Peak Quadriceps Muscle Torque and Electromyographic Output in Patients With Chronic Respiratory Disorders: Effects of Pulmonary Rehabilitation

Abstract

Objective

To assess the change in relation of the peak quadriceps electromyographic signal to the peak torque produced during a train of 5 isokinetic knee extensions (from 90 degrees below horizontal at a constant speed of 60 degrees/s) at baseline, and at 4 and 8 weeks of pulmonary rehabilitation.

Design

In this prospective observational study, isokinetic contractions were recorded during the extensions from the knee bent at 90 degrees to the horizontal plane against graded resistance. Peak quadriceps torque signal (Tq) and peak electromyographic signal (Eq) were recorded by dynamometry and surface electrodes placed at designated locations over the muscle group, respectively.

Setting

Physical therapy department in a tertiary care medical center.

Participants

Eighteen patients (9 restrictive lung disease, 6 chronic airflow limitation, 3 non-ILD restrictive; N=18) were compared with 11 healthy control subjects.

Interventions

Patients underwent an 8-week pulmonary rehabilitation program.

Main Outcome Measures

Comparisons of Tq, Eq, and Tq/Eq ratio among patients and controls were by analysis of variance. Associations between physiological variables were determined by multivariable Pearson's correlation.

Results

Compared with patients, controls exhibited a 22% higher baseline mean peak Eq (P<.05) and 76% higher mean peak Tq (P=.02) during knee extensions. Patients’ peak Eq/Tq was twice as high as in the controls (P=.02); at 4 weeks, Eq/Tq in patients decreased by 44% (P<.04) with no further decline at 8 weeks; changes in Eq/Tq of 5 of 6 patients paralleled changes in their respective St George's Respiratory Questionnaire scores. There was no change in Tq or Eq/Tq over time among the control cohort.

Conclusions

Eight weeks of pulmonary rehabilitation result in a decrease in Eq/Tq, indicating improvement in force generation of limb muscles, with the change occurring in the first 4 weeks.

Skeletal muscle dysfunction is an important systemic manifestation of chronic airflow limitation (CAL) and is a contributing factor in reduced exercise capacity, impaired quality of life (QOL), and higher health care utilization.¹ Current opinion suggests that the decrease in muscle strength is due to a reduction in cross sectional area and mass, while the loss of endurance is due to muscle fiber type changes2, 3, 4, 5, 6; both occur with chronic inflammation, muscle wasting and disuse, and oxidative stress.7, 8, 9, 10, 11 CAL patients develop a decrease in type I and increase in type II fibers compared with normal aging population.12, 13, 14 Muscle fiber type shift in CAL presents as 2 phenotypical patterns that exhibit different muscle histology (type I fiber proportion).¹³ In addition, isometric endurance of the quadriceps is reduced in patients with CAL and correlates with muscle oxidative activity.¹⁴ Quadriceps muscle dysfunction is associated with decreased survival,^14,15 poor functional status, and QOL.^6,15 Patients with idiopathic pulmonary fibrosis (an advanced form of interstitial lung disease) also exhibit sarcopenia with differences in indices of QOL, physical activities, psychological state, and depression compared with non-sarcopenic groups.¹⁶ More important, recovery of quadriceps muscle strength and endurance underlie much of the increase in exercise tolerance after multidisciplinary pulmonary rehabilitation (PR) for CAL.^15,17, 18, 19, 20 PR programs significantly benefit QOL and exercise tolerance in patients with chronic lung disease.20, 21, 22, 23 The programs are also important in the management of other conditions such as bronchiectasis, interstitial lung disease (ILD), and pulmonary hypertension and are recommended to improve fitness prior to lung transplantation.²⁴

Simultaneous recordings of surface electromyographic symbol (Eq) and torque of the quadriceps (Tq) during isometric and isokinetic contractions and their relation (Eq/Tq) have yielded important information with respect to muscle fatigue and recovery.25, 26, 27, 28, 29, 30 Eq/Tq increases as skeletal muscle fatigues under conditions of stress over time. This method has also been adapted for the evaluation of diaphragmatic fatigue in patients with chronic obstructive pulmonary disease (COPD) by measuring transdiaphragmatic pressure during inspiratory efforts.³¹ Relating changes in peak Tq to changes in peak Eq during isokinetic quadriceps contractions has not been investigated in patients with chronic respiratory illnesses undergoing PR. One would expect that PR results in a decrease in Eq/Tq as muscles require less neural stimulation to generate the same force (torque) as they did prior to PR.

The purpose of this study was to assess the relation of the peak quadriceps electromyographic (EMG) output to the torque produced during a train of 5 knee extensions over the course of an 8-week PR program in patients with chronic respiratory disorders, including restrictive diseases. Given that PR is known to improve QOL,^15,16 we also wished to assess if Eq/Tq correlated with quality-of-life as measured by the validated St. George's Respiratory Questionnaire (SGRQ),³² the total score of which has been used as an independent predictor of mortality also in patients with idiopathic pulmonary fibrosis.³³

Methods

Patients

This was a single center, prospective, observational study in which peak force generation of the quadriceps muscle group was simultaneously compared with its peak EMG output during maximal knee extensions. Measurements were assessed at intervals in clinically stable patients with chronic respiratory disorders undergoing PR. They were referred from the chest clinic if they had decreased functional capacity, activities of daily living, and spirometric evidence of respiratory impairment. The study was approved by the institutional review board (HS-19-00797) and all patients signed an informed consent prior to undergoing testing. The findings of this study were previously reported in part in an abstract elsewhere.³⁴ Since the publication of the abstract, additional patients have been studied with expansion of the physiological analysis in this paper. The study was conducted between January 2019 and January 2022. Patients unable to perform quadriceps muscle torque measurements due to musculoskeletal conditions such as arthritis and previous knee surgery were excluded from the study. All patients (testing negative for COVID 19) underwent pulmonary function testing according to American Thoracic Society/European Respiratory Society (ATS/ERS) guidelines and classified as having CAL or restrictive respiratory disease (with or without interstitial changes as seen on computed chest tomography) accordingly.³⁵

Once consented, all patients underwent a standard 8-week outpatient PR program conducted by the members of the Department of physical therapy.¹⁶ Patients are initially evaluated by a dedicated physical therapist and are requested to be seen ideally twice weekly thereafter to partake of the program that involves upper and lower body strength and endurance training. Many patients, however, are indigent, having limited financial resources and able to come at best once a week or every other week, mainly because of transportation issues, illness, or hospitalization. Return visits became more challenging after the onset of COVID19-imposed restrictions for face-to-face visits. Under these circumstances, patients were provided with instructions to continue the same exercises at home; they were subsequently contacted on a weekly basis by the physical therapist to inquire about their continued self-management. Those who chose to continue a home PR program were requested to come at 4 and 8 weeks after the initiation of PR for physiological evaluation. As such, some patients who were initiated in the study were unable to maintain their activities in the PT department or at home. Control subjects continued their regular daily activities in the physical therapy department; all underwent muscle testing at baseline and at 4 and 8 weeks. Individuals who were symptomatic or who tested positive for COVID19 were excluded from the study.

Equipment

Quadriceps force generation was recorded with a Biodex dynamometer (Biodex Medical Systems, 20 Ramsey Road, Shirley, New York) at the beginning, middle (at 4 weeks), and end of the 8-week PRP. The Biodex System 3 is an isokinetic dynamometer which provides constant velocity with accommodating resistance throughout a joint's range of motion. The resistance is produced via a hydraulic or electric servo-controlled mechanism at a constant velocity, in this case 60 degrees/second. The Biodex was periodically calibrated for angle and torque (Tq), using the procedure of Taylor et al.³⁶ The reliability and validity of the Biodex system has previously been tested by mechanical means³⁷ and during isokinetic testing of muscle strength in older men and women.³⁸ For the measurement of peak EMG output (Eq), surface electrodes were placed on the vastus lateralis muscle at the midpoint between the head of the greater trochanter and the lateral condyle of the femur. Electromyographic signals were recorded with a Vectra Genisys Therapy System (Chattanooga, DJO, LLC 1430 Decision St, Vista, CA). The same researcher (G.Y.) applied the dynamometer sensors and EMG electrodes in identical manner and position in every patient at every interval.

Testing procedure

All tests were performed on the right leg. Prior to testing, subjects performed a submaximal knee extension as warm-up to get used to the Biodex system. Isokinetic contractions were recorded by having the patient perform a train of 5 extensions from the knee bent at 90 degrees to 10 degrees below the horizontal plane (total angle 80 degrees) against graded resistance. The maneuvers were performed at a moderate angular velocity (60 degrees/s) as used in healthy older adults³⁸ and those with chronic illnesses.39, 40, 41 Higher angular velocities require rapid acceleration of the limb, are difficult to perform by older individuals and may lead to measurement errors.^42,43 Maximal isokinetic contractions about the knee were performed at the same time as on the dynamometer with recordings of knee joint movement, net moment, and EMG of quadriceps muscles. The means of 5 Eq and Tq signals during maximal contractions, respectively, were recorded. From these data, mean Eq/Tq was computed for each individual. The coefficient of variation for 5 peak contractions during knee extension was 7% for both EMG and torque signals. At the time of initial evaluation and completion of PR, patients were asked to complete a SGRQ.³³ Eleven healthy control subjects, consisting of staff from the rehabilitation center, underwent the same dynamometry and EMG measurements once for validation and to serve as a baseline comparison with patients with chronic respiratory disease; they did not undergo lung function testing or the SGRQ survey.

Statistical considerations

Descriptive data are shown as mean and standard deviation. Data analysis was performed for the cohort as a whole (n=18) and separately for patients with chronic airflow limitation (CAL, n=6) and interstitial lung disease (ILD, n=9); comparisons between baseline study group as a whole and the control group were by 2-tailed Student t test. Three patients who did not have CAL or ILD (by imaging) were not included in the subgroup analysis. Comparisons among cohorts who underwent physiological measurements at the 3 time periods were determined by analysis of variance (ANOVA) with adjustments for age, sex, and body mass index (BMI).⁴⁴ Associations between physiological variables were determined by multivariable Pearson's correlation, expressed as r². A P value of <0.05 was considered statistically significant for intergroup comparisons and for inter-variable associations.

Results

Comparison of baseline physiological measurements in original patients with those of control subjects

Nineteen patients initiated the PR program and underwent initial physiological evaluation of the quadriceps muscle group; 1 patient did not undergo respiratory function testing. All 6 patients with CAL were men while 8 of 9 patients with restrictive conditions were women. Mean age separately for women and men was 52 years. Six patients had pulmonary function data consistent with CAL; 6 had a restrictive pattern with imaging evidence of interstitial lung disease (ILD); 3 others had restrictive patterns without imaging evidence for ILD (fig 1). Nine of 11 control subjects were women. Table 1 lists baseline anthropometric and physiological variables for the 18 patients and 11 control subjects. Patients are subdivided into CAL and restrictive groups with differences among mean pulmonary functions as noted. Compared with patients with CAL, patients with restrictive lung disease exhibited lower forced vital capacity (FVC), forced expiratory volume in 1 second (FEV₁), and single breath carbon monoxide diffusion capacity (D_LCO) values. Patients with CAL exhibited plethysmographic lung volumes significantly higher than those with restrictive conditions, consistent with air trapping.

Fig 1.

Flow chart for patients screened, recruited and tested.

Table 1.

Characteristics and baseline physiologic data for 18 subjects with respiratory disease and 11 control subjects

All Respiratory Subjects (n=18)*		CAL (n=6)	Restrictives (n=9)	Controls (n=11)	P†
Age, years	55±13	63±8	49±14	49±14	0.05
M/F	10/9	6/0	1/8	2/9
BMI, kg/m2	28±6	29±7	30±3		NS
FVC, L	2±1	2.8±0.7	1.3±0.4		0.02
FVC, % pred.	56±21	68±18	48±18		0.025
FEV1, L	1.4±0.5	1.5±0.4	1.1±0.3		0.05
FEV1, % pred.	55±21	50±16	56±23		NS
FEV1/FVC, %	74±18	53±14	87±3		0.02
TLC, L	4.3±2	5.9±1.8	2.8±1.2		0.025
TLC, % pred.	83±27	100±27	61±17		0.01
RV, L	2±1.2	2.9±1.3	1.1±0.3		0.02
RV, %pred.	110±51	132±55	79±19		0.01
RV/TLC, %	47±12	49±9	44±6		NS
DLCO, mL/min/mmHg	11±6	13±4	7±4		0.01
DLCO, % pred.	50±21	57±10	38±17		0.01
Peak Eq, µv	264±73	235±106	275±54	322±129	0.05‡
Peak Tq, ft-lb	46±30	65±34	42±30	81±30	0.02
Peak Eq/Tq, µv/ft-lb	8.4±6.3	4.5±1.8	10.9±7.8	4.2±1.7	0.02

NOTE. Values represent mean ± SD.

Abbreviations: DLCO, single breath lung diffusion capacity; Eq, quadriceps EMG signal; NS, not significant; RV, residual volume; TLC, total lung capacity; Tq, quadriceps torque signal.

^⁎Three patients had non-interstitial combined obstructive and restrictive parenchymal disorders.

^†Difference between CAL and restrictives, ANOVA.

^‡Difference between controls and respiratory disease.

As can also be seen in table 1, the control group (n=11) exhibited a 22% higher baseline mean peak EMG output (P<.05) and 76% higher mean peak quadriceps muscle torque (P=.02) than the combined patient cohort during the train of isometric knee extensions. As a result, the study group as a whole had twice as high a mean peak Eq/Tq compared with the controls (P=.02). Among patients, mean peak Eq/Tq was significantly higher in the restrictive group than in the CAL patients and controls (P<.02). Baseline mean Eq/Tq in women was more than double than in men (P<.01).

Effects of pulmonary rehabilitation

Tables 2 and 3 show the same variables at the start, midpoint (4 weeks), and end (8 weeks) of the PR program for 8 patients who completed all 3 sets of measurements, and of the 11 control subjects, respectively (an additional patient completed the baseline and final testing but not the midterm and was not included in this analysis). All patients who underwent completed physiological testing continued home PR programs as instructed. As can be seen in table 2, among patients mean peak Tq had increased by 38% by midterm (although did not reach statistical significance) and remained unchanged by the end of the rehabilitation. By contrast, mean peak Eq/Tq had decreased by 44% midway through rehabilitation (P<.04) and remained unchanged after 8 weeks. In the control subjects, mean peak Eq, Tq, and Eq/Tq remained unchanged over the 8-week period (table 3). Patients who did not complete the remaining evaluations experienced acute illness, lack of transportation, or declined to continue muscle function evaluation.

Table 2.

Changes in peak Eq, Tq, and Eq/Tq from baseline to midterm (4 weeks) and final (8 weeks) measurements in 8 patients who completed all 3 sets of measurements

	Baseline	Midterm	Final	P*
Eq	281±55	311±115	337±110	NS
Tq	42±29	58±25	58±27	NS
Eq/Tq	11.1±8.1	6.5±3.3	6.7±2.8	0.04

NOTE. Values represent mean ± SD.

Abbreviations: NS, not significant.

^⁎ANOVA, multivariate.

Table 3.

Changes in Eq, Tq, and Eq/Tq from baseline to midterm and final measurements in 11 control subjects who underwent testing at 4 week intervals

	Baseline	Midterm	Final	P*
Eq	322±129	308±113	355±167	NS
Tq	81±30	81±32	82±26	NS
Eq/Tq	4.2±1.7	4.1±1.6	4.4±1.5	NS

NOTE. Values represent mean ± SD.

Abbreviations: NS, not significant.

^⁎ANOVA, multivariate.

Associations between Eq/Tq and respiratory function

In the entire cohort of 18 patients, the only statistically significant associations relevant to baseline peak Eq/Tq occurred with FVC (L) (r=-0.5, P=.03) and D_LCO (ml/min/mm Hg) (r=-0.59, P=.01). Patients in the CAL and ILD groups were too few for associations to achieve statistical significance.

Changes in SGRQ scores

Data for SGRQ at baseline and at completion of PR could be obtained for only 6 of the 8 patients who completed all 3 evaluations during the program (3 others completed only the baseline SGRQ assessment and were not included in this analysis). As noted in table 4, 4 patients showed improvement (decreases) of their SGRQ while 2 felt subjectively worse as determined by their SGRQ score. The mean total SGRQ score for the 6 patients at baseline and completion of the program was 66% and 59% predicted, respectively (NS, P=.28). All 4 patients whose Eq/Tq decreased after 8 weeks of PR exhibited improvements in their SGRQ scores. Of the remaining 2 patients, Eq/Tq of one increased by 3.1% as her SGRQ increased by 25%; the Eq/Tq of the sixth patient decreased by 37% while his SGRQ score increased by 2.4%. The change in Eq/Tq tended to be negatively associated with the baseline total SGRQ score but the number of patients was too few to reach statistical significance (fig 2).

Table 4.

Change in quality of life scores after completing PR

Patient	Starting SGRQ	Starting SGRQ %	Ending SGRQ	Ending SGRQ %	% Change
A	3147	78.9	2248.7	56.4	-28.6
D	2338.1	58.6	2919	73	24.6
E	2907.1	72.9	2082.8	52	-28.6
F	2516.9	63.1	2576.1	64.4	2.4
G	2230	55.9	2018.1	50.6	-9.5
H	2587	64.8	2280.1	57.1	-11.8

Fig 2.

Change in Eq/Tq after 8 weeks of pulmonary rehabilitation with respect to baseline total SGRQ scores (expressed as percent predicted). r²=0.383 (NS).

Discussion

The main findings in this study are (1) compared with healthy controls, patients with chronic respiratory illnesses exhibited higher pre-PR Eq/Tq ratios during a train of 5 maximal knee extensions against resistance, consistent with reduced quadriceps force generation; (2) at midpoint (4 weeks) the study cohort exhibited a significant decrease in mean Eq/Tq; the greatest relative improvement was noted at midpoint of rehabilitation which was sustained until the end of the program, with no further significant decrease in Eq/Tq at 8 weeks; (3) the Eq/Tq of 4 of 6 patients paralleled changes in their respective SGRQ scores. Meanwhile, there was no significant change in quadriceps muscle strength among the control cohort.

To our knowledge, this is the first study that has assessed the ratio of peak quadriceps muscle EMG signal to peak quadriceps muscle torque measured simultaneously during isokinetic knee extensions to monitor patients with chronic respiratory disorders undergoing PR.

Technical considerations

We considered a number of technical aspects and assumptions in preparing for and conducting the measurements described herein. Studies measuring EMG output and torque in healthy volunteers or rehabilitation patients to assess muscle fatigability during repetitive isometric contractions41, 42, 43, 44, 45, 46, 47, 48 have shown increases in EMG amplitude with respect to force generation during isometric skeletal muscle contractions as the muscle fatigues. Because the EMG reflects the electrical, not the mechanical aspects of a contraction, its interpretation requires caution.49, 50, 51 In isometric contractions, the relation between force and EMG amplitude is usually linear or close to linear in human knee extensors.^48,52,53 Others have reported a non-linear relation.54, 55, 56 While isokinetic muscle strength has not been studied extensively in patients with chronic respiratory disorders, it has been used across adult age ranges^38,57 and in non-respiratory diseases.^40,58 Isokinetic contractions pose additional complexities resulting from experimental conditions. Moderate knee extension velocities (60 degrees/s), as used in this study are recommended in order to minimize the effects of angular position, the angular displacement speed of the limb, intensity of muscle strength, and velocity of contraction which is related to the temporal variation of force.^23,50,59,60 In addition, the magnitude and timing of synergistic action of the 4 components of the quadriceps as well as co-contraction between antagonistic muscle groups can change the contribution of muscle strength on the net force generated at the joint.^26,61

Influence of patient characteristics

All patients in the CAL group were men, while 8 of 9 in the restrictive group were women. Our findings may have been influenced by sex, an important consideration as patients in the latter group exhibited a significantly higher peak Eq/Tq, indicating a higher predisposition for developing muscle fatigue. At least 1 study has shown that women exhibit a greater degree of fatigue during the first 30 repetitive isokinetic thigh contractions while performing an endurance test.⁶ In studies of muscle fiber type composition of the vastus lateralis, Staron et al⁶² showed that men have a higher proportion of fast oxidative type IIa fibers, whereas women have a higher proportion of slow oxidative type I fibers, which, in part, would explain the higher Eq/Tq in women; an increase in Eq/Tq enhances the likelihood of fatigue developing during repetitive contractions. Aging also results in a decrease in type II fiber size and oxidative capacity in healthy men and can be prevented by endurance training.^63,64 The capillary supply per unit type II fiber area is not affected by age but can be enhanced by training. Anthropometric indices correlate with skeletal muscle mass.

Relation of skeletal muscle force generation to respiratory function

We found statistically significant associations of baseline Eq/Tq with FVC and D_LCO (both as % predicted) in the group as a whole (n=18), similar to other studies evaluating muscle force generation and fatigue.^28,30,59, 60, 61 Decreased skeletal muscle mass is associated with diminished lung function and poorer prognosis in COPD.^15,17 Han et al⁶⁵ found that the prevalence of skeletal muscle atrophy in Chinese patients (mean age 67 years, BMI 22) with CAL was 7.3%, with age, sex, BMI, and severity of CAL factors influencing fat-free mass index and appendicular skeletal-muscle mass index. A reduction in FVC is associated with diminished respiratory muscle strength while a reduced D_LCO occurs with ventilation-perfusion mismatching and hypoxemia, contributing to respiratory muscle weakness. In the elderly, peripheral muscle strength is positively associated with maximal inspiratory and expiratory muscle strength.^66,67 In a study of 240,562 Korean subjects without known lung disease, Park et al⁶⁸ found that decreased skeletal muscle mass was independently associated with reduced FEV₁, FVC, and PEF after adjustment for age, sex, various metabolic factors, and health behavior-related factors. Similarly, Singer et al⁶⁹ showed that decreased (isometric) leg and respiratory muscle strength are independently associated with poor exercise capacity and lower extremity functioning across a spectrum of COPD severity.

Effect of pulmonary rehabilitation on skeletal muscle force generation

Eight of 9 patients who completed a final evaluation exhibited reductions in Eq/Tq after 8 weeks of PR. The ninth patient exhibited a 3.1% increase in Eq/Tq, a clinically negligible finding. These findings are consistent with previous reports of the effects of skeletal muscle fatigue on the relation of the EMG signal to torque generated during muscle contractions^{28,30,51,53,60} and its reversal with endurance training.^63,64

Relation of skeletal muscle strength to the SGRQ

Complete SGRQ data were collected in 6 subjects by the end of PR. As noted in table 3, the percent change in SGRQ score was variable with 2 patients reporting worsening symptoms at the completion of PRP. Interestingly, patients with a higher SGRQ score at the start of PR tended to exhibit an increase in Eq/Tq throughout rehabilitation as noted in figure 2. We surmise that those patients with worse QOL experience the greatest benefit from PR with respect to skeletal muscle strength and enhanced neural drive⁷⁰ but this will need to be confirmed in larger studies. That skeletal muscle function improved with PR likely as well in patients with ILD is consistent with findings in clinical trials showing improved functional capacity and QOL in this population.71, 72, 73

Clinical implications of findings

Our findings show that an 8-week PR program results in a reduction of Eq/Tq during a train of repetitive isokinetic contractions of the quadriceps. This finding is likely associated with decrease in skeletal muscle fatigue during exercise, in turn related to increase in type I muscle fibers and oxidative metabolism over time. Lower limb exercises result in increased endurance and strength during activities of daily living. Patients with advanced emphysema and pulmonary fibrosis benefit from such a program which reduces the risk of respiratory exacerbations and hospitalization.^1,11,20 This information lends additional quantitative physiological evidence to the benefits of such a program in clinical practice. With improvement in overall strength and endurance, the patient's psychological well-being (as assessed by a tool such as the SGRQ) will also improve, encouraging them to increase daily activities and functional capacity.

Opportunities for research

Larger studies incorporating simultaneous recordings of EMG and torque generation during isokinetic quadriceps contractions in patients with CAL and restrictive disorders undergoing PR should provide additional insight into the mechanisms of response. There are likely to be differences found in changes of Eq and Tq, depending on the underlying disorder, age, and sex. Exercise routines can then be tailored to the individual according to alterations in muscle physiology. Specific research relating Eq/Tq to outcomes of PR might include: (1) Relating changes in Eq/Tq with changes in tools for assessing fatigue such as the Manchester Chronic Obstructive Pulmonary Disease Scale⁷⁴ and the COPD Assessment Test (CAT).⁷⁵ Determine the same relation in patients with non-CAL disorders; (2) Comparison of changes in the ratio in patients undergoing home PR vs in-hospital or clinic PR programs. This would be important as studies show that outcomes with home-based exercise training are less effective in patients experiencing severe dyspnea,⁷⁶ possibly because of poor physical fitness and muscle deconditioning. These patients likely exhibit higher Eq/Tq than those participating in a hospital outpatient setting. Based on our findings, one would expect patients with lower Eq/Tq show a greater relative increase in the ratio provided they fully adhere to the PR program; (3) Compare change in Eq/Tq between patients completing the PR program with those who fail to do so (either partially complete or not adhere to the exercise program at all). This finding would show that the longer patients continued the exercise program, the lower the Eq/Tq would diminish; (4) Assess Eq/Tq after a comprehensive PR program for patients with ongoing symptoms of post-COVID-19 infection, and relate it changes to 6 minute walk test, the CAT, SGRQ, and the Post-COVID Functional Scale (PCFS)⁷⁷; (5) Measurements of hand grip strength before and after pulmonary rehabilitation should be correlated with changes in Eq/Tq as it may be an easier way to assess recovery of muscle mass and function than undergoing dynamometry and EMG analysis.

Limitations

There were limitations to this study. We used peak values of the EMG and torque signals rather than the areas under each curve, or the integrated signals, as used in many studies^23,24,26,41,44, 45, 46, 47, 48, 49, 50, 51, 52; we were unable to obtain integrated EMG signals using the Vectra Genisys system; instead corresponding peak EMG and torque values were assessed. Nevertheless, the same procedure of computing the means of the 5 peak EMG and torque signals were used consistently at each time period. An important question that needs exploration is assessment of the relation of Eq/Tq to the SGRQ among patients with different diseases, for example, CAL vs ILD. The numbers in each category were too small to achieve significant associations with pulmonary function, so larger studies are needed to demonstrate a relation as studies suggest that sarcopenia may be more severe and extensive in CAL than in ILD and have a greater effect on outcome.13, 14, 15, 16 This study was also affected by the COVID-19 pandemic during which our center experienced closures of most outpatient services including physical rehabilitation. For pragmatic reasons, most PR programs hold sessions only 2 or 3 times weekly,²⁰ but most programs instruct patients to exercise at home in between sessions, as we do. The frequency of exercise sessions to obtain an optimal training effect in ILD has not been established, but the regimen of twice-weekly supervised sessions with additional unsupervised sessions at home has been advocated by at some groups.¹⁶ Most patients did not complete the questionnaire at 1 or multiple points in the study; the few who did at baseline and final visit (n=6) exhibited, for the most part, parallel trends in the SGRQ score and Eq/Tq, suggesting an association between muscle endurance and QoL which needs to be confirmed with larger studies.

Conclusions

Eight weeks of PR results in a decrease in Eq/Tq, indicating improvement in contractile properties of limb muscles. Virtually all improvement occurs at midpoint (4 weeks) of rehabilitation which is sustained until the end of the program with no further significant decrease in Eq/Tq at 8 weeks. This approach provides a quantitative means of evaluating recovery of limb strength with respect to volitional neural stimulation in chronic respiratory patients undergoing PR.