group received AT alone. Outcomes included measures of forced vital capacity (FVC) and total lung capacity (TLC), 6 min walk test (6MWT), and Diabetic Quality of Life Questionnaire (DQoL). Between-group differences were analysed by ANCOVA.

Results:

At the end of the study, there were significant group effects for all outcome measures ($p<0.05$). Adjusted between-group differences favoured the MDR group for FVC (mean $\mathrm{\text{difference}}=3.17\%$; 95% CI: 1.07–5.27), TLC (mean $\mathrm{\text{difference}}=2.28\%$; 95% CI: 1.22–3.33), 6MWT distance (mean $\mathrm{\text{difference}}=22.58$ m; 95% CI: 10.17–34.99), and DQoL score (mean $\mathrm{\text{difference}}=0.94$; 95% CI: 0.31–1.57).

Conclusion:

Adding MDR to AT is more beneficial than AT only for enhancing ventilatory function, functional capacity and QoL in such elderly type 2 diabetic women. However, further studies with larger sample sizes and longer intervention durations are warranted to confirm these findings.

Introduction

Type 2 diabetes mellitus (T2DM), the adult form of diabetes, is a complex metabolic disorder primarily characterised by insulin resistance and $\beta$-cell dysfunction, which result in long-term hyperglycemia.¹ Elderly individuals aged $\ge 65$ years often experience the highest incidence rate of T2DM, reaching up to 21.8%.² Despite being more prevalent in men,³ women with T2DM usually present with higher rates of adverse effects and mortality.⁴ Impaired lung function is a common complication in individuals with T2DM,⁵ affecting approximately 23.7% of such cases,⁶ and most often manifests as a restrictive lung pattern⁷ due to respiratory muscle dysfunction resulting from hyperglycemia, inflammation, and oxidative stress.⁸ The presence of diabetes in the elderly can lead to greater deterioration in functional capacity than in non-diabetic individuals of the same age,⁹ owing to the progressive decline in skeletal muscle mass and strength,¹⁰ which decreases mobility levels¹¹ and quality of life (QoL).¹² The reduced mobility levels among elderly individuals with T2DM highlight the urgent need for alternative approaches to active exercise in managing diabetes-associated impairments.

One of the passive respiratory techniques that requires no physical activity is manual diaphragmatic release (MDR). Manual stretching of the diaphragm through MDR has been found to promote lung function in patients with both obstructive¹³ and restrictive lung disorders,¹⁴ in addition to enhancing functional capacity^13,15 and QoL¹⁶ in patients with obstructive lung disorders. Nevertheless, the benefits of MDR in patients with T2DM, specifically, have never been previously explored. To address this gap, this interventional study aimed to determine the effect of the manual MDR technique added to aerobic training (AT) on ventilatory function, functional capacity, and QoL, compared with AT alone in elderly women with T2DM.

Methods

This single-centre, parallel-group, randomised controlled trial was conducted in accordance with the CONSORT guidelines¹⁷ and adhered to the ethical principles outlined in the Declaration of Helsinki. The study was carried out between December 2024 and September 2025. Participants were recruited between December 2024 and March 2025 from the Endocrinology Department at Mansoura General Hospital in Egypt, through direct referrals from attending endocrinologists. Recruitment was limited to women because the collaborating clinic and physiotherapy unit predominantly serve elderly female patients, resulting in few eligible men being available during the recruitment period. Accordingly, the study objective and eligibility criteria were predefined for elderly women with T2DM. Of the 211 women screened for eligibility, 60 met the inclusion criteria and were enrolled in the study after providing written informed consent. All interventions were conducted at a private geriatric physical therapy unit equipped to accommodate elderly patients in Mansoura city in Egypt. The study protocol was reviewed and approved by the Ethics Committee for Human Scientific Study at Cairo University in Egypt and given the acceptance number P.T.REC/012/005439. With the assigned number of NCT06724744, the study has been registered through clinicaltrials.gov available at: https://clinicaltrials.gov/study/NCT06724744.

Sample size estimation

A pilot study was conducted with two groups, each comprising 12 women, to estimate the sample size based on the forced vital capacity (FVC) outcome measure. Pilot participants met the same eligibility criteria as the main trial and received the same intervention protocols for 8 weeks, but their data were not included in the final analysis. The study group, which received MDR and aerobic exercises, exhibited a post-study mean FVC (% of predicted value) of $78.73\pm 4.8\%$, while the control group, which received AT only, had a mean (% of predicted value) of $75.29\pm 4.1\%$. Inputting these values into the G*Power software (version 3.1, Germany) resulted in an estimated effect size (Cohen’s d) of 0.77. Using a two-tailed t-test for independent groups with a significance level ($\alpha =0.05$) and power ($1-\beta =0.80$), the required sample size was calculated to be 27 participants per group. Considering potential dropouts, a final sample size of 30 participants per group was adopted.

Randomisation, allocation and blinding

Simple randomisation using computer-generated numbers in Microsoft Excel software (version 2019, Microsoft Corporation, Redmond, WA, USA) was performed to equally allocate participants into two groups (1:1 allocation ratio). Allocation sequences were placed in sequentially numbered, sealed, opaque envelopes prepared by an independent endocrinologist. Each envelope was opened only after baseline assessments to ensure allocation concealment. The recruiters, three physiotherapists, were blinded to the allocation sequence. All outcome assessors, except the functional capacity assessor, as well as the statistician, were blinded to group allocation. Also, participants and supervisors of the interventions were aware of group assignments due to the nature of the interventions.

Participants

A total of 60 women with type 2 diabetes were included according to the following inclusion criteria: (a) A diagnosis of T2DM for more than 5 years; (b) An age range of 65–75 years was selected to include older adults who commonly exhibit diabetes-related ventilatory impairments while maintaining sufficient functional ability to safely participate in the interventions included in this trial; and (c) Restrictive lung function with total lung capacity (TLC) $<80\%$ of the predicted value.¹⁸ Exclusion criteria included male participants, other types of diabetes, smoking, chronic lung diseases, use of systemic or inhaled corticosteroids, diaphragmatic hernia, and musculoskeletal or neurological limitations to physical exercise.

Recruitment continued until the required sample size of 60 participants was reached. Participants were equally allocated to the MDR group, which received MDR plus AT, or the control group, which received aerobic exercises only. The study flow is presented in Fig. 1.

Fig. 1.

Participant flow chart. MDR $=$ Manual diaphragmatic release.

Evaluations

An initial assessment of all participants was performed by an endocrinologist to ensure eligibility for participation. Medical histories regarding any previous diseases or surgical interventions were documented. Current glycemic control levels, medications, and any associated comorbidities were also recorded. The baseline level of physical activity was documented using the validated 6-point physical activity scale.¹⁹ Additionally, anthropometric measurements, including weight, height, and body mass index, were obtained. All measures of outcomes were assessed at two points, baseline and after 8 weeks.

Primary outcome

Ventilatory function

Measures of FVC and TLC were assessed by a specialised technician according to the American Thoracic Society guidelines.²⁰ For FVC assessment, each participant was instructed to maximally inhale and exhale through the properly fixed mouthpiece of the lung function testing spirometer device (Spirobank II Basic, Italy). TLC was assessed using the whole-body plethysmography method. Participants breathed through a pneumotachograph while seated in a body plethysmograph (Q-Box, COSMED Srl, Italy). After achieving a stable end-expiratory level, the shutter was briefly closed, and participants performed gentle panting manoeuvres at 0.5–1.0 Hz to determine functional residual capacity. Once the shutter reopened, they performed a slow expiratory reserve volume manoeuvre followed by a slow inspiratory vital capacity manoeuvre. Functional residual capacity measurement was repeated up to five times or until three values varied by less than 5%. Residual volume was calculated as functional residual capacity minus expiratory reserve volume, and TLC was obtained by adding residual volume to the largest measured vital capacity. The highest values from three trials assessing FVC and TLC, expressed as percentages of the predicted measures, were used for statistical analysis. The normal percentage of both FVC and TLC is greater than 80% of the predicted values.¹⁸

Secondary outcomes

Functional capacity

As a reliable indicator of functional capacity in the elderly,²¹ all participants underwent the 6 min walk test (6MWT) following the standardisations of the American Thoracic Society.²² The test was first explained to each participant, while resting completely in a chair for 10 min before starting. The beginning and end of the testing area were marked by two cones placed 30 m apart in a long corridor. Each participant was instructed to walk as far as possible between the two points, while receiving standardised encouragement every minute and being monitored for any indications of test termination. Upon completion of the 6 min test, the total distance walked was recorded in meters. For elderly individuals, a distance of 386.5 m on the 6MWT is the cutoff value for limited functional capacity.²³

Quality of life

The Diabetic Quality of Life Questionnaire (DQoL) in its Arabic version, validated by Al-Qerem et al.,²⁴ was completed by each woman to assess her QoL. The Arabic DQoL comprises 29 items distributed across three domains: satisfaction, impact, and diabetes-related worries. Each item is rated on a 5-point Likert scale.²⁴ Domain-specific scores were calculated by averaging the items within each domain, and the overall DQoL score was determined as the mean of the three domain scores, with lower scores indicating poorer QoL and higher scores reflecting better QoL.²⁵

Interventions

Four physiotherapists, experienced in pulmonary rehabilitation, were responsible for developing, conducting, and supervising the interventions employed. Throughout the study, each participant’s blood glucose level was regularly measured prior to every session. Participants were also instructed to remain well hydrated before, during, and after the sessions. Interventions were performed in the morning, 2 h after breakfast, while wearing loose-fitting clothing. Adherence to anti-diabetic medications was strictly monitored and confirmed for all participants.

Aerobic training

For all women, 30–45 min of low-intensity aerobic cycling were performed three times weekly using an electrical cycle ergometer (MONARK 818 E, Sweden) for 8 weeks, following the standardisations of the American College of Sports Medicine.²⁶ The rate of perceived exertion on the 20-point Borg scale was maintained between 10 and 12 during cycling to ensure adherence to the low-intensity training zone.²⁷ Each session began with 10 min of self-stretching for the major muscle groups as a warm-up, followed by cycling exercises for 30 min during the first 4 weeks, which was progressed to 45 min in the last 4 weeks. 5 min of stretching exercises concluded each session for cool-down.

Manual diaphragmatic release

Only women in the MDR group received MDR interventions individually three times per week for 8 weeks. During each session, participants were placed in a crook-lying position, while the therapist stood at the head of the bed, facing the participant. The therapist positioned the hypothenar regions of both hands bilaterally beneath the costal cartilages of the 7th to 10th ribs. Participants were instructed to take deep inhalations as the therapist applied cephalic and lateral traction at the contact points to facilitate rib elevation. During expiration, the therapist increased the depth of manual contact beneath the costal margins while maintaining resistance.¹³ The MDR protocol was implemented progressively. In weeks 1–4, each session consisted of two sets of five deep breaths, which were increased to three sets of 10 repetitions per session during the final four weeks. To ensure procedural consistency, all four physiotherapists completed a standardised training workshop on the MDR protocol before participant treatment began, including supervised practice sessions and competency verification by the principal investigator. A written step-by-step procedural checklist was used during all sessions to standardise therapist hand placement, breathing coordination, and repetition structure. However, formal inter-rater reliability testing was not performed, which may have contributed to variability in technique.

Analysis of data

Statistical analyses were conducted using SPSS software (version 22; IBM Corp., Chicago, IL, USA). Data normality was confirmed using Shapiro–Wilk ($p>0.05$). Continuous variables are reported as means ± standard deviations and categorical variables as frequencies and percentages. Baseline characteristics were compared between groups using the independent two-tailed t-test for continuous variables and the chi-square test for categorical variables. Between-group comparisons of outcomes were performed using analysis of covariance (ANCOVA) with the post-study value as the dependent variable, the group as the fixed factor, and the corresponding baseline value as a covariate; thus, the reported mean difference (MD) represents baseline-adjusted effects rather than simple change-score difference. The assumption of homogeneity of regression slopes was assessed. Within-group changes were examined using paired t-tests. Between-group effects were expressed using the adjusted MD with 95% confidence intervals. Standardised effect sizes (Cohen’s d) were calculated from the adjusted mean differences²⁸ to quantify the magnitude of the treatment effect and interpreted as small ($d=0.2$), moderate ($d=0.5$), or large ($d=0.8$).²⁹ Missing post-study data were imputed using the expectation–maximisation algorithm,³⁰ and all analyses followed the intention-to-treat principle. Statistical significance was set at $p<0.05$.

Results

Two women dropped out of the MDR group based on their personal requests, and one from the control group due to travelling abroad. The total number of sessions actually attended in each group was calculated relative to the total planned sessions and revealed no significant difference between groups ($p>0.05$) (Table 1). Comparison of general, demographic, and clinical characteristics between groups also showed no significant differences ($p>0.05$) (Table 1). Adverse effects were monitored using a structured checklist at each session, and participants were also encouraged to report any symptoms spontaneously. No adverse effects, hyperglycemic or hypoglycemic events, or episodes of diabetic ketoacidosis were recorded as a result of the interventions.

Table 1.

Women’s characteristics.

Variables		MDR group ($n=30$)	Control group ($n=30$)	p-value
Age (years)		69.6 ± 2.97	70.2 ± 2.45	0.41
Weight (kg)		77.21 ± 5.35	78.51 ± 5.02	0.35
Height (cm)		166.64 ± 4.96	165.58 ± 5.15	0.34
BMI (kg/m2)		27 ± 2.21	27.51 ± 1.21	0.3
Marital status	Married	14 (46.66%)	16 (53.33%)	0.58
	Single	7 (23.33%)	4 (13.33%)
	Widowed	4 (13.33%)	7 (23.33%)
	Divorced	3 (10%)	2 (6.66%)
Highest education	Primary school	16 (53.33%)	13 (43.33%)	0.31
	Secondary school	10 (33.33%)	10 (33.33%)
	University	2 (6.66%)	6 (20%)
Duration of diabetes (years)		12.1 ± 3.96	12.58 ± 3.55	0.63
Glycemic control	HbA1c (%)	7.28 ± 0.66	7.3 ± 0.5	0.91
	RBS (mg/dL)	148.32 ± 21.04	141.27 ± 15.48	0.15
Mean arterial pressure (mmHg)		110.57 ± 10.51	111.86 ± 10.78	0.64
Resting heart rate (beat/minute)		80.1 ± 5.69	81.13 ± 5.69	0.49
Total blood cholesterol (mmol/L)		221.85 ± 20.3	216.51 ± 18.47	0.3
Associated co-morbidities	Hypertension	19 (63.33%)	23 (76.66%)	0.32
	Dyslipidemia	25 (83.33%)	20 (66.66%)	0.06
	Knee osteoarthritis	15 (50%)	13 (43.33%)	0.34
	Chronic kidney disease	9 (30%)	8 (26.66%)	0.7
Diabetes treatments	Oral hypoglycemic	23 (76.66%)	27 (90%)	0.21
	Insulin	11 (36.66%)	17 (56.66%)	0.87
Other medications	Anti-hypertensive	17 (56.66%)	23 (76.66%)	0.12
	Statins	21 (70%)	16 (53.33%)	0.11
	Non-steroidal analgesics	10 (33.33%)	13 (43.33%)	0.48
6-point physical activity scale		2.67 ± 1	2.86 ± 1.1	0.5
Attendance to sessions		91.07%	93.39%	0.83

Notes: Data are reported as means ± standard deviations, frequencies, and percentages. Continuous variables were analysed with the independent t-test, and categorical variables with the chi-square test. BMI $=$ Body mass index; HbA1c $=$ Glycated haemoglobin; MDR $=$ Manual diaphragmatic release; RBS $=$ Random blood sugar

Ventilatory function

For FVC, 50% of participants in the study group scored greater than 80%, compared to only 25.8% in the control group at post-study assessment. Following the intervention, ANCOVA test revealed a significant group effect for FVC ($F=9.16$, $p=0.004$, partial ${\eta }^2=0.14$), favouring the MDR group ($\mathrm{\text{MD}}=3.17\%$; 95% $\mathrm{\text{CI}}=1.07$–5.27) (Table 2). For TLC, 82.14% of participants in the study group scored greater than 80%, compared to 43.84% in the control group at post-study assessment. This was accompanied by a significant group effect for TLC ($F=18.76$, $p=0.001$, partial ${\eta }^2=0.25$) favouring the MDR group ($\mathrm{\text{MD}}=2.28\%$; 95% $\mathrm{\text{CI}}=1.22$–3.33) (Table 2).

Table 2.

Outcome measures.

Outcome		MDR ($n=30$)	Control group ($n=30$)	MD (95% CI)	Cohen’s d	p-value
FVC (%)	Pre-study	73.28 ± 7.03	71 ± 6.86			0.004**
	Post-study	78.77 ± 5.5*	75.59 ± 6.69*	3.17 (1.07, 5.27)	0.52
TLC (%)	Pre-study	74.6 ± 2.64	74.44 ± 2.41			0.001**
	Post-study	82.16 ± 2.57*	79.87 ± 1.37*	2.28 (1.22, 3.33)	1.11
6MWT (m)	Pre-study	397.21 ± 53.69	394.72 ± 57.96			0.001**
	Post-study	441.96 ± 22.27*	419.38 ± 26.42*	22.58 (10.17, 34.99)	0.92
DQoL	Pre-study	2.32 ± 1.09	2.41 ± 1.11			0.004**
	Post-study	3.91 ± 1.11*	2.97 ± 1.26	0.94 (0.31, 1.57)	0.79

Notes: Data are expressed as mean values with standard deviations. A p-value is for ANCOVA group effect. *Indicates significant change compared to pre-study, while **indicates a significant group effect. CI $=$ Confidence interval; Cohen’s $d=\mathrm{\text{Cohen}}$’s effect size; DQoL $=$ Diabetes quality of life questionnaire; FVC $=$ Forced vital capacity; MD $=$ Mean difference; MDR $=$ Manual diaphragmatic release; 6MWT $=$ Six-minute walk test; TLC $=$ Total lung capacity.

Functional capacity

At baseline, 42.85% of women in the MDR group did not exceed 386.5 m on the 6MWT, compared to 48.27% in the control group. At post-study assessment, only one participant (3.5%) in the MDR group failed to exceed 386.5 m, compared to three participants (10.3%) in the control group. Comparing the adjusted post-study scores of 6MWT between groups was significant ($F=13.3$, $p=0.001$, partial ${\eta }^2=0.19$) favouring the MDR group ($\mathrm{\text{MD}}=22.58$ m; 95% $\mathrm{\text{CI}}=10.17$ to 34.99) (Table 2).

Quality of life

Following the intervention, the percentage of participants with a DQoL score above the median value of 3 increased from 35.71% to 89.28% in the MDR group, compared with an increase from 37.93% to 58.62% in the control group. ANCOVA revealed a significant group effect for the adjusted post-study DQoL scores ($F=9.01$, $p=0.004$, partial ${\eta }^2=0.14$). The MD was 0.94 (95% CI: 0.31–1.57), favouring the MDR group (Table 2).

Discussion

This study is the first to investigate the effect of MDR intervention on ventilatory function, functional capacity, and QoL in elderly women with T2DM. Eight weeks of MDR added to AT led to significantly greater improvements in lung function measures (FVC and TLC), functional capacity, and QoL compared to the control group, which received AT only.

The main finding of this trial is that MDR resulted in significantly greater post-study FVC (Cohen’s $d=0.52$, $p=0.004$) and TLC (Cohen’s $d=1.11$, $p=0.001$) compared to the control group. However, the adjusted MD for FVC exceeded the lower threshold of the established MCID range (2–6%),³¹ and the lower bound of its 95% CI (1.07) fell below this range. Therefore, the clinical significance of the finding should be interpreted with caution. Although the MCID of TLC for restrictive lung disorders has not been previously established, a between-group difference of 2.28%, corresponding to the large effect size of 1.11, may be considered clinically meaningful. Similar to our findings, MDR has been shown to significantly improve FVC after 12 weeks in individuals exposed to occupational lung hazards³² and after 30 days in patients with non-specific low back pain.¹⁴ MDR was found to alleviate the restrictive chest wall pattern by effectively enhancing diaphragmatic motion³³ and chest expansion,³⁴ which are often reduced in diabetic individuals due to altered collagen/elastin ratio³⁵ and weakened respiratory muscles.³⁶

Functional capacity is a major concern in elderly diabetic individuals. This trial found that MDR resulted in greater post-study scores of the 6MWT compared to the control group (Cohen’s $d=0.92$, $p=0.001$). In context, the MD between groups at post-study (i.e., 22.58 m) exceeded the pre-determined MCID for older adults of 17.8 m,³⁷ confirming the clinical effectiveness of MDR. Similar to our results, MDR has previously been shown to enhance 6MWT distance in various patient populations, including those with chronic obstructive lung disorders,¹³ post-COVID syndrome,³⁸ and individuals exposed to work-related respiratory hazards.³²

Elderly diabetic individuals typically present with reduced QoL.³⁹ This trial utilised the diabetes-specific DQoL questionnaire to assess QoL. After 8 weeks, MDR led to significant greater DQoL score at post-study compared to the control group (Cohen’s $d=0.79$, $p=0.004$). Although the MCID for DQoL has not been previously established, a mean between-group difference of 0.94 (representing 18.8% of the total DQoL score), together with its corresponding effect size of 0.79, may be considered clinically meaningful. Similarly, MDR has been found to enhance QoL in patients with gastroesophageal reflux disease after 4 weeks,⁴⁰ and in asthmatic patients after 6 weeks.¹⁶

Study limitations

This study has several limitations. It was conducted at a single centre and included only elderly women, which limits the generalisability of the findings to men and to broader and more diverse populations. Cultural, biological, and behavioural differences between sexes may influence both diaphragmatic mechanics and response to manual therapy; therefore, the present findings should not be directly extrapolated to male patients with T2DM. The eight-week intervention period may not have been sufficient to fully capture changes in all outcomes. A further limitation is the absence of inter-rater reliability assessment among the four therapists delivering MDR, which may have introduced variability in the execution of the technique. Additionally, the lack of blinding for participants and intervention supervisors could have introduced bias. Moreover, the 6MWT assessor was not blinded to group allocation, which may have introduced performance or measurement bias despite the use of standardised instructions and encouragement procedures. Although baseline comorbidities, medications, and physical activity levels were comparable between groups, changes in medication regimens or lifestyle behaviours during the intervention period could not be fully controlled and may have influenced outcomes. Finally, social and financial stressors may have influenced QoL assessments.

Conclusion

Adding MDR to AT produced greater improvements in ventilatory function, functional capacity, and QoL compared with AT alone in elderly women with T2DM. While these findings suggest that MDR may be a useful adjunct in rehabilitation programs for this population, the results should be interpreted with caution because the study was limited by its single-centre design, relatively short intervention period, and inclusion of women only. Larger multi-centre trials including both sexes with longer follow-up are needed to confirm and generalise these findings.

Acknowledgments

Authors are grateful to endocrinologists in the recruitment sites and physiotherapists working in the place of intervention.

Conflict of Interest

The authors assert that they have no conflicts of interest to disclose.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Author Contributions

SL Elgayar contributed to conceptualisation, study design, methodology, data collection, formal analysis, data interpretation, paper drafting, critical revision of the paper, and final approval of the version to be published.

ORCID

Saher Lotfy Elgayar https://orcid.org/0009-0004-5284-6267