Respiratory muscle weakness and respiratory muscle training in severely disabled multiple sclerosis patients☆1☆2☆3☆4☆5☆6☆7☆8☆9

Article Outline

• Abstract
• Materials and methods
  • Part 1
  • Part 2
    • Pulmonary function test
    • Anthropometric data
    • Respiratory muscle strength
    • Cough efficacy and functional status
    • Statistics
• Results
  • Part 1
    • Pulmonary function, cough efficacy, and functional status
    • Respiratory muscle strength
    • Correlation analysis
  • Part 2
• Discussion
  • Criticism of the methods
  • Respiratory muscle dysfunction, impaired FVC, and cough efficacy
  • Respiratory muscle training
• Acknowledgements
• References
• Copyright

Abstract 

Gosselink R, Kovacs L, Ketelaer P, Carton H, Decramer M. Respiratory muscle weakness and respiratory muscle training in severely disabled multiple sclerosis patients. Arch Phys Med Rehabil 2000;81:747-51. Objective: To evaluate the contribution of respiratory muscle weakness (part 1) and respiratory muscle training (part 2) to pulmonary function, cough efficacy, and functional status in patients with advanced multiple sclerosis (MS). Design: Survey (part 1) and randomized controlled trial (part 2). Setting: Rehabilitation center for MS. Patients: Twenty-eight bedridden or wheelchair-bound MS patients (part 1); 18 patients were randomly assigned to a training group (n = 9) or a control group (n = 9) (part 2). Intervention: The training group (part 2) performed three series of 15 contractions against an expiratory resistance (60% maximum expiratory pressure [PEmax]) two times a day, whereas the control group performed breathing exercises to enhance maximal inspirations. Main Outcome Measures: Forced vital capacity (FVC), inspiratory and expiratory muscle strength (PImax and PEmax), neck flexion force (NFF), cough efficacy by means of the Pulmonary Index (PI), and functional status by means of the Extended Disability Status Scale (EDSS). Results: Part 1 revealed a significantly reduced FVC (43% ± 26% predicted), PEmax (18% ± 8% predicted), and PImax (27% ± 11% predicted), whereas NFF was only mildly reduced (93% ± 26% predicted). The PI (median score, 10) and EDSS (median score, 8.5) were severely reduced. PEmax was significantly correlated to FVC, EDSS, and PI (r = .77, −.79, and −.47, respectively). In stepwise multiple regression analysis, PEmax was the only factor contributing to the explained variance in FVC (R² = .60), whereas body weight (R² = .41) was the only factor for the PI. In part 2, changes in PImax and PEmax tended to be higher in the training group (p = .06 and p = .07, respectively). The PI was significantly improved after 3 months of training compared with the control group (p < .05). After 6 months, the PI remained significantly better in the training group. Conclusions: Expiratory muscle strength was significantly reduced and related to FVC, cough efficacy, and functional status. Expiratory muscle training tended to enhance inspiratory and expiratory muscle strength. In addition, subjectively and objectively rated cough efficacy improved significantly and lasted for 3 months after training cessation. © 2000 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation

Keywords:  Multiple sclerosis, Respiratory function tests, Rehabilitation, Respiratory muscles, Training

MULTIPLE SCLEROSIS (MS) is a primary disorder of the central nervous system that may affect motor pathways and cause muscle weakness. Recent reports address the importance of respiratory muscle weakness in these patients.¹, ², ³ Respiratory complications are common in the terminal stages of MS and contribute to mortality in these patients.⁴, ⁵, ⁶ When respiratory motor pathways are involved, respiratory muscle weakness frequently occurs in MS and may impair the performance of coughing. Aspiration, pneumonia, or even acute ventilatory failure may ensue.

Because pulmonary complications contribute importantly to morbidity and mortality in MS,⁷ it appears worthwhile to study respiratory muscle function and its relationship with forced vital capacity (FVC), cough efficacy, and functional status. In addition, strategies to improve respiratory muscle function are likely to be important in reducing deterioration of pulmonary function and perhaps in improving cough efficacy and survival.

The first part of the study was thus designed to examine the contribution of respiratory muscle weakness to this impaired pulmonary function and health status. It was concluded that expiratory muscle strength was more affected than inspiratory muscle strength. In addition, expiratory muscle weakness was significantly related to FVC, cough efficacy, and functional status. Because of the observed expiratory muscle weakness and its relationship to FVC and cough efficacy, the second part of the study was performed to investigate the effects of expiratory muscle training on respiratory muscle function, FVC, and cough efficacy.

Back to Article Outline

Materials and methods 

Part 1 

Twenty-eight bedridden or wheelchair-bound MS patients (mean age, 58 ± 14yrs; mean duration of the disease, 27 ± 13yrs) were studied to examine the contribution of respiratory muscle weakness to their impaired health status. FVC, inspiratory and expiratory muscle strength (PImax and PEmax), Pulmonary Index (PI), and Extended Disability Status Scale (EDSS) were measured. All patients were in a clinically stable condition for at least 4 weeks (no recent infection or exacerbation). All tests were performed in the early afternoon. Data are summarized in table 1.

Table 1: Patient characteristics in part 1

Abbreviations: BMI, body mass index; EDSS, Extended Disability Status Scale; FVC, forced vital capacity; PImax, maximum inspiratory pressure; PEmax, maximum expiratory pressure.

Part 2 

Twenty-one bedridden or wheelchair-bound MS patients were included in the second part of the study to evaluate the effects of expiratory muscle training on FVC, PImax and PEmax, and the PI (table 2). Power analysis based on the results of Smeltzer and coworkers⁸ showed that at least 5 patients were needed in each group. Because our patient group was more severely disabled, we included 21 patients. Three patients dropped out from the study because of lack of cooperation. Patients were randomized by random numbers in sets of 10 patients. The patients in the training group (n = 9) performed three series of 15 expiratory contractions (60% PEmax) with the Threshold^a adapted for expiratory loading two times a day. The control group (n = 9) was instructed to perform breathing exercises to enhance maximal inspirations. These breathing exercises were routinely part of the physiotherapy treatment in the MS center. No specific feedback was given to the control group, and the breathing exercises were not supervised. For these reasons we considered this a control treatment. The measurements were repeated after 3 months of training and 3 months after the training period.

Table 2: Patient characteristics of the control group and the training group in part 2

Abbreviations: BMI, body mass index; EDSS, Extended Disability Status Scale; FVC, forced vital capacity; PImax, maximum inspiratory pressure; PEmax, maximum expiratory pressure.

The study was approved by the Medical Ethical Board of the National Multiple Sclerosis Centre Melsbroek.

Pulmonary function test 

All patients underwent spirometry in the sitting position with determination of the FVC (Ohmeda 5420 Volume monitor^b). Spirometry was repeated until no further improvement of recordings was obtained. Sufficient rest periods were left between tests until the patients' respiratory rates were back to baseline and their subjective feeling was that they could continue. The highest value obtained was related to the normal values of Quanjer and colleagues.⁹

Anthropometric data 

Body weight and height were taken from recent patient records. Body mass index was calculated as the ratio of body weight and height squared (kg/m²).

Respiratory muscle strength 

All patients underwent determination of PImax and PEmax. These pressures were measured with a portable electronic manometer (Mouth pressure device^c).¹⁰ A mouthpiece or, in patients with more severe weakness, a face mask was held firmly over the mouth and nose of the patient by the investigator to prevent air leaking. PImax was measured from residual volume (RV), whereas PEmax was measured near total lung capacity (TLC). Tests were repeated until no further improvement was obtained and at least three attempts differed less than 5%. Sufficient rest periods were provided between tests until patients' respiratory rates were back to baseline and their subjective feeling was that they could continue. The highest values were taken for analysis.

Isometric hand-held dynamometer (MicroFET^d) measurements of neck flexion force (NFF) were performed according to van der Ploeg and associates.¹¹ The measurements were related to their reference values.¹¹

Cough efficacy and functional status 

The PI¹² was used to assess objectively and subjectively rated cough efficacy. Scores range from 4 to 11; a score of 4 indicates normal cough efficacy, whereas a score of 11 is the most abnormal cough efficacy. The impact of the neurologic disorder on functional status was assessed with the EDSS.¹³ Scores range from 0 to 10; a score of 0 indicates no disability; 5, ambulatory without assistance; 8, wheelchair bound or bedridden; and 10, dead.

Statistics 

Statistical analysis was performed on the data obtained from the 28 patients who participated in part 1 of the study and from the 18 patients who participated in part 2. Pearson's correlation coefficients between variables were calculated. For the variables significantly related to FVC, PI, and functional status, a stepwise multiple regression analysis was performed. Variables used in the model included age, gender, body weight, height, duration of illness, respiratory muscle function, and pulmonary function. Only variables significantly contributing in single correlation were retained.

Comparison between the control group and training group was done with unpaired t tests. Within-group comparison was performed with paired t test analysis. All statistical analysis was performed using SAS statistical package.^e Limits of significance were set as p < .05.

Back to Article Outline

Results 

Part 1 

Pulmonary function, cough efficacy, and functional status 

A significantly reduced FVC (1.4 ± 0.9L, 43 ± 26% predicted) was observed. The median PI score was 10 (ranging from 6 to 11) indicating severely impaired cough efficacy. The high median score on the EDSS (8.5, ranging from 6.5 to 9.5) (table 1) confirmed the markedly reduced mobility of the patients, nearly all being wheelchair bound or bedridden.

Respiratory muscle strength 

The PEmax (29 ± 15cmH₂O, 18 ± 8% predicted) was significantly more reduced than the PImax (25 ± 14cmH₂O, 27 ± 11% predicted) (p < .01). Mean isometric neck flexion strength was only mildly reduced (93 ± 36N, 93 ± 26% predicted). Two patients failed to perform a proper inspiratory and expiratory maneuver, whereas in seven patients one of the maneuvers was impossible to perform. Three patients failed inspiratory pressure measurement, and four failed expiratory pressure measurement. The FVC in these seven patients was significantly lower compared with those who were able to perform the respiratory pressure maneuvers (19 ± 8 vs 50 ± 26% predicted, p < .01).

Correlation analysis 

The EDSS was significantly correlated to FVC (% predicted) (r = −.87, p < .001), PEmax (% predicted) (r = −.79, p < .001) (fig 1), and the PI (r = .45, p < .05).

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 1. 

  Relation between expiratory muscle strength (PEmax), measured as percentage predicted (% pred) and Extended Disability Status Score (EDSS).

The PI was significantly correlated with body weight (r = −.64, p < .001), PEmax (% predicted) (r = −.47, p < .02), and FVC (% predicted) (r = −.36, p = .05). The FVC (% predicted) was significantly correlated with PEmax (% predicted) (r = .77, p < .001) (fig 2), NFF (% predicted) (r = .56, p < .005), and PImax (% predicted) (r = .48, p < .01).

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 2. 

  Relation between expiratory muscle strength (PEmax), measured as percentage predicted (% pred), and forced vital capacity (FVC).

In stepwise multiple regression analysis, PEmax was the only factor contributing to explained variance in FVC (R² = .60), whereas body weight (R² = .41) was the only factor for the PI.

Part 2 

Patient characteristics between training and control groups did not differ initially (table 2). The PImax was significantly improved after 3 months of training compared with baseline (Δ−9 ± 9cmH₂O, p < .05), but not compared with the control group (p = .06, table 3).

Table 3: Changes at 3 and 6 months after training

Values are expressed as mean ± SD.

Abbreviations: VC, vital capacity; PImax, maximum inspiratory pressure; PEmax, maximum expiratory pressure.

Changes in PEmax were higher in the training group, but the improvement failed to reach statistical significance compared with baseline (Δ8 ± 14cmH₂O, p = .08) as well as compared with the changes in the control group (p = .07). In addition, the PI improved significantly compared with baseline and with the control group (p < .05). No significant changes were observed in pulmonary function (0.2 ± 0.5L, p = .41). Subgroup analysis showed that the improvement in PImax in the training group was significantly associated with the initial EDSS (r = −.76, p < .02; fig 3).

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 3. 

  Relation between changes in maximal inspiratory pressure (PImax), measured as percentage predicted (% pred), after training and the initial Expanded Disability Status Score (EDSS) in the training group.

After 6 months, the PI remained significantly better in the training group (p < .05), whereas the improvements of PImax tended to be higher, but failed to reach statistical significance (p = .16).

Back to Article Outline

Discussion 

It was concluded from the first part of the study that in wheelchair-bound and bedridden MS patients, expiratory muscle strength was more affected than inspiratory muscle strength. Expiratory muscle strength was significantly related to FVC, cough efficacy, and functional status.

From the second part of the study, it was concluded that expiratory muscle training tended to enhance both inspiratory and expiratory muscle strength and significantly improved the objectively and subjectively rated cough efficacy, which lasted for 3 months after training cessation.

Criticism of the methods 

The assessment of FVC and respiratory muscle strength is effort and motivation dependent. Leakage during the maneuvers was prevented by the use of a face mask, which was held firmly by the investigator. Maneuvers were repeated until no further improvement was observed. Although careful instruction was given and patients were well motivated, full cooperation with tests was in some patients very difficult. In particular, timing of the onset of the maneuvers was sometimes difficult. Some patients failed to follow specific instructions to perform maximal inspiratory or expiratory mouth pressures at a specific lung volume (RV or TLC). In seven patients, all with a FVC below 600mL, data on PImax or PEmax were missing because of an inability of the patient to perform the test. Values below 7cmH₂O were not accepted by the mouth pressure measurement device. Therefore, the observed respiratory muscle weakness might have been underestimated.

The effects of inspiratory muscle training might have been studied in more detail when TLC and RV were measured. However, a body plethysmograph was not available in the center where the patients were hospitalized. Obviously, bedridden patients would be unable to perform these tests.

Respiratory muscle dysfunction, impaired FVC, and cough efficacy 

Expiratory muscles were more affected than inspiratory muscles, as observed previously.¹, ³, ¹², ¹⁴, ¹⁵ Obviously, the demyelinating process affected abdominal muscles more than inspiratory muscles. Neck flexors, acting as accessory inspiratory muscles, had preserved strength but could not compensate for the obvious weakness of primary inspiratory muscles. The reduced inspiratory muscle strength is partly explained by the increased RV. Consequently, the PImax maneuver was performed with the inspiratory muscles operating at a shorter length, thereby reducing the force-generating capacity. In addition, deconditioning might also contribute to respiratory muscle weakness.¹⁶, ¹⁷, ¹⁸ This study's findings were in agreement with previous findings of the positive relationship between expiratory muscle strength, and EDSS and pulmonary function.¹² Pulmonary function was also related to EDSS.¹, ³

Expiratory muscle weakness is associated with difficulty in coughing in neuromuscular disorders,¹², ¹⁹, ²⁰ shown by the significant correlation between PEmax and PI. The relationship was weaker than that observed by Smeltzer and associates,¹² but their study also included patients with mild disease. Obviously, our patients had more severe disability as illustrated by the high EDSS (median 8.5 vs 7.0 in the study by Smeltzer). The association between PI and body weight might be explained by disease severity, which might be reflected in body weight due to loss of muscle mass, malnutrition, and cachexia.

Respiratory muscle training 

The observed tendencies of enhanced respiratory muscle function in the training group might be influenced by the somewhat better baseline characteristics of the training group. Although no statistically significant differences were observed between both groups, the training group was somewhat younger, had a better preserved FVC and respiratory muscle function, and had 7 years' less history of MS. These factors could have contributed to somewhat more reserve to obtain training effects. We were, however, unable to identify clear differences in baseline patient characteristics between responders (those with improved respiratory muscle function) and nonresponders. Only a reduced training response of PImax was associated with a higher EDSS (fig 2).

Our results differ in some respect with those of Smeltzer et al.⁸ The relative improvement in PEmax (35%) was similar, but our results did not reach statistical significance (p = .08). In contrast to the tendency of an improvement of PImax in this study, Smeltzer observed only a marginal, nonsignificant increase of PImax (6%). These differences might be explained by differences in patient population, training protocol, or both. The patients in Smeltzer's study had a shorter duration of disease (14 vs 27yrs), probably a lower EDSS score (range, 6.5 to 9.5; median score not reported), and substantially higher respiratory pressures (PImax 47% predicted, PEmax 37% predicted). Expiratory muscle loading protocol in our study was set at 60% of PEmax and a low repetition number was chosen, as strength rather than endurance capacity was aimed to improve. Effective coughing needs explosive expiratory muscle contraction. Smeltzer⁸ aimed at a similar high-intensity protocol, but the obtained intensity was not reported. It is not likely that the training intensity in our study was too low, because subjectively our patients experienced a high training intensity. The lack of significance might be caused by the very low expiratory muscle strength (18% predicted), probably indicating severe expiratory muscle impairment and reduced trainability. We speculate that the poor training response of the expiratory muscles might be related to the severity of the demyelination process. Demyelination of the central nervous system might affect the innervation of expiratory muscles, thus reducing the force-generating capacity of the expiratory muscles.

An alternative explanation for the improved inspiratory muscle strength is a reduction of RV. Expiratory muscle training might have reduced expiratory lung volume, thus allowing the inspiratory muscles to operate at a more advantageous part of the length-tension relationship. Unfortunately data on RV were not available, but the absence of a relationship between changes in FVC and changes in respiratory muscle function makes this assumption unlikely.

The tendency of an improvement of PImax as a result of expiratory muscle training was surprising. We speculate that during expiratory loading, in the absence of sufficient expiratory muscle strength, patients increased their inspiratory lung volume and hence, elastic recoil pressure to overcome the expiratory pressure. This stimulus was probably high enough to cause a training response of the inspiratory muscles. Alternatively, a decrease of RV caused by expiratory muscle training might put the inspiratory muscles on a more advantageous part of their length-tension relationship. Because measurements of TLC and RV were not performed, only speculations can be made.

Interestingly, in addition to improved respiratory muscle function, FVC did not change significantly. This conclusion follows the findings of the uncontrolled study by Olgiati and coworkers.²¹ As expected in patients of Olgiati's study who had only moderate muscle weakness, vital capacity remains unchanged.²² However, our study included patients with severe respiratory muscle weakness, and accordingly a restrictive pulmonary dysfunction. Improvement of PImax and PEmax was expected to be associated with a proportional lung volume increase in patients with severe muscle weakness. The lack of improvement might perhaps also be explained by the absence of specificity of respiratory muscle training. We have no data on the actual lung volumes during expiratory muscle training, but we might speculate that these were not adequate to improve lung function. Inspiratory muscles need to be trained near TLC to improve inspiratory capacity, whereas expiratory muscles need training near RV to enhance expiratory capacity. Inspiratory muscles were probably loaded near TLC, but training was not targeted at these muscles. Expiratory muscles were not loaded near RV, but probably near TLC.

The PI decreased significantly indicating improved cough efficiency. This was not studied in previous studies, but seems important because it may contribute to a reduced incidence of pulmonary complications. Importantly, this improvement was maintained during the follow-up period. Our follow-up was too short to obtain reliable results on the incidence of pulmonary complications. In addition, we were unable to reconstruct reliable data from the period before the study period from patient records. Future studies in patients with severe MS should focus on inspiratory muscle training. Enhancing inspiratory reserve volume increases elastic recoil pressure and alveolar pressure, which might contribute to improved cough efficacy and prevent pulmonary complications. Improvement of inspiratory muscle function might be obtained through inspiratory resistive training or through the use of incentive spirometry. Incentive spirometry may load the inspiratory muscles near TLC, but has the disadvantage that the actual training intensity is not known. Inspiratory muscle training allows accurate control of the actual training intensity.

In conclusion, expiratory muscle strength was significantly reduced and related to FVC, cough efficacy, and functional status. Expiratory muscle training tended to enhance inspiratory and expiratory muscle function. In addition, subjectively and objectively rated cough efficacy improved significantly and lasted for 3 months after training cessation.

Back to Article Outline

Acknowledgements 

The authors gratefully thank Mrs. R. Schepers for her expert help with statistical analysis, and Mrs. V. Debusschere, G. DeDekker, and N. Demolon for their assistance in data collection and supervision of the training.

Back to Article Outline

References 

1. Smeltzer SC, Utell MJ, Rudick RA, Herndon RM. Pulmonary function and dysfunction in multiple sclerosis. Arch Neurol. 1988;45:1245–1249
   • View In Article
   • MEDLINE
2. Foglio K, Clini E, Facchetti D, Vitacca M, Marangoni S, Bonomelli M, et al.  Respiratory muscle function and exercise capacity in multiple sclerosis. Eur Respir J. 1994;7:23–28
   • View In Article
   • MEDLINE
   • CrossRef
3. Buyse B, Demedts M, Meekers J, Vandegaer L, Rochette F, Kerkhofs L. Respiratory dysfunction in multiple sclerosis: a prospective analysis of 60 patients. Eur Respir J. 1997;10:139–145
   • View In Article
   • MEDLINE
   • CrossRef
4. Sadovnick AD, Eisen K, Ebers GC, Paty DW. Cause of death in patients attending multiple sclerosis clinics. Neurology. 1991;41:1193–1196
   • View In Article
   • MEDLINE
   • CrossRef
5. Phadke JG. Survival pattern and cause of death in patients with multiple sclerosis: results from an epidemiological survey in north east Scotland. J Neurol Neurosurg Psychiatry. 1987;50:523–531
   • View In Article
   • MEDLINE
   • CrossRef
6. Midgard R, Riise T, Kvale G, Nyland H. Disability and mortality in multiple sclerosis in western Norway. Acta Neurol Scand. 1996;93:307–314
   • View In Article
   • MEDLINE
   • CrossRef
7. Gosselink R, Kovacs L, Decramer M. Respiratory muscle involvement in multiple sclerosis. Eur Respir J. 1999;13:449–454
   • View In Article
   • MEDLINE
   • CrossRef
8. Smeltzer SC, Lavietes MH, Cook SD. Expiratory training in multiple sclerosis. Arch Phys Med Rehabil. 1996;77:909–912
   • View In Article
   • Abstract
   • Full-Text PDF (497 KB)
   • CrossRef
9. Quanjer PH, Tammeling GJ, Pedersen OF, Peslin R, Yernault JC. Lung volumes and forced expiratory flows. Eur Respir J. 1993;6(Suppl 16):5–40
   • View In Article
   • MEDLINE
10. Hamnegard CH, Wragg S, Kyroussis D, Aquilina R, Moxham J, Green M. Portable measurement of maximum mouth pressures. Eur Respir J. 1994;7:398–401
    • View In Article
    • MEDLINE
    • CrossRef
11. van der Ploeg RJO, Fidler V, Oosterhuis JHGH. Hand-held myometry: reference values. J Neurol Neurosurg Psychiatry. 1991;54:244–247
    • View In Article
    • MEDLINE
    • CrossRef
12. Smeltzer SC, Skurnick JH, Troiano R. Respiratory function in multiple sclerosis: utility of clinical assessment of respiratory muscle function. Chest. 1992;101:479–484
    • View In Article
    • MEDLINE
    • CrossRef
13. Kurtzke JF. Rating neurological impairment in multiple sclerosis: an Expanded Disability Status Scale (EDSS). Neurology. 1983;33:1444–1452
    • View In Article
    • MEDLINE
    • CrossRef
14. Tantucci C, Massucci M, Piperno R, Betti L, Grassi V, Sorbini CA. Control of breathing and respiratory muscle strength in patients with multiple sclerosis. Chest. 1994;105:1163–1170
    • View In Article
    • MEDLINE
    • CrossRef
15. Olgiati R, Jacquet J, di Prampero PE. Energy cost of walking and exertional dyspnea in multiple sclerosis. Am Rev Respir Dis. 1986;134:1005–1010
    • View In Article
    • MEDLINE
16. Kent-Braun JA, Sharma KR, Weiner MW, Miller RG. Effects of exercise on muscle activation and metabolism in multiple sclerosis. Muscle Nerve. 1994;17:1162–1169
    • View In Article
    • MEDLINE
    • CrossRef
17. Kent-Braun JA, Sharma KR, Miller RG, Weiner MW. Postexercise phosphocreatine resynthesis is slowed in multiple sclerosis. Muscle Nerve. 1994;17:835–841
    • View In Article
    • MEDLINE
    • CrossRef
18. Kent-Braun JA, Ng GY, Castro M, Weiner MW, Gelinas D, Dudley GA, et al.  Strength, skeletal muscle composition, and enzyme activity in multiple sclerosis. J Appl Physiol. 1997;83:1998–2004
    • View In Article
    • MEDLINE
19. Polkey MI, Lyall RA, Green M, Nigel P, Moxham J. Expiratory muscle function in amyotrophic lateral sclerosis. Am J Respir Crit Care Med. 1998;158:734–741
    • View In Article
    • MEDLINE
20. Szeinberg A, Tabachnik E, Rashed N, McLaughlin FJ, England S, Bryan CA, et al.  Cough capacity in patients with muscular dystrophy. Chest. 1988;94:1232–1235
    • View In Article
    • MEDLINE
    • CrossRef
21. Olgiati R, Girr A, Hugi L, Haegi V. Respiratory muscle training in multiple sclerosis: a pilot study. Schweiz Arch Neurol Psychiatr. 1988;140:46–50
    • View In Article
22. De Troyer A, Borenstein S, Cordier R. Analysis of lung volume restriction in patients with respiratory muscle weakness. Thorax. 1980;35:603–610
    • View In Article
    • MEDLINE
    • CrossRef