Feasibility of Low-Load Resistance Training Using Blood Flow Restriction for People With Advanced Multiple Sclerosis: A Prospective Cohort Study

Abstract

Objective

The objective of this study was to determine the feasibility of low-load resistance training with blood flow restriction (BFR) for people with advanced disability due to multiple sclerosis (MS).

Methods

In this prospective cohort study, 14 participants with MS (Expanded Disability Status Scale [EDSS] score = 6.0 to 7.0; mean age = 55.4 [SD = 6.2] years; 71% women) were asked to perform 3 lower extremity resistance exercises (leg press, calf press, and hip abduction) bilaterally twice weekly for 8 weeks using BFR. Feasibility criteria were as follows: enrollment of 20 participants, ≥80% retention and adherence, ≥90% satisfaction, and no serious adverse events related to the intervention. Other outcomes included knee extensor, ankle plantar flexor, and hip abductor muscle strength, 30-Second Sit-to-Stand Test, Berg Balance Scale, Timed 25-Foot Walk Test, 12-Item MS Walking Scale, Modified Fatigue Impact Scale, Patient-Specific Functional Scale, and daily step count.

Results

Sixteen participants consented, and 14 completed the intervention, with 93% adherence overall. All participants were satisfied with the intervention. A minor hip muscle strain was the only intervention-related adverse event. There were muscle strength improvements on the more-involved (16%–28%) and less-involved (12%–19%) sides. There were also changes in the 30-Second Sit-to-Stand Test (1.9 repetitions; 95% CI = 1.0 to 2.8), Berg Balance Scale (5.3 points; 95% CI = 3.2 to 7.4), Timed 25-Foot Walk Test (−3.3 seconds; 95% CI = −7.9 to 1.3), Modified Fatigue Impact Scale (−8.8 points; 95% CI = −16.5 to −1.1), 12-Item MS Walking Scale (−3.6 points; 95% CI = −11.5 to 4.4), Patient-Specific Functional Scale (2.9 points; 95% CI = 1.9 to 3.8), and daily step count (333 steps; 95% CI = −191 to 857).

Conclusion

Low-load resistance training using BFR in people with MS and EDSS scores of 6.0 to 7.0 appears feasible, and subsequent investigation into its efficacy is warranted.

Impact

Although efficacy data are needed, combining BFR with low-load resistance training may be a viable alternative for people who have MS and who do not tolerate conventional moderate- to high-intensity training because of more severe symptoms, such as fatigue and weakness.

Lay Summary

Low-load strength training with BFR was feasible in people who have advanced disability due to MS. Using BFR may provide an alternative for people with MS who do not tolerate higher intensity training due to more severe symptoms, such as fatigue and weakness.

Introduction

Multiple sclerosis (MS) is an autoimmune condition that causes demyelination and axonal death in the central nervous system, and can result in a variety of symptoms depending on lesion location.¹ Muscle weakness is one of the most common impairments for people with MS and is key contributor to impaired mobility.^2,3 As disability advances so does the prevalence of muscle weakness such that 80% of those who require a unilateral assistive device (Expanded Disability Status Scale [EDSS] scores of 6.0) and essentially all those using a wheelchair (EDSS scores of 7.0–8.0) have muscle weakness.²

Resistance training is at the core of best-practice exercise guidelines for people with MS.⁴ There is strong evidence that moderate- to high-intensity resistance training improves muscle strength in people with MS and mild to moderate disability (EDSS scores of <6.0).^5,6 However, resistance training has not been thoroughly investigated in those with more advanced disease (EDSS scores of ≥6.0), who may require unique approaches to resistance training as they cannot always tolerate higher intensities because of more severe weakness and other MS-related symptoms, such as fatigue.^7,8 Traditional modalities like weight training machines may create additional barriers because of physical access limitations and inability to perform the exercises even with the lowest resistance settings.

Blood flow restriction (BFR) training has the potential to address the unique needs of people with advanced disability due to MS. BFR uses a pressure cuff to partially occlude blood flow of an exercising limb, causing muscle hypoxia and metabolic stress, which triggers anaerobic metabolism, protein synthesis, and satellite cell proliferation, even with low-intensity resistance training (20%–30% of the 1-repetition max [1RM]).^9–11 Low-intensity BFR training has been shown to be as effective as higher intensity training without BFR (70%–80% of the 1RM) at increasing strength in people with musculoskeletal conditions and in older adults.^9–11 Importantly, for people with MS and EDSS scores of ≥6.0, resistance training at low intensities with BFR is less demanding on muscles, connective tissues, and joints, which can result in less delayed onset muscle soreness and fatigue, and therefore it improves tolerance.¹² In addition, because of lower loads, resistance training with BFR can be more easily dosed and progressed for people with severe weakness using simple and accessible modalities, such as exercise bands, weight cuffs, or portable resistance machines.¹²

BFR is an increasingly common clinical intervention among rehabilitation professionals,¹³ and as many as 40% of clinicians using BFR have applied it in people with neurologic conditions.¹⁴ However, there is very little literature on BFR training in people with neurologic conditions, including MS. One study has demonstrated tolerance in a single session of low-load resistance training with BFR in 15 people with MS (EDSS score of 1.87 [SD = 1.51]).¹⁵ Two case reports have described improvements in strength and function using low-load resistance training with BFR in individuals with MS and EDSS scores of 3.0 and 5.5.^16,17 Finally, a pilot study in 22 participants with MS (EDSS scores of 5.5–6.5) applied BFR during a walking task and reported improved walking speed, but did not measure muscle strength changes.¹⁸ These early trial and case report data are promising, but to inform clinical use and efficacy trials, more study is needed to determine the feasibility and initial efficacy of resistance training with BFR in people who have advanced disability due to MS.

The objective of this study was to determine the feasibility of low-load resistance training with BFR for people with MS and advanced disability based on enrollment, retention, adherence, satisfaction, and safety. In addition, initial efficacy and responsiveness for muscle strength, functional mobility, patient-reported outcomes, and daily step count were measured before and after intervention to inform future study design.

Methods

This prospective cohort study (NCT04633759) was approved by the local institutional review board and all participants provided informed consent. Participants were recruited primarily from the Rocky Mountain MS Center at the University of Colorado Anschutz Medical Campus. Between March 2021 and August 2022, 26 potential participants were consecutively screened for eligibility (Fig. 1). Individuals who had MS and were 18 to 70 years old were included if they had an EDSS score of 6.0 (unilateral assistance required to walk at least 100 m), 6.5 (bilateral assistance required to walk at least 20 m), or 7.0 (essentially restricted to a wheelchair, but can self-propel and transfer alone). Individuals were excluded if they had an EDSS score of ≥7.5 (essentially restricted to a wheelchair) or ≤ 5.5 (able to walk >100 m without walking aid or rest). Individuals were also excluded if they had a history of deep venous thrombosis, pulmonary embolism, vascular disease, thrombophilia, or other clotting disorders, had comorbidities that would interfere with their ability to exercise safely, were undergoing supervised resistance training, had used BFR in the 3 months prior to enrollment, or had MS-related exacerbation or medication changes in the month prior to enrollment.

Figure 1.

Prospective participant and study participant flow diagram. AD = assistive device.

Intervention

Intervention is described using the Template for Intervention Description and Replication and Consensus on Exercise Reporting Template checklists (Suppl. Tab. S1).^19,20 All participants were asked to perform 3 lower extremity resistance exercises (leg press, calf press, and hip abduction (Fig. 2) bilaterally twice weekly for 8 weeks while using BFR (Delfi PTSII; Delfi Medical Innovations, Vancouver, British Columbia, Canada). The muscle groups targeted were selected as they have all been found to be important contributors of functional mobility in people with MS.^3,21,22 Intervention was delivered face-to-face in a physical therapist gym used for clinical research and supervised by a licensed physical therapist who had undergone training specific to the Delfi PTSII BFR device.

Figure 2.

Standard exercise positions and set-up with Shuttle Mini-Press for (a) knee extension, (b) ankle plantarflexion, and (c) hip abduction.

The pressure cuff was placed on the most distal portion of the thigh, and individualized limb occlusion pressure (LOP) was determined in supine every session for each participant. During each exercise, the cuff was inflated between 60% and 80% maximal LOP, which is considered the therapeutic range of BFR as there is diminished venous outflow to induce a metabolic effect while still allowing arterial flow.¹² Exercises followed recommended dosing for BFR training: 1 set of 30 repetitions followed by 3 sets of 15 at 20% to 30% of the 1RM and 60% to 80% of the maximal LOP.¹² The cuff stayed inflated for all 4 sets, including 30-second rests between sets, but was deflated between the different exercisesEach exercise took 5 to 8 minutes to complete and was followed by a 2- to 3-minute rest before beginning the next exercise, resulting in sessions lasting approximately 60 minutes. Dosing intensity was determined by a perceived exertion scale (where 0 = extremely easy and 10 = extremely hard), and 20% to 30% of the 1RM was estimated by a participant rating of 2 or 3 out of 10 (easy to somewhat easy) during 10 repetitions without any occlusion.²³ The starting LOP ranged from 60% to 80% of LOP, but the goal was to increase LOP to 80% during the first 2 weeks.

Resistance was provided by the Shuttle Mini-Press (Shuttle Systems, Bellingham, WA, USA) (Fig. 2), a portable resistance trainer that allows for precise dosing (0–100 lb) and can be used from a standard chair, a low mat table, or even a wheelchair. Leg and calf presses were performed in a seated position from a standard height chair, and hip abduction was performed in the supine position on a standard low mat table (Fig. 2). A custom-built lift was added to the Shuttle Mini-Press to facilitate the range of motion needed for the calf press (Fig. 2b). If participants were unable to perform an exercise in the standard positions or with the minimal resistance of the Shuttle Mini-Press, exercises could be performed in alternate positions or with manually applied resistance. Exercises were performed 1 limb at a time in a standardized order: the leg press was first, followed by the calf press and hip abduction. Warm-up activities varied based on the participants’ wide range of functional abilities and activity tolerance; some participants walked for several minutes prior to starting exercises, whereas others performed 1 to 2 minutes of unresisted leg-press motion on the resistance trainer. Standard calf and thigh stretches were provided as a cool-down. Dosing and intensity were reassessed every 2 weeks and progressed as tolerated. Fidelity was assessed for each participant at the first intervention visit, at visit 5, and on an ongoing basis via chart review by a second physical therapist experienced in delivering BFR.

Outcomes

Participant age, sex, race, ethnicity, MS type, and EDSS level were collected to characterize the sample. Feasibility was defined a priori by enrolling 20 participants, retaining at least 80% of participants, adherence to at least 80% of intervention sessions, at least 90% of participants reporting being at least “satisfied” with the intervention on a 7-point Likert scale (0 = completely unsatisfied; 7 = completely satisfied), and no serious adverse events associated with the intervention.^24,25 Feasibility criteria, including sample size, were chosen to inform process and resource needs for a future efficacy trial, and to determine if changes to intervention protocol would be needed to maximize participant tolerance and satisfaction.^24–26 Resting blood pressure and heart rate, and fatigue and pain level measured on a visual analog scale (0 = no fatigue/pain; 10 = maximal fatigue/pain) were also assessed before and immediately after each exercise session. LOP was also recorded for each exercise at each session.

Muscle strength was measured by handheld dynamometry (Lafayette Manual Muscle Tester; Lafayette Instrument Co, Lafayette, IN, USA) and recorded in kilograms. In people with MS and EDSS scores of <6.0, muscle strength measured by handheld dynamometry is reliable, and minimal detectable change (MDC) values for knee extension and hip abduction have been reported to be 20% to 27% and 28% to 30%, respectively.²⁷ Knee extension was measured in a seated position, and hip abduction and ankle plantarflexion were measured in the supine position. A “make” protocol was used, meaning the examiner matched the force of the participant isometrically.²⁸ The more-involved limb was determined at baseline assessment and defined as the limb with worse muscle strength.

Functional mobility was measured by the 30-Second Sit-to-Stand Test (30STS), Berg Balance Scale (BBS), and Timed 25-Foot Walk Test (T25FW). The 30STS counts the number of times a patient can transfer from sitting to standing in 30 seconds and is a reliable and valid assessment in people with MS, and the MDC has been reported to be 1.13 repetitions.²⁹ The BBS assesses task-specific balance and activities of daily living function and is a reliable and valid outcome in people with MS, and the MDC has been reported to be 3 points.^30,31 The T25FW is the standard measure of gait speed in MS and correlates with disability and function, and meaningful change is commonly cited as an improvement of at least 20%.^32,33 For both 30STS and T25FW, upper extremity and/or assistance device use was recorded, and the same level of assistance was used at the follow-up visit.

Patient-reported outcomes were also measured before and after the intervention. The 12-Item MS Walking Scale (MSWS-12) was used to measure the impact of MS on an individual’s walking ability, with higher scores indicating greater impact.³⁴ The Modified Fatigue Impact Scale (MFIS) was used to measure fatigue perception, with higher scores indicating worse fatigue.³⁵ Meaningful change was defined as an 8-point decrease on the MSWS-12^34,36 and the MFIS.³⁷ For the Patient-Specific Functional Scale (PSFS), participants were asked to identify 3 activities related to walking, mobility, or moving around that they had difficulty performing because of their MS and rate them on a scale of 0 (unable to perform activity) to 10 (able to do activity without any difficulty).^38,39 An increase of at least 2.5 points on the PSFS corresponds with clinically meaningful change in people with MS.⁴⁰ Finally, step count was measured for each participant using a thigh-mounted activPAL (activPAL3; PAL Technologies Ltd, Glasgow, Scotland, UK) worn continuously for 10 days at each data collection time point.

Data Analysis

Study feasibility was determined by comparison to the predefined criteria in areas of recruitment, retention, adherence, satisfaction, and safety.²⁵ Changes in muscle strength, functional mobility, and patient-reported outcomes were analyzed using paired t tests and reported with 95% CIs of the mean difference. Results were compared to meaningful change or MDC values when available. All data analyses were performed with SPSS Statistics 28 (IBM, Armonk, NY, USA).

Role of Funding Source

Funders played no role in the design, conduct, or reporting of this study.

Results

Sixteen people consented to participate. One participant was excluded after a previously undisclosed neurological condition was discovered in the chart, and a second withdrew before starting the intervention because of transportation issues. In total, 14 participants began the intervention (EDSS scores of 6.0–7.0; 55.4 [SD = 6.2] years old; 71% women) (Tab. 1), and all 14 completed postintervention assessments. Although the study did not meet recruitment goals, all other feasibility criteria were achieved. Retention was 93%, and adherence was 93% for the 14 participants who started the intervention. All participants were satisfied with the intervention (6 rated it 6/7 [very satisfied], and 5 rated it 7/7 [completely satisfied]).

Table 1.

Per Participant Baseline Characteristics, Adherence, Satisfaction, and LOP, Ordered by EDSS Scores^a

Participant	EDSS Score	MS Type	Duration of MS (y)	Ethnicity/ Race	Age (y)	Sex	BMI (kg/m 2 )	No. (%) of Visits Completed	Satisfaction	Average LOP of Less-Involved Limb (%)	Average LOP of More-Involved Limb (%)	Achieved 80% LOP in All Exercises
1	6.0	SPMS	22	Not Hispanic/White	65	F	15.8	16 (100)	Satisfied	67	71	No
2	6.0	RRMS	19	Not Hispanic/White	58	F	21.8	16 (100)	Completely satisfied	75	75	Visit 11
3	6.0	SPMS	3	Not Hispanic/White	63	M	25.3	16 (100)	Very satisfied	78	78	Visit 4
4	6.0	SPMS	35	Not Hispanic/White	63	F	24.0	12 (75)	Very satisfied	76	76	Visit 6
5	6.0	RRMS	4	Not Hispanic/White	53	F	36.0	15 (94)	Completely satisfied	79	79	Visit 4
6	6.5	RRMS	28	Not Hispanic/White	46	F	23.9	10 (63)	Completely satisfied	80	80	Visit 1
7	6.5	PPMS	4	Not Hispanic/White	53	M	31.2	16 (100)	Very satisfied	80	80	Visit 1
8	6.5	PPMS	20	Not Hispanic/White	56	M	22.5	16 (100)	Completely satisfied	80	80	Visit 1
9	6.5	RRMS	33	Not Hispanic/White	59	M	28.3	15 (94)	Satisfied	73	73	Visit 11
10	7.0	RRMS	21	Not Hispanic/White	44	F	20.9	16 (100)	Very satisfied	80	80	Visit 2
11	7.0	SPMS	19	Not Hispanic/White	49	F	20.0	16 (100)	Completely satisfied	80	79	Visit 2
12b	7.0	RRMS	20	Not Hispanic/White	56	F	25.0	16 (100)	Very satisfied	61	61	No
13	7.0	SPMS	30	Not Hispanic/White	55	F	24.6	12 (75)	Very satisfied	77	77	Visit 6
14	7.0	PPMS	9	Not Hispanic/White	55	F	19.9	16 (100)	Satisfied	79	79	Visit 6
Total	6.5 (SD = 0.4)	3 PPMS, 5 SPMS, 6 RRMS	19.1 (SD = 10.7)	100% not Hispanic/White	55.4 (SD = 6.2)	10 F (71%)	24.2 (SD = 5.1)	14.8 (93)	100% satisfaction	76 (SD = 5.7)	76 (SD = 5.2)	4.2 (SD = 3.7) visits

^a BMI = body mass index; EDSS = Expanded Disability Status Scale; F = female; LOP = limb occlusion pressure; M = male; MS = multiple sclerosis; PPMS = primary progressive MS; RRMS = relapsing–remitting MS; SPMS = secondary progressive MS.

^b Impairments limited the ability to perform 1 exercise with the more-involved limb.

All participants tolerated at least 60% LOP at the first visit, and all but 2 participants were eventually able to tolerate 80% LOP during the study (Tab. 1). Before and immediately after exercise sessions, there were negligible, nonsignificant changes in systolic blood pressure (125.0 [SD = 19.1]–125.1 [SD = 19.7] mm Hg), diastolic blood pressure (79.4 [SD = 14.8]–80.4 [SD = 15.1] mmHg), and heart rate (77.8 [SD = 13.2]–78.3 [SD = 12.3] mm Hg) (Suppl. Tab. S2). After exercise sessions, there was a mean decrease in pain of −0.3 of 10 points (95% CI = −0.4 to −0.2) and an increase in fatigue of +0.6 of 10 points (95% CI = 0.4 to 0.6). All participants completed all exercises during each session except 1 participant, who was unable to perform the ankle plantarflexion exercise on the more-involved side for the entire duration of the intervention because of severe weakness (0/5 on the manual muscle test). One participant could not tolerate lying supine for a period long enough to complete the hip abduction exercise, so the exercise was performed in the standing position.

In total, there were 13 documented adverse events (Suppl. Tab. S3). The most common adverse event was a fall (n = 7), but none occurred during or immediately before or after an intervention session, and none caused an injury requiring medical attention. One participant reported increased lower extremity edema the day following an intervention session, but the study neurologist examined the participant and determined that the swelling was mild and typical for the participant. One participant was diagnosed with COVID-19 and missed 3 sessions but reported an uncomplicated recovery and resumed the intervention. Two participants experienced muscle strains. One, a hip abductor strain, was likely related to the intervention, and exercise load had to be decreased on that side for the remainder of the intervention. Finally, 1 participant arrived to a session with a blood pressure of >180/100, which resulted in cancelation of the exercise session. Although the participant was asymptomatic, the participant’s caregiver brought her to a nearby emergency department at the physical therapist’s recommendation, and blood pressure medications were adjusted before the next session.

Muscle strength changes are displayed in Figure 3. Knee extension strength improved 2.4 kg (16%) on the more-involved side (95% CI = 0.4 to 4.5) and 2.6 kg (12%) on the less-involved side (95% CI = 0.8 to 4.4). Ankle plantarflexion strength improved 3.8 kg (28%) on the more-involved side (95% CI = 0.5 to 7.1) and 3.8 kg (18%) on the less-involved side (95% CI = 1.1 to 6.5). Hip abduction strength improved 1.2 kg (25%) on the more-involved side (95% CI = 0.0 to 2.3) and 1.3 kg (19%) on the less-involved side (95% CI = 0.7 to 2.2). On average, resistance and occlusion pressure also progressed in the overall sample for all exercises (Tab. 2). In addition, all participants increased resistance load and/or occlusion pressure individually (Suppl. Tabs. S4–S6).

Figure 3.

Muscle strength changes in knee extension (Ext), ankle plantarflexion (PF), and hip abduction (Abd) in more and less-involved limbs.

Table 2.

Average Resistance and Occlusion Pressure Progression From Baseline to Final Visit^a

	Mean (SD) at:
Exercise	First Visit		Eighth Visit		Final Visit
	Resistance, kg	% Maximal LOP	Resistance, kg	% Maximal LOP	Resistance, kg	% Maximal LOP
More-involved leg press	7.8 (4.4)	70.4 (8.4)	11.7 (5.5)	77.6 (4.2)	14.0 (6.6)	79.3 (2.7)
Less-involved leg press	13.8 (4.3)	69.7 (7.9)	17.5 (5.2)	77.8 (4.6)	20.6 (6.7)	78.6 (3.6)
More-involved calf press	8.8 (5.5)	72.0 (8.5)	15.8 (10.8)	78.5 (3.8)	22.0 (11.6)	80.0 (0.0)
Less-involved calf press	19.9 (9.8)	69.7 (7.9)	25.4 (9.4)	77.9 (4.3)	31.0 (9.6)	79.3 (2.7)
More-involved hip abductionb	2.2 (1.7)	70.4 (8.4)	3.0 (2.1)	77.6 (4.2)	4.5 (3.7)	80.0 (0.0)
Less-involved hip abductionb	1.3 (2.3)	69.7 (7.9)	2.6 (4.2)	77.9 (4.3)	4.4 (6.9)	79.2 (2.8)

^a LOP = limb occlusion pressure.

^b One participant performed hip abduction in a standing, closed-chain position because of the inability to tolerate the supine position and was able to use considerably more resistance than the rest of the participants, who performed hip abduction in a supine, open-chain position.

Mobility and patient-reported outcomes are reported in Table 3. The number of repetitions on the 30STS improved by 1.9 (95% CI = 1.0 to 2.8), BBS scores improved by 5.3 points (95% CI = 3.2 to 7.4), and T25FW decreased by−3.3 seconds (95% CI = −7.9 to 1.3). Daily step count improved by 333 steps (95% CI = −191 to 857). The MFIS improved by −8.8 points (95% CI = −16.5 to −1.1), the MSWS-12 improved by −3.6 points (95% CI = −11.5 to 4.4), and the PSFS improved by 2.9 points (95% CI = 1.9 to 3.8).

Table 3.

Changes in Functional Performance and Patient-Reported and Physical Activity Outcomes^a

Outcome	Mean (SD)		Mean Difference (95% CI)
	Before Intervention	After Intervention
Functional mobility
30STS, reps	8.9 (3.3)	10.8 (3.3)	1.9 (1.0 to 2.8)
BBS	34.9 (13.9)	40.1 (12.5)	5.3 (3.2 to 7.4)
T25FW, s	22.7 (27.4)	19.3 (20.3)	−3.3 (−7.9 to 1.3)
Patient reported
MFIS	36.6 (15.9)	27.8 (18.4)	−8.8 (−16.5 to −1.1)
MSWS-12	78.9 (21.0)	75.3 (22.4)	−3.6 (−11.5 to 4.4)
PSFS	3.9 (1.1)	6.7 (1.8)	2.9 (1.9 to 3.8)
Physical activity
Daily step count	2478 (2115)	2811 (2484)	333 (−191 to 857)

^a 30STS = 30-Second Sit-to-Stand Test; BBS = Berg Balance Scale; MFIS = Modified Fatigue Impact Scale; MSWS-12 = 12-Item MS Walking Scale; PSFS = Patient-Specific Functional Scale; reps = repetitions; T25FW = Timed 25-Foot Walk Test.

Discussion

This study demonstrated the feasibility of low-load resistance training using BFR in people who have advanced disability due to MS. In addition, there were improvements in muscle strength, functional mobility, self-reported fatigue, and patient-identified goals. People with MS and more severe disease may have difficulty tolerating conventional moderate- to high-intensity resistance training and/or accessing traditional resistance training devices. BFR may be an important way to overcome tolerance and access barriers, and this study supports the need for a phase II efficacy trial of low-load resistance training with BFR in people with MS and advanced disability.

Two previous trials have examined conventional progressive resistance training in people with advanced disability due to MS.^7,41,42 In a randomized trial, Coote et al⁴¹ compared moderate-intensity resistance training with and without neuromuscular electrical stimulation. Although there were significant within-group strength improvements in both groups, 8 of 18 participants were lost to follow-up (44%) in the resistance training-only group: 1 because of nonadherence, 2 because of musculoskeletal injury, 2 because of fatigue, and 3 because of relapse. In a cohort study, Filippi et al⁴² enrolled 67 patients with MS (40 with EDSS scores of ≥5.0) in a moderate-intensity resistance training program using traditional weight training machines. However, very few of the participants with EDSS scores of ≥5.0 were able to complete knee flexion exercise or testing (5/40; 13%) or knee extension exercise or testing (17/40; 43%). Meanwhile, in our study, all participants who started the BFR intervention completed it, only 1 participant was unable to perform 1 exercise because of weakness, and all participants used progressively harder intensities during the intervention. Although more study is needed, our results support the idea that BFR training is feasible and well tolerated in people with MS and advanced disability, possibly because low-loads can accommodate more severe weakness and allow the use of more accessible training devices.

In our study, there was a slight decrease in pain after each session and only a small increase in self-reported fatigue. This is consistent with previous finding by Freitas et al,¹⁵ which demonstrated that in 15 people with MS and mild to moderate disability, 1 session of low-load resistance training with BFR resulted in significantly less perceived exertion and delayed onset muscle soreness than 1 session of moderate-intensity resistance training without BFR. A recent systematic review concluded that there may be a hypoalgesic effect with load–load resistance training using BFR in people rehabilitating from a musculoskeletal injury.⁴³ The mechanisms behind this phenomenon are not well-understood, but pain in MS is highly prevalent,⁴⁴ and future research may consider examining the effects of BFR on people with MS and pain that limits ability to exercise. Meanwhile, these collective data suggest that BFR could be an important alternative to conventional resistance training when pain and/or fatigue limit tolerance to higher loads.

There have been historical concerns about the safety of BFR, especially in people with comorbid health conditions and disabilities.^45,46 However, the rate of serious adverse events like venous thrombosis, pulmonary embolism, and rhabdomyolysis is low (0.008%–0.055%).⁴⁵ In our study, we excluded people with any history of clotting disorder, and no serious events occurred that were related to the intervention. There was 1 participant who experienced a hypertensive crisis; however, it occurred prior to the intervention session, and the participant was able to resume intervention after a medication adjustment. Although future studies in people with MS should continue to closely track adverse events and monitor vital signs, our results provide early evidence that BFR training can be safe even in people with severe MS.

We did not meet our recruitment target of 20 participants, and we were only able to identify 26 total potential participants over an 18-month period. Much of our recruitment occurred during a time when there were restrictions related to COVID-19, which may have resulted in potential participants being less willing to even consider volunteering for an in-person exercise study. Indeed, people with MS have had a higher risk of developing severe COVID-19 compared with the general population, and up to 25.40 increased odds of death if they have advanced disability.^47–49 In addition, during the first 2 years of the COVID-19 pandemic, people with MS had increased availability of telemedicine and telerehabilitation.⁵⁰ Although this was only 1 trial with a small sample, it is important to consider expanding recruitment timeline and resources for future in-person exercise studies.

This study was not designed to evaluate efficacy, but it is notable that there were positive changes in all muscle strength variables following intervention. Although values for important change in muscle strength for people with MS and EDSS scores of ≥6.0 are not established, the improvements we measured on the more-involved limb (16%–28%) do correspond to MDC values for handheld dynamometry in people with EDSS scores of <6.0.²⁷ In addition, all participants increased resistance load and/or occlusion pressure (Suppl. Tabs. S4–S6). Our study also extends findings from case reports documenting improved strength following low-load resistance training with BFR in people with MS and mild to moderate disability (EDSS scores of 3.0 and 5.5).^16,17 Together, these data are consistent with literature in people with musculoskeletal injury and in older adults, where efficacy of low-load resistance training with BFR is well studied^9–11 and support the further investigation of the efficacy of BFR to improve muscle strength in people with MS and advanced disability.

In our study we saw similar changes for strength in hip abduction (proximal to the cuff) as we did for knee extension and plantarflexion (distal to the cuff). BFR is thought to have the most effect in muscle groups distal to the cuff placement, and though there is some evidence that BFR training can improve muscles proximal to the cuff in healthy adults,^51,52 more research is needed to determine the relevance of cuff placement on strength outcomes in people with MS. However, because weakness in MS is not limited to the extremities, the ability to improve more proximal muscle groups, which are known to be important for postural control and mobility,^22,53 is intriguing and may be an interesting area of future study.

Muscle strength is a key contributor to mobility in people with MS,³ and our study measured improvements in the 30STS and BBS. Importantly, our measured changes also exceeded published values for the MDC on the 30STS (1.13 repetitions)²⁹ and minimal clinically important difference on the BBS (3 points)³¹ in people with MS. Meanwhile, improvements in walking speed were relatively small and did not approach the commonly accepted meaningful change value of 20% in people with MS,³³ which could be due to the wide range of walking ability and assistive device use in this study. In individuals with an EDSS score of 7.0, walking is not the main form of mobility, and therefore a walking assessment may be less responsive. Accordingly, the changes in the MSWS-12 were also small in this study and did not reach a meaningful improvement of 8 points.^34,36 Furthermore, 3 participants achieved the maximal score on the MSWS-12, indicating a floor effect for the outcome. In contrast, there was an improvement in MFIS scores in our study corresponding to published values for meaningful change of 4 to 8 points³⁷ and consistent with prior evidence demonstrating improved fatigue following resistance training in people with MS.^6,54 Finally, the PSFS improvement found in our study was in the range of meaningful change values of 2.5 points for the PSFS in people with MS.⁴⁰ Resistance training in people with advanced disability due to MS has the potential to improve mobility and other self-reported outcomes, further supporting future investigation into the efficacy of BFR in people with advanced MS.

In addition to mobility and self-reported outcomes, there was a small increase in number of daily steps in our study. This change, however, did not approach meaningful change values of at least 779 steps per day for people with MS.⁵⁵ People in our study took very few steps in general, which is similar to prior work showing that people who have MS and ambulate with walkers and canes take only 1972 (SD = 1260) and 3539 (SD = 2170) steps per day, respectively.⁵⁵ Our data reinforce the importance of improving physical activity in people with advanced disability due to MS, and more study is needed in this area. As it is likely necessary to include a behavioral change intervention alongside physical intervention when the goal is to improve physical activity,^56,57 future investigations of BFR training may consider incorporating behavioral change intervention or education to translate functional improvements to real-world participation changes.

This study was designed to measure feasibility, which is an important step when studying a new intervention in a new population. However, although determining feasibility of recruitment, retention, adherence, satisfaction, and tolerance is generally considered important prior to studying efficacy, there are not universal cutoffs on how best to evaluate these criteria.^24–26 Therefore, it could be that more stringent criteria would have provided a different conclusion. In addition to feasibility, this study did report improvements in muscle strength, mobility, and patient-reported outcomes. However, it was not designed or powered for efficacy, so we cannot conclude that the BFR was the cause of these changes and it could be that low-intensity resistance training by itself can improve strength and mobility in people with advanced disability due to MS. As such, we caution against widespread clinical adoption of BFR in people with MS without additional safety and efficacy evidence, especially in individuals who can tolerate conventional, moderate- to high-intensity resistance training, which is known to be effective.⁶

Our study had additional limitations. First, we estimated dosing intensity based on perceived exertion instead of determining 1RM via performance. We did not think 1RM would be feasible or safe for all participants, and importantly, perceived exertion scales have been shown to be reliable in people with MS.^58,59 A second limitation was that a standardized warm-up was not provided because of the different levels of fatigue and activity tolerance among participants, and this may have contributed to the 2 muscle strains experienced in the study. A third limitation was the use of handheld dynamometry. Although reliable, handheld dynamometry does have more variability than electromechanical dynamometry and results can be influenced by tester strength, although in this study most participants were weak enough that this was likely less of a concern.²⁸ A final limitation is that the entire study sample identified as White and not Hispanic and therefore may not be as generalizable to other races and/or ethnicities. Racial and ethnic diversity in MS research continues to be a major problem,⁶⁰ and future studies need to increase efforts and outreach to increase representation.

In conclusion, these results support the safety and feasibility of low-load resistance training using BFR in people with MS and EDSS scores of 6.0 to 7.0 and support further study into efficacy on outcomes of muscle strength, mobility, and self-reported fatigue. BFR may be a viable alternative for people who have MS and who do not tolerate conventional moderate- to high-intensity training because of more severe symptoms such as fatigue and weakness.

Author Contributions

Mark M. Mañago (Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Visualization, Writing—original draft, Writing—review & editing), Evan T. Cohen (Conceptualization, Methodology, Writing—original draft, Writing—review & editing), Enrique Alvarez (Conceptualization, Funding acquisition, Investigation, Methodology, Resources, Supervision, Writing—original draft, Writing—review & editing), Emily R. Hager (Data curation, Funding acquisition, Project administration, Writing—original draft, Writing—review & editing), Johnny G. Owens (Conceptualization, Funding acquisition, Investigation, Methodology, Resources, Writing—original draft, Writing—review & editing), and Michael Bade (Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Resources, Supervision, Validation, Visualization, Writing—original draft, Writing—review & editing)

Funding

This study was funded by the Consortium of Multiple Sclerosis Centers Pilot Award Program; National Center Medical Rehabilitation Research, National Institute of Child Health and Human Development, and National Institute Neurological Disorders and Stroke, National Institutes of Health (K12 HD055931); and Colorado Clinical & Translational Science Institute (NIH/NCATS UL1-TR001082, TL1-TR002535). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health, the US Department of Veterans Affairs, or the United States Government.

Data Availability

The authors confirm that the data supporting the findings of this study are available within the article [and/or] its supplementary material.

Ethics Approval

This study was approved by the Colorado Multiple Institutional Review Board (NCT04633759).

Disclosures and Presentations

The authors completed the ICMJE Form for Disclosure of Potential Conflicts of Interest and reported no conflicts of interest. Johnny G. Owens is a paid consultant for Delfi Medical Innovations, INC and the Major Trauma Research Consortium. Delfi Medical Innovations also provided the BFR devices for use in the trial. E. Alvarez has received compensation for activities such as advisory boards, lectures, and consultancy with the following companies and organizations: Actelion/Janssen, Alexion, Bayer, Biogen, Celgene/BMS, EMD Serono/Merck, Genentech/Roche, Genzyme, Novartis, and TG Therapeutics and research support from: Biogen, Genentech/Roche, Novartis, TG Therapeutics, Patient-Centered Outcomes Research Initiative, National Multiple Sclerosis Society, National Institutes of Health, and Rocky Mountain MS Center. An abstract reporting on some of these results was accepted as a platform and was presented at the 2023 American Physical Therapy Association Combined Sections Meeting; February 23–25; San Diego, California. Some of these results were presented in an educational session at the 2023 Consortium of Multiple Sclerosis Centers Annual Meeting; May 31 to June 3; Aurora, Colorado.