Effects of respiratory muscle training on respiratory function, trunk control and fatigue in multiple sclerosis Authors: Marta de la Plaza San Frutos, Alberto Bermejo Franco, Doris Dalila Albán Calle, Fausto David Pérez Villena, Federico Salniccia, Cecilia Estrada Barranco Published: 2026 ABSTRACT Multiple sclerosis (MS) is a chronic condition affecting the central nervous system, often causing fatigue, postural instability, and respiratory dysfunction. This study aims to evaluate the effects of a combined inspiratory and expiratory respiratory muscle training (RMT) program on respiratory function, trunk control, balance, and clinical fatigue in patients with MS. A quasi-experimental clinical trial was conducted with 27 participants diagnosed with MS. The experimental group (EG) underwent respiratory muscle training (RMT) using the Threshold Inspiratory Muscle Trainer (IMT) and the Positive Expiratory Pressure (PEP) devices (Respironics®, Phillips), in addition to standard therapy, while the control group (CG) received standard therapy only. Pre- and post-intervention measurements included spirometry, maximum inspiratory pressure (MIP), maximum expiratory pressure (MEP), ultrasound for abdominal muscles thickness, diaphragm thickness and diaphragmatic excursion, and scales for balance (Berg Balance Scale) and trunk control (Trunk Impairment Scale). The EG showed significant improvements in the FEV1/FVC ratio, MEP, and trunk control scores, particularly in the dynamic and coordination subscales of the Trunk Impairment Scale. However, no significant changes were observed in fatigue levels or abdominal muscles thickness. The RMT program enhanced respiratory function and trunk control in MS patients, demonstrating its potential as a therapeutic intervention. Further studies are needed to explore the long-term effects of RMT in MS rehabilitation. Introduction Multiple sclerosis (MS) is a chronic autoimmune disease of the central nervous system that affects approximately 1.3 million people worldwide, with a female-to-male ratio of 3:11,2. Although its etiology remains unclear, MS prevalence shows marked geographical variability, being higher in Europe, Australia, and North America (50–200 per 100,000 population)3. The disease course and clinical manifestations are highly heterogeneous; however, demyelinating lesions frequently involve the brainstem, spinal cord, and periventricular white matter4, leading to a wide range of neurological impairments. Among the most disabling impairments in people with MS are muscle weakness, balance disorders, impaired postural control, and fatigue, all of which significantly limit functional independence and participation in daily activities5,6. Trunk control is often compromised due to altered neuromuscular activation of deep stabilizing muscles and delayed anticipatory postural responses7, which increases fall risk and negatively impacts functional mobility. Given the central nervous system involvement characteristic of MS, impaired neural drive and inefficient activation of trunk and respiratory muscles may contribute to respiratory dysfunction, further exacerbating deficits in postural control and balance. Respiratory dysfunction is common in MS and is characterized by reduced inspiratory and expiratory muscle strength, altered breathing patterns, and decreased ventilatory efficiency8. These impairments may negatively influence core stability, as the diaphragm plays a key role not only in respiration but also in lumbopelvic stabilization through its synergistic activation with the abdominal and pelvic floor muscles9,10. Consequently, respiratory muscle weakness may compromise trunk control and balance, particularly during tasks requiring higher postural demands. In addition to inspiratory dysfunction, expiratory muscle weakness has been widely reported in people with MS11, impairing cough effectiveness and secretion clearance, and increasing the risk of respiratory complications such as aspiration pneumonia and acute respiratory failure, particularly in advanced stages of the disease3,4,11. These findings highlight the clinical relevance of targeting both inspiratory and expiratory muscles within rehabilitation programs, not only to improve respiratory efficiency but also to address broader functional limitations associated with the disease. In this context, fatigue represents one of the most prevalent and disabling symptoms in MS, affecting up to 80% of patients, and is commonly defined as a subjective and persistent feeling of physical and/or mental exhaustion that is disproportionate to activity level and not relieved by rest5. In the present study, fatigue is conceptualized as clinical or symptomatic fatigue, assessed using the Modified Fatigue Impact Scale (MFIS), which reflects the perceived impact of fatigue on physical, cognitive, and psychosocial functioning. Previous studies have suggested complex interactions between fatigue, postural control, balance, and respiratory function3,7,11. However, existing evidence is inconsistent, and the relationship between clinical fatigue and objective measures of balance and trunk control remains insufficiently understood. Approximately 50% of individuals with MS experience at least one fall per year, primarily due to impaired postural control and reduced core stability12–14. Core stability relies on the coordinated activation of the diaphragm, pelvic floor muscles, abdominal wall, and posterior spinal musculature, including the multifidus and thoracolumbar fascia12. Adequate activation of this muscular system is essential for maintaining balance and efficient movement, particularly in populations with neuromuscular dysfunctions10. Several studies have highlighted the potential benefits of respiratory muscle training (RMT) in neuromuscular disorders, suggesting improvements in respiratory function and postural control10,12,13. However, despite growing interest, there remains a lack of conclusive evidence regarding the effects of combined inspiratory and expiratory muscle training on trunk control, balance, and clinical fatigue in individuals with MS, particularly within structured home-based interventions14. Therefore, the aim of this study was to evaluate the effects of a home-based RMT program targeting both inspiratory and expiratory muscles on respiratory function, trunk control, balance, deep abdominal muscle activation, and clinical fatigue in people with MS. Methodology Study design A quasi-experimental clinical trial was conducted with a control group, with pre- and post-intervention measurements. Approval was obtained from the research ethics committee of the Hospital Clínico San Carlos, Madrid (Spain) COD:23/685-E), The research was conducted according to the framework of the Helsinki declaration15. Registration was made in ClinicalTrials.gov (NCT06405451) on 8 August 2024. To ensure the methodological quality of the study, the criteria established by the TREND guide were followed16. Procedure The sample was obtained through the Multiple Sclerosis Foundation of Madrid (FEMM) (Spain) in the context of a collaboration agreement between the European University, the Madrid College of Pharmacists, Sanitas Foundation and Real Madrid Foundation. The interested parties were summoned for information about the research, the signing of the informed consent form and the first assessment. Participants were recruited from clinical and rehabilitation centers through clinician referrals and outreach efforts. All recruitment procedures were conducted in coordination with the participating institutions to ensure that eligibility criteria were satisfied. The study followed a non-randomized design. Group assignment was determined by the clinical setting in which participants received rehabilitation care, with participants attending the Multiple Sclerosis Foundation allocated to the control group and those treated at the Sanitas rehabilitation center allocated to the experimental group. This approach reflects the organization of rehabilitation services. Although randomization was not feasible, strict inclusion criteria, the use of validated outcome measures, and the assessment of baseline differences between groups were applied. Participants were adults with multiple sclerosis aged 18–70 years, with a disease duration of more than two years, stable medical treatment for at least six months prior to the intervention, no cognitive impairment, the ability to understand instructions, a Mini-Mental State Examination score ≥ 24, and a Kurtzke Expanded Disability Status Scale (EDSS) score ≤ 6. Exclusion criteria included musculoskeletal, cardiovascular, respiratory, or metabolic disorders; disease exacerbation or symptom worsening within the previous three months; and treatment with intravenous or oral corticosteroids within six months before or during the intervention. Outcomes To assess the morphology of the core musculature, the ultrasound scanner (VSCAN PRIM®; GE HealthCare Technologies, Inc.; Chicago, Illinois, United States) was used. Ultrasound measurements were taken of the Transversus Abdominis (TA) at rest and contraction, Internal Oblique (IO) and External Oblique (EO) at rest and contraction, Diaphragm at rest (DX rest) and contraction (DX cont) and Diaphragmatic Excursion (DEX). To avoid bias during measurement we used the RUSI standard, which takes three measurements of each of the scans and records the mean. Fatigue was assessed using the MFIS, a validated self-report questionnaire that evaluates the perceived impact of fatigue on daily functioning in individuals with multiple sclerosis. The MFIS consists of 21 items across three domains: physical, cognitive, and psychosocial. Participants also completed quantitative balance scales such as the Berg Balance Scale (BBS)17 and the Trunk Impairment Scale (TIS)18, which are validated and reliable tools for assessing balance and trunk control in individuals with multiple sclerosis. The BBS assesses balance through 14 items with a time delay between 10 and 15 min for its application, it provides a value according to its performance in each item from 0 (not achievable) to 4 (normal), with a total score ranging from a minimum of 0 (not able to maintain balance) to a maximum of 56 points indicating (good ability) to maintain balance17. The TIS assesses trunk control with 17 items scored between 0 (minimum control), 2 and 3 (maximum control) for a total of 23 points, and consists of three sub-scales that consider the ability of Static Balance (TIS-Static), Dynamic Balance (TIS- DYN) and Coordination (TIS-COOR)18. Respiratory function was assessed by a spirometer (Easyone Connect®; ndd Medical Technologies; Zürich, Switzerland), from which the Forced Vital Capacity (FVC), the maximum expiratory force in the first second (FEV1) and the ratio between FEV1/FVC were obtained19 a and interpreted according to established international guidelines. Respiratory muscle strength was measured using the Micro RPM® (Micro‑Direct, Inc.; Lewiston, Maine, United States) device which measures Maximun Inspiratory Pressure (MIP), Maximun Expiratory Pressure (MEP)20. All assessments (including spirometry, MIP/MEP, and fatigue) were conducted by an experienced physiotherapist blinded to group allocation and not involved in the delivery of the intervention. The supervised RMT sessions were conducted by a different therapist. Intervention Both the control group (CG) and the experimental group (EG) received 45 min of conventional treatment based on the guidelines developed by the National Institute for Health and Care Excellence (NICE) twice a week21. To ensure consistency and replicability of the intervention, the training protocol was structured according to the FITT principle (Frequency, Intensity, Time, and Type), as described below: The RMT was performed twice daily, six days per week, for 6 weeks (Frequency). Each session consisted of five sets of 10–15 repetitions, with 30 s of rest between sets (Time). The intensity of the training was individualized and progressively adjusted on a weekly basis, ranging between 40% and 60% of each participant’s maximum inspiratory and expiratory pressure (MIP and MEP), in accordance with the protocol proposed by Klefbeck et al.22. The intervention was designed as a home-based respiratory muscle training program using the Threshold IMT and PEP devices (Respironics®, Philips; Murrysville, Pennsylvania, United States). Training was performed in a seated position, with feet flat on the floor, without back support, and with the trunk positioned at a 90° angle to the hips. A nasal clip was used to prevent air leakage during exercises. The CG received 45 min of conventional treatment focused on postural control, stretching, mobility, and general conditioning exercises two days a week. Training was performed in a seated position, with feet flat on the floor, without back support, and with the trunk positioned at a 90° angle to the hips. A nasal clip was used to prevent air leakage during exercises (Type). Adherence during the home-based intervention was monitored by direct contact with patients. Participants For the calculation of the sample size, the software G*Power 3.1 was used. An effect size of d = 1.0 was assumed, based on previous evidence showing clinically relevant improvements in respiratory muscle strength after respiratory muscle training in people with multiple sclerosis and other neurological populations. Considering a standard deviation of 0.50 L, a significance level of 5% (α=0.05) and a statistical power of 80%, a sample size of 12 participants per group was established to detect significant differences between the two groups. Although participants covered a wide age range (18 to 70 years), strict inclusion criteria described previously were applied. In addition, no statistically significant differences in age were observed between the control and experimental groups (Mann–Whitney U = 79.5, p = 0.608), indicating homogeneity in this demographic variable and reducing the risk of age acting as a confounding factor. Statistical analysis To analyze the efficacy of the intervention, a comparison was made between the initial and final scores of the CG and the EG. Given that the data did not follow a normal distribution, non-parametric tests were employed. Specifically, the Wilcoxon signed-rank test was used to assess within-group differences (pre- and post-intervention scores) for both groups. For between-group comparisons, change scores were calculated for each outcome as the difference between post-intervention and baseline values (Δ = post − pre). These change scores were then compared between groups using the Mann–Whitney U test. Effect sizes were interpreted as small (r = 0.1), medium (r = 0.3), and large (r = 0.5), according to conventional thresholds proposed by Cohen (1988) and commonly applied to r statistics23. A significance level of p < 0.05 was established for all analyses. Results Thirty participants were recruited in the FEMM, of whom twenty-seven were selected to participate. Three were discarded as they did not meet the inclusion criteria. The subjects were divided into two groups, fifteen of which were assigned to the CG and twelve to the EG. Eleven subjects in the EG completed the intervention, one did not complete the intervention due to family circumstances and three subjects in the CG did not attend the final measurement (Fig. 1). Fig. 1Participants selection. Participants selection. The characteristics of the sample and baseline comparisons between groups are presented in Tables 1 and 2, respectively. No statistically significant differences were observed in FVC, FEV1, FEV1/FVC, MIP, MEP, MFIS, BBS, DX rest, DX cont, or DEX at baseline. However, the experimental group presented significantly lower baseline values in trunk impairment (TIS and TIS DYN). Table 1Descriptive characteristics of the sample.CGEGn = 15M(RI)n = 12M(RI)Age49 (20)46 (14.5)Height169 (10)168 (13)Weight71 (14)76.5 (38.25)Type of MSPP = 3; SP: 1; RR: 11PP = 2; SP: 1; RR: 9SexF = 11; M = 4F = 6; M = 6EDSS4.0 (1.63)4.5 (1.88)CG: Control Group; EG: Experimental Group; MS: Multiple Sclerosis; F: Female; M: Male; P-P: Primary Progressive; S-P: Secondary Progressive; R-R: Relapsing-Remitting Multiple Sclerosis; EDSS: Expanded Disability Status Scale. Table 1Descriptive characteristics of the sample.CGEGn = 15M(RI)n = 12M(RI)Age49 (20)46 (14.5)Height169 (10)168 (13)Weight71 (14)76.5 (38.25)Type of MSPP = 3; SP: 1; RR: 11PP = 2; SP: 1; RR: 9SexF = 11; M = 4F = 6; M = 6EDSS4.0 (1.63)4.5 (1.88)CG: Control Group; EG: Experimental Group; MS: Multiple Sclerosis; F: Female; M: Male; P-P: Primary Progressive; S-P: Secondary Progressive; R-R: Relapsing-Remitting Multiple Sclerosis; EDSS: Expanded Disability Status Scale. Descriptive characteristics of the sample. CG: Control Group; EG: Experimental Group; MS: Multiple Sclerosis; F: Female; M: Male; P-P: Primary Progressive; S-P: Secondary Progressive; R-R: Relapsing-Remitting Multiple Sclerosis; EDSS: Expanded Disability Status Scale. Table 2Baseline comparison of outcome variables.CG (n = 15)EG (n = 12)Initial M (IR)Initial M (IR) p FVC88 (21)91 (36.5)0.696FEV181 (24.5)80 (36.75)0.696FEV1/FVC80 (14.5)71.5 (24.75)0.067MIP56 (43)39.5 (31.5)0.329MEP62 (40.5)51.5 (36.75)0.751DX rest0.15 (0.05)0.15 (0.14)0.787DX cont0.3 (0.11)0.28 (0.07)0.525DEX5 (3.27)5.96 (1.81)0.291TIS15.5 (4)12 (6)0.008*TIS STATIC4 (1)3 (2)0.060TIS DYN7.5 (1)6 (4)0.046*TIS COOR4 (2)3 (2)0.128BBS42.5 (8.75)40 (26)0.265MFIS46 (15)43.5 (27.2)0.826CG: Control Group; EG: Experimental Group; FVC: forced vital capacity; FEV1: forced expiratory volume in 1 s; MIP: maximal inspiratory pressure; MEP: maximal expiratory pressure; DX rest: diaphragm thickness at rest; DX cont: diaphragm thickness during contraction; DEX: diaphragmatic excursion; TIS: Trunk Impairment Scale; TIS STATIC: Trunk Impairment Scale static; TIS DYN: Trunk Impairment Scale dynamic; TIS COOR: Trunk Impairment Scale coordination; BBS: Berg Balance Scale; MFIS: Modified Fatigue Impact Scale; M (IR): median (interquartile range). Between-group comparisons were performed using the Mann–Whitney U test. Statistical significance was set at p < 0.05. Table 2Baseline comparison of outcome variables.CG (n = 15)EG (n = 12)Initial M (IR)Initial M (IR) p FVC88 (21)91 (36.5)0.696FEV181 (24.5)80 (36.75)0.696FEV1/FVC80 (14.5)71.5 (24.75)0.067MIP56 (43)39.5 (31.5)0.329MEP62 (40.5)51.5 (36.75)0.751DX rest0.15 (0.05)0.15 (0.14)0.787DX cont0.3 (0.11)0.28 (0.07)0.525DEX5 (3.27)5.96 (1.81)0.291TIS15.5 (4)12 (6)0.008*TIS STATIC4 (1)3 (2)0.060TIS DYN7.5 (1)6 (4)0.046*TIS COOR4 (2)3 (2)0.128BBS42.5 (8.75)40 (26)0.265MFIS46 (15)43.5 (27.2)0.826CG: Control Group; EG: Experimental Group; FVC: forced vital capacity; FEV1: forced expiratory volume in 1 s; MIP: maximal inspiratory pressure; MEP: maximal expiratory pressure; DX rest: diaphragm thickness at rest; DX cont: diaphragm thickness during contraction; DEX: diaphragmatic excursion; TIS: Trunk Impairment Scale; TIS STATIC: Trunk Impairment Scale static; TIS DYN: Trunk Impairment Scale dynamic; TIS COOR: Trunk Impairment Scale coordination; BBS: Berg Balance Scale; MFIS: Modified Fatigue Impact Scale; M (IR): median (interquartile range). Between-group comparisons were performed using the Mann–Whitney U test. Statistical significance was set at p < 0.05. Baseline comparison of outcome variables. CG: Control Group; EG: Experimental Group; FVC: forced vital capacity; FEV1: forced expiratory volume in 1 s; MIP: maximal inspiratory pressure; MEP: maximal expiratory pressure; DX rest: diaphragm thickness at rest; DX cont: diaphragm thickness during contraction; DEX: diaphragmatic excursion; TIS: Trunk Impairment Scale; TIS STATIC: Trunk Impairment Scale static; TIS DYN: Trunk Impairment Scale dynamic; TIS COOR: Trunk Impairment Scale coordination; BBS: Berg Balance Scale; MFIS: Modified Fatigue Impact Scale; M (IR): median (interquartile range). Between-group comparisons were performed using the Mann–Whitney U test. Statistical significance was set at p < 0.05. To analyse the effect of the intervention on variables related to respiratory function, baseline and final scores were compared in both groups. The CG did not show significant changes in any of the respiratory variables studied, including DX morphology and DEX. In contrast, the EG showed significant improvements in the FEV1/FVC ratio and MEP after the intervention (Table 3). Table 3Respiratory outcomes.CGEGInitialM(IR)FinalM (IR) p InitialM(IR)FinalM (IR) p FVC88(21)81 (26)0.66691 (36.5)89.5 (40.5)0.609FEV181 (24.5)81 (23.5)0.81380 (36.75)87 (45.5)0.388FEV1/FVC80 (14.5)74 (11)0.42371.5 (24.75)81.5 (20.75) 0.003 MIP56 (43)54 (51.5)0.91639.5 (31.5)54 (40.25)0.814MEP62 (40.5)59 (32.5)0.90651.5 (36.75)74 (31.25) 0.002 DX rest0.15 (0.05)0.16 (0.05)0.4090.15 (0.14)0.16(0.14)0.371DX cont0.3 (0.11)0.28 (0.11)0.5040.28 (0.09)0.28 (0.07)0.874DEX5 (3.27)5.18 (2.02)0.6255.96 (1.81)6.45 (1.45) 0.002 FVC: forced vital capacity; FEV1: forced expiratory volume in 1 s; MIP: maximal inspiratory pressure; MEP: maximal expiratory pressure; DX rest: diaphragm thickness at rest; DX cont: diaphragm thickness during contraction; DEX: diaphragmatic excursion; M (IR): median (interquartile range). Statistical significance was set at p < 0.05. Table 3Respiratory outcomes.CGEGInitialM(IR)FinalM (IR) p InitialM(IR)FinalM (IR) p FVC88(21)81 (26)0.66691 (36.5)89.5 (40.5)0.609FEV181 (24.5)81 (23.5)0.81380 (36.75)87 (45.5)0.388FEV1/FVC80 (14.5)74 (11)0.42371.5 (24.75)81.5 (20.75) 0.003 MIP56 (43)54 (51.5)0.91639.5 (31.5)54 (40.25)0.814MEP62 (40.5)59 (32.5)0.90651.5 (36.75)74 (31.25) 0.002 DX rest0.15 (0.05)0.16 (0.05)0.4090.15 (0.14)0.16(0.14)0.371DX cont0.3 (0.11)0.28 (0.11)0.5040.28 (0.09)0.28 (0.07)0.874DEX5 (3.27)5.18 (2.02)0.6255.96 (1.81)6.45 (1.45) 0.002 FVC: forced vital capacity; FEV1: forced expiratory volume in 1 s; MIP: maximal inspiratory pressure; MEP: maximal expiratory pressure; DX rest: diaphragm thickness at rest; DX cont: diaphragm thickness during contraction; DEX: diaphragmatic excursion; M (IR): median (interquartile range). Statistical significance was set at p < 0.05. Respiratory outcomes. 0.16 (0.14) FVC: forced vital capacity; FEV1: forced expiratory volume in 1 s; MIP: maximal inspiratory pressure; MEP: maximal expiratory pressure; DX rest: diaphragm thickness at rest; DX cont: diaphragm thickness during contraction; DEX: diaphragmatic excursion; M (IR): median (interquartile range). Statistical significance was set at p < 0.05. The effect of the intervention on balance, fatigue and ultrasound variables of the abdominal musculature was analysed. The EG showed significant differences in global TIS and TIS-COOR scores. However, no significant differences were found in abdominal muscle thickness measured by ultrasound, either at rest or during contraction, and fatigue levels remained unchanged. Similarly, the CG did not show significant changes in any balance, fatigue, or ultrasound variables of the abdominal musculature (Table 4). Table 4Trunk stability outcomes.CGEGInitialM(IR)FinalM(IR) p InitialM(IR)FinalM(IR) p TIS 15.5 (4)13 (5.75)0.12212 (6)16 (1) 0.002 TIS STATIC 4 (1)4 (0.75)13 (2)4 (0)0.076 TIS DYN 7.5 (1)6 (3.75)0.0726 (4)8 (2)0.064 TIS COOR 4 (2)2.5 (1.75)0.0683 (2)4 (1) 0.024 BBS 42.5 (8.75)41.5 (9.25)0.87840 (26)46 (30)0.53 MFIS 46 (15)37 (39)0.24943.5 (27.2)38 (39.75)0.158 Tr rest 0.31 (0.12)0.3 (0.13)0.7500.15 (0.05)0.18 (0.07)0.630 Tr cont 0.45 (0.34)0.44 (0.26)0.2800.25 (0.33)0.28 (0.26)0.080 EO rest 0.44 (0.43)0.43 (0.44)0.9000.3 (0.35)0.28 (0.33)0.390 EO cont 0.63 (0.44)0.62 (0.38)0.6900.4 (0.28)0.42 (0.26)0.690 IO rest 0.57 (0.41)0.57 (0.39)0.7500.48 (0.33)0.49 (0.33)0.130 IO cont 0.85 (0.42)0.85 (0.46)0.6900.57 (0.32)0.58 (0.38)0.750TIS: Trunk Impairment Scale; TIS STATIC: Trunk Impairment Scale static; TIS DYN: Trunk Impairment Scale Dynamic; TIS COOR: Trunk Impairment Scale Coordination; BBS: Berg Balance Scale; MFIS: Modified Fatigue Impact Scale; TR rest: Transverse thickness at rest; TR cont: Transverse thickness at contraction; EO rest: External Oblique Thickness at rest; EO cont: External Oblique thickness at contraction; IO rest: Internal Oblique thickness at rest; IO cont: Internal Oblique thickness at contraction; M: median; IR: interquartile range. Statistical significance value p < 0.05. Table 4Trunk stability outcomes.CGEGInitialM(IR)FinalM(IR) p InitialM(IR)FinalM(IR) p TIS 15.5 (4)13 (5.75)0.12212 (6)16 (1) 0.002 TIS STATIC 4 (1)4 (0.75)13 (2)4 (0)0.076 TIS DYN 7.5 (1)6 (3.75)0.0726 (4)8 (2)0.064 TIS COOR 4 (2)2.5 (1.75)0.0683 (2)4 (1) 0.024 BBS 42.5 (8.75)41.5 (9.25)0.87840 (26)46 (30)0.53 MFIS 46 (15)37 (39)0.24943.5 (27.2)38 (39.75)0.158 Tr rest 0.31 (0.12)0.3 (0.13)0.7500.15 (0.05)0.18 (0.07)0.630 Tr cont 0.45 (0.34)0.44 (0.26)0.2800.25 (0.33)0.28 (0.26)0.080 EO rest 0.44 (0.43)0.43 (0.44)0.9000.3 (0.35)0.28 (0.33)0.390 EO cont 0.63 (0.44)0.62 (0.38)0.6900.4 (0.28)0.42 (0.26)0.690 IO rest 0.57 (0.41)0.57 (0.39)0.7500.48 (0.33)0.49 (0.33)0.130 IO cont 0.85 (0.42)0.85 (0.46)0.6900.57 (0.32)0.58 (0.38)0.750TIS: Trunk Impairment Scale; TIS STATIC: Trunk Impairment Scale static; TIS DYN: Trunk Impairment Scale Dynamic; TIS COOR: Trunk Impairment Scale Coordination; BBS: Berg Balance Scale; MFIS: Modified Fatigue Impact Scale; TR rest: Transverse thickness at rest; TR cont: Transverse thickness at contraction; EO rest: External Oblique Thickness at rest; EO cont: External Oblique thickness at contraction; IO rest: Internal Oblique thickness at rest; IO cont: Internal Oblique thickness at contraction; M: median; IR: interquartile range. Statistical significance value p < 0.05. Trunk stability outcomes. TIS: Trunk Impairment Scale; TIS STATIC: Trunk Impairment Scale static; TIS DYN: Trunk Impairment Scale Dynamic; TIS COOR: Trunk Impairment Scale Coordination; BBS: Berg Balance Scale; MFIS: Modified Fatigue Impact Scale; TR rest: Transverse thickness at rest; TR cont: Transverse thickness at contraction; EO rest: External Oblique Thickness at rest; EO cont: External Oblique thickness at contraction; IO rest: Internal Oblique thickness at rest; IO cont: Internal Oblique thickness at contraction; M: median; IR: interquartile range. Statistical significance value p < 0.05. The Table 5 presents the median, interquartile range (IQR), Mann-Whitney U test results, and effect sizes (r) for the post-pre differences (Δ) across multiple variables. Significant differences were observed in several outcomes. In respiratory function, ΔFEV1/FVC showed a large effect size (r = 0.59, p = 0.003), indicating a substantial improvement in the experimental group. The most notable change was found in ΔMEP (r = 0.74, p < 0.001). Additionally, ΔTIS and ΔTIS-COOR both showed large effect sizes (r = 0.75 and r = 0.58, respectively), highlighting significant improvements in trunk impairment and coordination within the experimental group (p < 0.001 and p = 0.004, respectively). The variable ΔTIS-DYN also demonstrated a significant improvement, with a large effect size (r = 0.52, p = 0.010). For ΔDEX, a significant differences were observed was observed (r = 0.55, p = 0.007). No significant differences were found in fatigue or abdominal muscle thickness. Table 5Outcomes changes.M (IR) cgM (IR) EGU de Mann-Whitney p EFFECT SIZE (r)ΔFEV1/FVC0.00 (8.00)10.00 (10.00)24.0000.0030.59 ΔMEP 1.00 (14.00)12.00 (9.00)10.000 < 0.001 0.74 ΔTIS 0.00 (4.00)4.00 (3.00)8.500 < 0.001 0.75 ΔTIS DYN 0.00 (3.00)2.00 (3.00)28.000 0.010 0.52 ΔTIS Coor 0.00 (1.00)0.00 (1.00)26.000 0.004 0.58 ΔDEX 0.40 (0.82)-0.02 (0.93)25.000 0.007 0.55 Δ: post–pre difference; FEV1/FVC: ratio of forced expiratory volume in 1 s to forced vital capacity; MEP: maximal expiratory pressure; DEX: diaphragmatic excursion; TIS: Trunk Impairment Scale; TIS DYN: Trunk Impairment Scale, dynamic subscale; TIS COOR: Trunk Impairment Scale, coordination subscale; M (IR): median (interquartile range). Effect sizes are reported as r and were calculated as r = Z/√N for Mann–Whitney U tests. Statistical significance was set at p < 0.05. p-values < 0.001 are reported as thresholds. Table 5Outcomes changes.M (IR) cgM (IR) EGU de Mann-Whitney p EFFECT SIZE (r)ΔFEV1/FVC0.00 (8.00)10.00 (10.00)24.0000.0030.59 ΔMEP 1.00 (14.00)12.00 (9.00)10.000 < 0.001 0.74 ΔTIS 0.00 (4.00)4.00 (3.00)8.500 < 0.001 0.75 ΔTIS DYN 0.00 (3.00)2.00 (3.00)28.000 0.010 0.52 ΔTIS Coor 0.00 (1.00)0.00 (1.00)26.000 0.004 0.58 ΔDEX 0.40 (0.82)-0.02 (0.93)25.000 0.007 0.55 Δ: post–pre difference; FEV1/FVC: ratio of forced expiratory volume in 1 s to forced vital capacity; MEP: maximal expiratory pressure; DEX: diaphragmatic excursion; TIS: Trunk Impairment Scale; TIS DYN: Trunk Impairment Scale, dynamic subscale; TIS COOR: Trunk Impairment Scale, coordination subscale; M (IR): median (interquartile range). Effect sizes are reported as r and were calculated as r = Z/√N for Mann–Whitney U tests. Statistical significance was set at p < 0.05. p-values < 0.001 are reported as thresholds. Outcomes changes. Δ: post–pre difference; FEV1/FVC: ratio of forced expiratory volume in 1 s to forced vital capacity; MEP: maximal expiratory pressure; DEX: diaphragmatic excursion; TIS: Trunk Impairment Scale; TIS DYN: Trunk Impairment Scale, dynamic subscale; TIS COOR: Trunk Impairment Scale, coordination subscale; M (IR): median (interquartile range). Effect sizes are reported as r and were calculated as r = Z/√N for Mann–Whitney U tests. Statistical significance was set at p < 0.05. p-values < 0.001 are reported as thresholds. Discussion Findings from this study suggest that RMT in individuals with MS can be effective in improving respiratory function, trunk control, and diaphragmatic functionality. The effect size indicated that this intervention was particularly relevant in enhancing the FEV1/FVC ratio, MEP, TIS, and its sub-variables, including TIS-DYN and TIS-COOR, as well as MIP in the EG. It should be noted that the experimental group presented lower baseline values in trunk impairment (TIS and TIS DYN) compared to the control group, likely related to the non-randomized allocation based on treatment setting. This difference should be considered when interpreting trunk-related outcomes. In the present study, participants in the intervention group performed respiratory muscle training for six weeks, and the observed improvements in the FEV1/FVC ratio align with those reported by Martín Sánchez et al.24.Both studies support the hypothesis that low- to moderate-resistance inspiratory muscle training (20%–60%) can enhance pulmonary function and ventilatory mechanics. Therefore, RMT may increase the ability to expel air from the lungs more rapidly and effectively by improving respiratory muscle efficiency and overall ventilatory mechanics7. Our results show an improvement in the FEV1/FVC ratio, while no statistically significant differences were observed in FEV1 and FVC values. Although the FEV1/FVC ratio is commonly used to detect airflow obstruction, in this context its increase should be interpreted as reflecting improved ventilatory mechanics24. The findings of this study suggest an improvement in pulmonary function following the intervention. Previous studies, such as those by Martín Sánchez et al.24 and Elisabeth Westerdahl et al.26, did demonstrate statistically significant improvements in FVC. This discrepancy may be attributed to the participants in Martín Sánchez’s study undergoing inspiratory muscle training for 12 weeks, while the participants in Westerdahl’s study used a PEP device for eight weeks, suggesting that increasing the duration of the intervention may favorably modify FVC and FEV124–26. Furthermore, the values for MIP in this study did not show significant changes after the intervention, whereas an improvement in maximal expiratory pressure (MEP) was evident in the intervention group. Previous literature presents contradictory results regarding this aspect; for example, the systematic review published by Rietberg et al.8, which evaluated inspiratory and expiratory muscle training interventions, also reported no changes in MIP, in contrast to studies like that of Huang et al.27, which indicated that daily inspiratory muscle training with progressive resistance significantly improves MIP. Some authors have suggested that expiratory muscles may adapt more rapidly to resistance training than inspiratory muscles, although the underlying mechanisms remain incompletely understood and evidence is still limited8,27,28. In terms of DEX, the present study found a significant diaphragmatic displacement toward inspiration following the intervention. This aligns with the findings of Tanaka Montoya et al.29, which suggest that diaphragmatic movement contributes to the exchange of a greater volume of air during inspiration and expiration. Our findings indicate that the intervention improved the ventilatory efficiency of the diaphragm during inspiratory contraction. The improvement observed in MEP coincides with the systematic review published by Ferreira et al.30, which highlights the importance of RMT in enhancing inspiratory and expiratory strength in persons with MS and amyotrophic lateral sclerosis. The review included nine articles and a total of 194 persons, revealing statistically significant improvements in MIP, MEP, and FEV1 compared to the control group, concluding that RMT may serve as an adjunct therapy in the rehabilitation of neurodegenerative diseases, improving ventilatory function and respiratory strength. Furthermore, the improvement in MEP values not only reflects enhanced respiratory function but is also linked to the person´s physical capacity and quality of life31. There was no statistical significance concerning abdominal muscle thickness or diaphragm thickness at contraction and at rest, in line with the results of Şahin Hamza et al.32, who found no difference in diaphragm thickness between 45 persons with MS and healthy subjects, indicating that persons with MS and healthy individuals have similar diaphragm thickness. This suggests that the basic structure of the diaphragm is not affected by MS. However, healthy subjects showed a higher level of change and a significantly greater rate of thickening compared to individuals with MS, suggesting that, although diaphragm thickness is similar, the ability of the diaphragm to contract and expand during breathing is reduced in individuals with MS, indicating that diaphragmatic dysfunction relates more to function than to structure. This may reflect alterations in neuromuscular activation, motor unit recruitment, or timing and coordination of diaphragmatic contraction, rather than true structural hypertrophy or atrophy. According to our results, the intervention based on RMT was associated with improvements in both respiratory function (FEV1/FVC and MEP) and trunk control (TIS total score and subscales), suggesting that RMT may concurrently influence these two domains. Cetisli Korkmaz et al.33, in their study evaluating trunk control comparing healthy individuals and those diagnosed with MS, found that TIS scores were highly correlated with dynamic, coordination, and total sub-scores. These findings support the relationship between respiratory function and trunk control34 and are in line with the hypothesis that RMT may contribute to improvements in trunk control and balance. On the other hand, our results did not show significant changes in BBS scores in either group. While both the TIS and the BBS are validated and reliable tools for assessing postural control and balance in individuals with MS, they differ in scope and sensitivity. The TIS assesses selective trunk movements and coordination and has been shown to be particularly sensitive to changes following core-specific interventions33. In contrast, the BBS evaluates more complex, functional balance tasks related to daily activities, which may require longer or more comprehensive interventions to produce measurable changes35. This distinction may explain why improvements in trunk control as measured by the TIS did not immediately translate into functional balance gains detected by the BBS. RMT did not result in significant improvements in clinical fatigue as measured by the MFIS. This finding is consistent with previous studies suggesting that fatigue in MS is a multifactorial symptom influenced by central nervous system mechanisms, psychological factors, sleep disturbances, and disease-related processes, and may therefore be less responsive to isolated physical interventions such as respiratory muscle training6,34. Moreover, changes in perceived fatigue may require longer interventions or multimodal approaches targeting both physical and central components of fatigue6. Previous studies have suggested that fatigue in MS is more strongly associated with balance impairments than with respiratory function34. The lack of changes in balance, despite improvements in trunk control, indicates that addressing fatigue may require the addition of a more targeted intervention, specifically designed to complement RMT. The findings presented in this study have significant clinical implications. The training program undertaken by individuals with MS was home-based and independent, supervised only by physical therapists, which enhances its feasibility and reproducibility. Implementing programs to improve balance at home presents the challenge of fall risk during unsupervised exercises35. The proposed training program involves simple exercises performed in a seated position, allowing all persons to participate regardless of their baseline balance status, without increasing fall risk, and yet significantly improving not only respiratory function but also trunk control. Clinical relevance and future research directions The improvements observed in trunk control, as reflected by increases in the TIS total score and its subcomponents, suggest potentially meaningful clinical changes. Although a specific minimal clinically important difference (MCID) for the TIS in individuals with multiple sclerosis (MS) has not yet been firmly established, evidence from recent studies in neurological populations provides useful reference values. For example, MCID estimates for the TIS in stroke populations range from approximately 3 to 8,5 points, depending on the methodology applied (36; 37). These findings suggest that even relatively small changes in TIS scores may correspond to meaningful improvements in postural control. In individuals with MS, trunk control is closely associated with key functional domains such as balance, gait, and overall mobility, reinforcing its relevance as a determinant of physical function38. Consequently, the magnitude of change observed in this study may translate into improvements in postural stability and functional performance, even though these outcomes were not directly assessed. Furthermore, the concept of MCID has become increasingly important in rehabilitation research, as it allows the interpretation of whether statistically significant changes are also clinically meaningful from both clinician and patient perspectives39. From a clinical standpoint, these findings support the integration of targeted trunk-focused interventions into rehabilitation programs for individuals with MS. Improvements in trunk control may enhance proximal stability, facilitate more efficient limb movements, and contribute to better balance strategies, with potential implications for functional independence and fall risk reduction. Future research should aim to establish disease-specific MCID values for the TIS in the MS population to improve the clinical interpretability of intervention effects. Additionally, studies incorporating longer follow-up periods and functional outcome measures, such as gait performance, balance, and participation, are warranted to determine whether improvements in trunk control translate into meaningful real-world benefits. The inclusion of patient-reported outcome measures would further enhance the understanding of the clinical impact of these interventions. Limitations This study has several limitations. First, the non-randomized design may have introduced selection bias, despite the use of strict inclusion criteria. Future studies should aim to employ randomized controlled designs to enhance methodological robustness and generalizability. Second, the sample size calculation was based on an assumed large effect size (d = 1.0), which may be optimistic and may have limited the ability to detect moderate effects. Therefore, the results should be interpreted with caution. Third, the intervention duration was relatively short, which may have been insufficient to elicit changes in some outcomes, particularly structural or long-term adaptations. Future studies should consider longer interventions to better capture the full effects of respiratory muscle training. Finally, the sample was heterogeneous, particularly in terms of age range. Although no statistically significant age differences were found between groups, future research may benefit from stratifying participants by age to explore possible age-related variations in response to the intervention. Conclusions Training with respiratory muscle devices such as the Threshold IMT and Threshold PEP. (Respironics®, Philips) for home use may contribute to improvements in respiratory function, particularly in the FEV1/FVC ratio and peak expiratory pressure, while maintaining maximum inspiratory pressure, and may also be associated with enhanced trunk control. However, these findings should be interpreted with caution due to the non-randomized design, the small sample size, and the absence of significant changes in some outcomes such as fatigue and functional balance. Therefore, the results should be considered exploratory. Further randomized controlled trials with larger samples and longer follow-up periods are needed to confirm these findings and to better understand the long-term effects of RMT in individuals with MS.