Efficacy of suspension exercise training in spinal cord injury: a systematic review and meta-analysis

Abstract

Background

Spinal cord injury (SCI) refers to damage sustained by the spinal cord due to direct or indirect external factors, resulting in a variety of motor, sensory, and sphincter dysfunctions. Suspension exercise training (SET), which utilizes neuromuscular activation techniques, has demonstrated lasting benefits for musculoskeletal disorders. The objective of this systematic review and meta-analysis was to rigorously evaluate the safety and effectiveness of SET as an intervention for SCI.

Methods

The authors conducted a comprehensive search across eight databases to identify all relevant randomized controlled trials (RCTs) on October 20, 2024. Data were extracted, the literature was reviewed, and the methodological quality of the included trials was evaluated. This meta-analysis was performed using Stata version 14.0.

Results

Thirteen studies involving a total of 883 patients were included. Outcomes: Lower extremity motor function score of the lower extremity motor function score of the American Spinal Injury Association Motor Score (ASIA-LEMS) [weighted mean difference (WMD) = 3.653, 95% confidence interval (CI) (2.351:4.956), P < 0.05], the gait parameters [step length standardized mean difference (SMD) = 0.655,95% CI(0.462:0.849), P < 0.05; step speed SMD = 1.057,95% CI(0.533:1.582), P < 0.05; step frequency SMD = 1.000,95% CI(0.512:1.489), P < 0.05], the Berg Balance Scale score(BBS) [WMD = 6.154,95% CI(5.019:7.289), P < 0.05]. The Modified Barthel Index score (MBI) [WMD = 12.475, 95% CI (5.855:19.094), P < 0.05], and the Modified Ashworth Score(MAS) [WMD=0.756, 95% CI (0.879:0.632), P < 0.05]. Significant improvements were observed in all of the aforementioned outcome indicators, suggesting that SET is likely to be effective in restoring function in patients with SCI.

Conclusion

This study has demonstrated that SET is likely to improve motor and balance functions, as well as activities of daily living, in patients, while also contributing to a reduction in muscle tone. However, this meta-analysis is limited by the quality of the studies included; therefore, further high-quality research is essential to validate these findings.

Systematic review registration

PROSPERO; identifier: CRD42024606161.

Clinical trial number

Not applicable.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13018-025-06044-z.

Background

SCI represents a significant neurological impairment that leads to motor, sensory, and autonomic dysfunction [1]. This condition has profound implications for both the physical and the psychological well-being of affected individuals. The global incidence of SCI ranges from 10.4 to 83 cases per million individuals annually, contributing to a substantial socioeconomic burden. Prompt intervention during the early postinjury phase is crucial for facilitating long-term functional recovery [2, 3].

SET is an innovative rehabilitation approach that focuses on enabling the body to engage in movement while in an unbalanced state. This method enhances core muscle strength and patient stability, which, in turn, aids in improving balance and motor function [4]. SET consists of both a diagnostic system and a therapeutic system. As an effective neuromuscular feedback reconstruction method, SET emphasizes the concept of active training. The core of the technique relies on neuromuscular activation methods to stimulate ‘dormant’ or ‘deactivated’ muscles, re-establishing their normal functioning patterns and neural control mechanisms [5]. One study demonstrated that the CHORDATA^® method, which involves suspension and pendulous exercises, is beneficial for improving muscle activation, flexor and extensor torques, and muscle structure, as well as for regaining functionality in patients with traumatic spinal cord injury (TSCI) [6]. Recent studies have indicated that SET can lead to enhancements in muscle mass, strength, and functional performance among older adults [7]. Compared with other treatment modalities, SET is emerging as a promising noninvasive therapy for the treatment of skeletal and muscular disorders, as it can more effectively motivate patients and improve their motor function. However, limited evidence exists regarding the application of SET for the functional improvement of patients with SCI. Consequently, the objective of this study was to assess the efficacy and safety of SET in the context of SCI treatment through a meta-analysis, thereby providing a foundation for clinical rehabilitation practices in this population.

Methods

This systematic review and meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) guidelines. The study protocol has been registered with PROSPERO. Registration number: CRD42024606161.

Literature search

PubMed, Web of Science, Embase, The Cochrane Library, CBM, CNKI, Wanfang, and VIP were utilized as data sources to search for RCTs of SET for the treatment of SCI, covering the period from the inception of the database until October 20, 2024. The following medical subject headings and keywords were employed in the search strategy: “spinal cord injuries,” “spinal cord injury,” “traumatic myelopathies,” “spinal cord trauma,” “suspension exercise training,” “suspension training,” “suspension exercise,” “sling exercise training,” and “sling exercise therapy.” The detailed search strategy is presented in Supplemental Table 1.

Inclusion and exclusion criteria

Trials were included in the study if the following criteria were met: (a) the studies were RCTs; (b) the included patients met the diagnostic criteria of the American Spinal Injury Association (ASIA) disability classification [8] and were diagnosed with SCI through magnetic resonance imaging (MRI) or computed tomography (CT) examination; (c) interventions: the control group received conventional rehabilitation or medication, while the experimental group was given SET in addition to the control group’s treatment; (d) the outcomes included at least one of the following outcome measures: the lower extremity motor function score of the ASIA-LEMS, gait parameters (step length, step speed, and step frequency), BBS, MBI, and MAS.

Trials were excluded if they met any of the following criteria: (a) repeated published experiments; (b) non-RCTs; (c) animal studies; (d) review articles; (e) literature with unclear outcome measures, incomplete data on outcome metrics, or unavailable and insufficient data for meta-analysis; (f) literature for which the original text could not be accessed.

Two researchers participated in the literature screening process, conducted their evaluations independently and utilized EndNote for literature management.

Data extraction and management

Data extraction was performed independently by two authors using Excel software. The extracted data included the study identification number, sample size, details of the intervention, duration of the intervention, outcomes, and patient characteristics, such as age, percentage of male participants, and duration of the disease.

Risk of bias assessment

The risk of bias in each included study was assessed independently by two researchers using the ROB 2 tool [9]. The RoB 2 tool is structured into five bias domains: bias arising from the randomization process, bias due to deviations from intended interventions, bias due to missing outcome data, bias in measurement of the outcome, and bias in selection of the reported result. Studies were evaluated as low, with some concerns and a high risk of bias. Any disagreements were resolved through discussion.

Quality of evidence

The degree of certainty regarding the outcome was assessed via the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) method [10]. Evidence from RCTs initially represents high-quality evidence; however, it can be downgraded on the basis of five factors: risk of bias, inconsistency, indirectness, imprecision, and publication bias. For the sake of transparency and simplicity, the GRADE system categorizes the quality of evidence into four levels: high, moderate, low, and very low [11]. The level of evidence for each meta-analysis was assessed independently by two reviewers, who then conferred with a third reviewer to resolve any uncertainties. Supplementary Table 2 provides information on GRADE scores.

Data analysis

This meta-analysis was conducted using Stata 14.0, which employs WMD or SMD to represent continuous variables, along with the calculation of 95% CIs. Cochrane’s Q test (p < 0.10) and I² measures [12] were employed to assess heterogeneity, and if the included trials exhibited no heterogeneity or low heterogeneity (I² < 50%, P > 0.1), a meta-analysis was conducted using the fixed-effects model; if there was a high degree of heterogeneity in the included trials (I² ≥ 50%, P ≤ 0.1), subgroup analysis or sensitivity analysis was performed to investigate the sources of heterogeneity in the outcome indicators. This was done to explore and address the sources of heterogeneity. If heterogeneity could not be resolved, a random-effects model was employed for analysis. Publication bias was assessed using Egger’s regression test [13]. Forest plots were created to illustrate the statistical results of the meta-analysis, and differences were considered statistically significant when the P value < 0.05.

Results

Search results and study characteristics

A total of 113 studies were initially searched, and 13 RCTs were eventually included in this meta-analysis [14–26], involving a total of 883 patients. In these RCTs, the average age of the experimental group ranged from 36.92 to 58.50 years, and the average age of the control group ranged from 34.71 to 56.40 years. Among the RCTs included, the experimental groups in six studies used routine rehabilitation training and SET [14, 16, 19, 22, 23, 25], whereas the experimental groups of the other seven studies included other therapies [15, 17, 18, 20, 21, 24, 26]. The duration of the intervention ranged from 1.5 to 3 months.

The PRISMA flowchart depicts the study selection procedure (Fig. 1).The baseline clinical characteristics of the included studies are summarized in Table 1.

Fig. 1.

PRISMA flow diagram

Table 1.

Summary of studies included in the meta-analysis

Study	Country	Design	EG						CG						Outcome
			Sample size	Age (year)	Male (%)	Intervention	Disease duration	Time	Sample size	Age (year)	Male (%)	Intervention	Disease duration	Time
Song(2023)	China	RCT	35	36.92 ± 4.35	60.0	Routine rehabilitation training + SET	1.96 ± 0.37 months	2 months	35	37.14 ± 4.41	57.1	Routine rehabilitation training	1.92 ± 0.34 months	2 months	ASIA, FMA - L, BBS, MBI, complication rate
Ma et al.(2023)	China	RCT	20	45.37 ± 7.02	55.0	Basic drug therapy + routine rehabilitation training + SET	43.07 ± 11.95 days	3 months	20	44.74 ± 6.83	60.0	Basic drug therapy + routine rehabilitation training	41.23 ± 10.83 days	3 months	MAS, BBS, ASIA-LEMS, Holden, MBI
Chen et al.(2023)	China	RCT	58	45.73 ± 6.61	51.7	Routine rehabilitation training + SET	2.01 ± 0.48 months	2 months	58	43.47 ± 5.54	55.2	Routine rehabilitation training	1.82 ± 0.55 months	2 months	Holden, step length, step speed, step frequency, BBS, ODI, ASIA, complication rate
Ma et al.(2022)	China	RCT	30	38.41 ± 5.76	56.7	Chinese herbal therapy + SET	2.4 ± 0.65 months	2 months	30	34.71 ± 6.61	50.0	Chinese herbal therapy	2.5 ± 0.81 months	2 months	MMT, BBS, MBI, step length, step speed, step frequency, ADL
Niu et al.(2022)	China	RCT	30	42 ± 12	63.3	Routine rehabilitation training + electroacupuncture + SET	38.10 ± 5.76 days	1.5 months	30	43 ± 12	73.3	Routine rehabilitation training + electroacupuncture	35.01 ± 5.64 days	1.5 months	ASIA, MBI, MAS, MEP
Li(2022)	China	RCT	38	48.59 ± 9.06	57.9	Routine rehabilitation training + SET	NA	3 months	37	48.52 ± 9.03	54.1	Routine rehabilitation training	NA	3 months	ASIA, step length, step speed, step frequency, BBS
Chen et al.(2021)	China	RCT	60	54.45 ± 6.67	45.0	Regular weight loss walking training + SET	3.13 ± 0.84 months	3 months	60	55.21 ± 7.23	46.7	Regular weight loss walking training	3.27 ± 0.78 months	3 months	MMT, 10MWT,6MWT, WISCI II, step length, step speed, step frequency, BBS, ADL, FIM
Gong et al.(2021)	China	RCT	40	37.34 ± 5.75	70.0	Routine rehabilitation training + EMG biofeedback training + SET	7.56 ± 2.24 months	2 months	40	36.96 ± 6.02	67.5	Routine rehabilitation training + EMG biofeedback training	7.51 ± 2.33 months	2 months	FAC, MAS, BBS, FMA
Wang et al.(2021)	China	RCT	32	43.3 ± 5.4	53.1	Routine rehabilitation training + SET	1.84 ± 0.78 months	2 months	32	45.0 ± 6.7	59.4	Routine rehabilitation training	2.03 ± 0.94 months	2 months	Step length, step speed, step frequency, ASIA, Static equilibrium function
Wang et al.(2020)	China	RCT	25	37.22 ± 7.46	84.0	Routine rehabilitation training + SET	23.84 ± 4.13 days	3 months	25	36.95 ± 8.21	88.0	Routine rehabilitation training	24.47 ± 3.79 days	3 months	FCA, ASIA-LEMS, SCI-FAI
Yin et al.2020	China	RCT	12	58.5 ± 10.9	83.3	Routine rehabilitation training + Conventional medication + MOTOmed virtual scenario training + SET	3.5 ± 0.5 months	2 months	13	56.4 ± 10.7	76.9	Routine rehabilitation training + Conventional medication + MOTOmed virtual scenario training	3.4 ± 0.5 months	2 months	ASIA-LEMS, TCT, BBS, 6MWT
Zhang et al.(2020)	China	RCT	44	37.51 ± 5.11	61.4	Routine rehabilitation training + SET	2.02 ± 0.48 months	2 months	44	38.05 ± 5.32	63.6	Routine rehabilitation training	2.05 ± 0.43 months	2 months	ASIA, BBS, MBI, ASIA-LEMS, Holden
Han et al.(2018)	China	RCT	17	53.4	52.9	Conventional neurotrophic therapy + routine rehabilitation training + SET	NA	3 months	18	52.7	55.6	Conventional neurotrophic therapy + routine rehabilitation training	NA	3 months	TUG, BBS

EG: experimental group; CG: control group; ASIA: American Spinal Cord Injury Association Scale; FMA – L: lower limb Fugl Meyer motor function assessment scale; BBS: Berg Balance Scale score; MBI: Modified Barthel Index score; MAS: Modified Ashworth Score; ASIA-LEMS: Lower extremity motor function score of the American Spinal Injury Association Motor Score; Holden: Holden walking ability classification; ODI: Oswestry disability index; MMT: Manual Muscle Test; ADL: Activity of Daily Living Scale; MEP: motor evoked potential; 10MWT: 10-minute walk test; 6MWT: 6-minute walk test; WISCI II: Walking Index for Spinal Cord Injury; FIM: Functional independent measures; FAC: Functional Ambulation Category Scale; FMA: Fugl-Meyer Assessment scale; FCA: Functional comprehensive assessment; SCI-FAI: Spinal cord injury-functional ambulation inventory; TCT: Trunk control test; TUG: Timed up and go test;

NA: Not appliable

The methodological quality of thirteen studies was assessed using the ROB 2 tool, all of which were considered some concerns. Of the thirteen studies, eight studies [16, 18, 19, 21, 22, 24–26] utilized the random number table method for grouping, while two studies [15, 20] mentioned randomization without providing specific details. Three studies [14, 17, 23] did not use a random allocation method. None of the studies reported allocation concealment, nor did any of the studies exhibit issues with baseline heterogeneity between groups. All studies were considered some concerns in the randomization process. All subjects in the studies were aware of their assigned interventions, and each study employed suitable analytical methods to assess the impact of the intervention assignments. All studies had access to outcome data for nearly all subjects. There was little to no impact on the outcome measures, even when subjects were aware of which intervention they were receiving. None of the studies selectively reported results. None of the studies exhibited domains of high risk, and there were domains with some concerns. So all studies were considered as some concerns. See Fig. 2. and Fig. 3. Although ROB 2 has revised certain terminology to improve clarity, some entries may still be insufficiently defined. This can lead to varying interpretations and judgments of the same issue among different assessors, ultimately affecting the consistency and reliability of the assessment results. In addition, ROB 2 lacks clear guidelines and methods for accurately assessing the magnitude and direction of bias.

Fig. 2.

Risk of bias graph

Fig. 3.

Risk of bias summary

The GRADE methodology was employed to assess the quality of the 13 included studies. The GRADE scores ranged from moderate to very low certainty, as detailed in Supplemental Table 2. One of the outcome indicators was rated as moderate, three were rated as low, and three were rated as very low.

Outcome of efficacy

ASIA-LEMS

Four RCTs [15, 23–25] reported ASIA-LEMS, and the heterogeneity test yielded I² = 3.3% < 50% and P = 0.376 > 0.1, suggesting that there was a small degree of heterogeneity among the RCTs, and a fixed-effects model was chosen for meta-analysis. Significant differences in ASIA-LEMS were found between the groups [WMD = 3.653, 95% CI (2.351–4.956), Z = 5.5, P < 0.05], which indicated that SET may improve the lower limb motor function of patients with SCI. See Fig. 4. The evidence supporting ASIA-LEMS was deemed to have a low certainty.

Fig. 4.

Forest plot of the effect of SET on ASIA-LEMS

Gait parameters (step length, step speed, and step frequency)

Five papers [16, 17, 19, 20, 22] recorded gait parameters (step length, step speed, and step frequency).

Step length

The step length was tested for heterogeneity to obtain I² = 13.6% < 50% and P = 0.327 > 0.1, suggesting that there was a small degree of heterogeneity among the RCTs, and a fixed-effects model was chosen for meta-analysis. The results revealed significant differences in step length between the groups[SMD = 0.655, 95% CI (0.462–0.849), Z = 6.64, P < 0.05], which indicated that SET may increase the step length of patients with SCI. See Fig. 5. The evidence supporting step length was deemed to have a moderate certainty.

Fig. 5.

Forest plot of the effect of SET on step length

Step speed

Step speed was tested for heterogeneity to obtain I² = 84.5% > 50% and P < 0.1, suggesting that there was a large heterogeneity among the RCTs, and a random-effects model was chosen for meta-analysis. The results revealed significant differences in step speed between the groups[SMD = 1.057, 95% CI (0.533–1.582), Z = 3.95, P < 0.05], which indicated that SET may increase the walking speed of patients with SCI. See Fig. 6. The evidence supporting step speed was deemed to have a very low certainty.

Fig. 6.

Forest plot of the effect of the SET on the step speed

To explore the sources of heterogeneity, subgroup analyses were performed according to the duration of the intervention. The results revealed that heterogeneity in each subgroup was not significantly reduced after grouping compared with the original subgroup, and the duration of intervention was not a major source of heterogeneity. See Fig. 7.

Fig. 7.

Forest plot of subgroup analysis of the effect of SET on step speed(by duration of the intervention)

To ensure the reliability of the results, sensitivity analyses were conducted using a random-effects model, which revealed no significant change in the results and no change in the direction of the combined results, indicating that the results were relatively stable and reliable. See Fig. 8.

Fig. 8.

Step speed sensitivity analysis

Step frequency

The step frequency was tested for heterogeneity to obtain I² = 82.4% > 50% and P < 0.1, suggesting that there was high heterogeneity among the RCTs, and a random-effects model was chosen for the meta-analysis. The results revealed significant differences in step frequency between the groups[SMD = 1.000, 95% CI (0.512–1.489), Z = 4.02, P < 0.05], which indicated that SET may increase the walking frequency of patients with SCI. See Fig. 9. The evidence supporting step frequency was deemed to have a very low certainty.

Fig. 9.

Forest plot of the effect of the SET on the step frequency

To explore the source of heterogeneity in this outcome indicator, subgroup analyses were performed according to the duration of the intervention. The results showed that heterogeneity in each subgroup was instead greater after grouping than before, suggesting that the duration of the intervention was not the main source of heterogeneity. See Fig. 10.

Fig. 10.

Forest plot of subgroup analysis of the effect of SET on step frequency (by duration of the intervention)

To ensure the reliability of the results, a random-effects model was used to conduct sensitivity analysis. There was no significant change in the results, and the direction of the combined results did not change, indicating that the results were more stable and reliable. See Fig. 11.

Fig. 11.

Step frequency sensitivity analysis

BBS

Ten papers [14–17, 19–21, 24, 25], and [26] recorded the BBS, and the heterogeneity test yielded I² = 66.9% > 50% and P = 0.001 < 0.1, suggesting that there was a large degree of heterogeneity among the RCTs, and a random-effects model was chosen for meta-analysis. The results revealed significant differences in BBS between the groups[WMD = 6.154, 95% CI (5.019–7.289), Z = 10.63, P < 0.05], which indicated that SET may improve the balance function of patients with SCI. See Fig. 12. The evidence supporting BBS was deemed to have a low certainty.

Fig. 12.

Forest plot of the effect of SET on the BBS

To explore the source of heterogeneity, subgroup analyses were performed on the basis of the mean age of the patients and duration of the intervention. The results showed that after patients were grouped by the mean age and duration of intervention, the heterogeneity of each subgroup was not significantly reduced from the original comparison, and the heterogeneity of each group was still greater than 50%, suggesting that neither the mean age nor the duration of intervention was a major source of heterogeneity. See Fig. 13 and supplemental Fig. 1.

Fig. 13.

Forest plot of subgroup analysis of the effect of SET on BBS (by mean age)

To ensure the reliability of the results, a random-effects model was used to conduct sensitivity analysis, with no significant change in the results and no change in the direction of the combined results, indicating that the results are more stable and reliable. See Supplemental Fig. 2.

MBI

Five papers [14, 15, 17, 18, 25] reported the MBI, and the heterogeneity test yielded I² = 93.2% > 50% and P < 0.1 for the Q-test, suggesting that there was a great deal of heterogeneity among the RCTs. The random-effects model was selected for the meta-analysis. The results showed that significant differences in MBI were found between the groups[WMD = 12.475, 95% CI (5.855–19.094), Z = 3.69, P < 0.05], which indicated that SET may improve activities of daily living in patients with SCI. See Supplemental Fig. 3. The evidence supporting MBI was deemed to have a very low certainty.

To ensure the reliability of the results, one or more studies were phased out, and a random-effects model was used to conduct sensitivity analysis, which revealed no significant change in the results and no change in the direction of the combined results, indicating that the results were relatively stable and reliable. See Supplemental Fig. 4.

MAS

Three papers [15, 17, 21] reported MAS, and the heterogeneity test yielded I² = 6.6% < 50% and P = 0.343 > 0.1, suggesting that there was a small degree of heterogeneity between the RCTs, and a fixed-effects model was chosen for the meta-analysis. Significant differences in MAS were found between the groups[WMD = -0.756, 95% CI (-0.879 to -0.632), Z = 12.01, P < 0.05], which indicated that SET may reduce muscle tone in patients with SCI. See Supplemental Fig. 5. The evidence supporting MAS was deemed to have a low certainty.

Publication bias

Publication bias was assessed using Egger’s test for the main outcome measures. The results revealed the following: ASIA-LEMS P = 0.365 > 0.05; step length P = 0.43 > 0.05; step speed P = 0.752 > 0.05; step frequency P = 0.468 > 0.05; and BBS P = 0.908 > 0.05.The results indicated that there was no significant publication bias in this study. See Table 2.

Table 2.

Assessment of publication bias for primary outcome measures

Outcomes	Std_EFF	Coef.	Std.Err.	t value	95% CI	P value
ASIA-LEMS	Bias	1.101311	0.9469034	1.16	-2.972886,5.175507	0.365
Step length	Bias	-3.190108	3.504186	-0.91	-14.34199,7.961775	0.43
Step speed	Bias	3.188866	9.201274	0.35	-26.09369,32.47143	0.752
Step frequency	Bias	5.434621	6.563318	0.83	-15.45279,26.32203	0.468
BBS	Bias	0.1842619	1.546348	0.12	-3.381623,3.750147	0.908

Adverse effects

The lack of adverse effects reported in the included literature may be due to the high safety profile of SET, with few to no adverse effects in patients. Owing to the small sample size of the included studies, the safety of SET needs to be further explored.

Discussion

This research represents an inaugural comprehensive systematic review and meta-analysis evaluating the effectiveness of SET in individuals with SCI. In this study, the authors reported significant differences in ASIA-LEMS, gait parameters, BBS, MBI and MAS between the experimental and control groups, indicating that SET is likely to contribute to the functional recovery of patients with SCI. According to Egger’s test, no evidence of publication bias was detected in any of the pooled studies. Furthermore, the results of the sensitivity analyses for each outcome were robust and reliable. Other outcome metrics, such as holden walking ability classification and the TUG, were addressed in the included studies. These outcome indicators were not included in the results in this study because the number of papers that included these outcome indicators was small and the data were insufficient for meta-analysis.

In this study, four papers [15, 23–25] reported ASIA-LEMS score. The ASIA score serves as an international standard for assessing the degree of functional impairment in SCI patients. SET has been demonstrated to alleviate motor dysfunction in patients, likely because of its capacity to enhance neuromuscular coordination among trunk muscles and the primary muscle groups of the body, thereby contributing to improved neuromuscular function [27].

Five studies [16, 17, 19, 20, 22] recorded gait parameters, including step length, step speed and step frequency. The significant improvements in these parameters among the experimental groups compared with those of the control groups indicate that the patients’ walking abilities have been enhanced and that motor function is gradually recovering. In the subgroup analysis of step speed, the authors reported that the intervention time was not the main source of heterogeneity. The source of heterogeneity was likely because the control groups had different interventions, with Ma (2022) [17] using traditional Chinese medicine and Chen (2021) [20] using weight-loss walking training. In the subgroup analysis of step frequency, intervention time was not the main source of heterogeneity. Analyzing the source of heterogeneity by reading through the whole article may be the same as the source of heterogeneity in step speed, which is significant because of the different interventions in the control groups.

Ten papers [14–17, 19–21, 24–26] reported the BBS, and the experimental groups showed a significant improvement in balance function compared with the control groups. This may be because SET enhances spinal stability by strengthening and coordinating trunk muscle groups, thereby improving the balance, control and stability of the body during exercise [28]. The BBS is a thorough assessment tool designed to evaluate a patient’s dynamic and static balance in both seated and standing postures. Balance control is a multifaceted motor skill that necessitates intricate collaboration between the nervous system and muscular system to achieve effective balance regulation [29]. Enhancing patients’ balance function is advantageous for facilitating the execution of functional movements and for augmenting ambulation capabilities. In the subgroup analysis, neither age nor intervention time was a major source of heterogeneity. Reading through the whole article to analyze the source of heterogeneity could be due to the different interventions for patients in the control groups or the different durations of the patients’ disease, with two papers [19, 26] unable to obtain the duration of the patients’ disease, and one paper [21] with an average duration of the patients’ disease of approximately 7 months, which was significantly greater than the duration of the patients’ disease in the rest of the literature.

Five papers [14, 15, 17, 18, 25] reported a significant improvement in the ability to perform activities of daily living in the MBI, and experimental groups compared with the control groups. SCI increases the time required for self-care and reduces mobility, which can seriously affect patients’ activities of daily living [30]. Improving patients’ capacity to engage in activities of daily living has the potential to alleviate the burden on caregivers and facilitate more effective reintegration of patients into society. Subgroup analyses revealed that average age was not a significant contributor to the observed heterogeneity. These results may be due to the different frequencies of suspension training exercises performed by the patients. In Song (2023) [14], the frequency of patients performing SET was 3 times a day, 5 days a week, whereas in Ma (2022) [17] and Zhang(2020) [25], the frequency of patients performing SET was once a day, 6 days a week and once a day, 5 days a week, respectively.

Three papers [15, 17, 20] reported a significant decrease in muscle tone in the MAS in the experimental groups compared with the control groups. The main motor symptoms in patients with SCI are muscle atrophy and spasticity, and patients often lose motor function due to spasticity [31], and a decrease in muscle tone is beneficial for reducing motor dysfunction.

SET has the potential to rehabilitate functional impairments resulting from neurological disorders [32]. reported that SET can improve balance function in children with cerebral palsy. In terms of motor function, it has been shown [33] that SET can improve the lower limb motor function and walking ability of patients, enhance their lower limb motor function, improve their neurological function and improve their quality of life, which is consistent with the results of these analyses.

SET encompasses both diagnostic and therapeutic methodologies. By identifying weak links through targeted testing, this training approach aims to facilitate muscle relaxation, improve mobility, and enhance sensory and motor functions [34]. In clinical treatment, SET can be used as a supplement to traditional rehabilitation therapy to enhance the therapeutic effect [14]. In addition, SET can be combined with other treatments, such as electroacupuncture therapy [18] and biofeedback therapy [35], to improve patient dysfunction and positively affect patient movement, balance and other functions.

The present study is subject to several limitations: (1) All included studies were from Chinese databases, weakening the comprehensiveness and universality of the synthesis. It is hoped that more high-quality cross-national RCTs will be available in the future to compensate for this shortcoming. (2) None of the 13 studies reported the occurrence of adverse reactions during treatment. (3) The studies did not provide information regarding follow-up assessments. (4) Most outcomes presented in this study exhibit low to very low certainty. There remains a deficiency of high-quality multicenter RCTs have investigated the efficacy of SET for the treatment of SCI.

Conclusion

In summary, the findings of this study indicate that SET is likely to be effective in ameliorating dysfunction in individuals with SCI. However, given the current limitations in the literature, there is a pressing need for additional high-quality randomized controlled trials with larger sample sizes to further validate the effectiveness of SET for this patient population.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analysis

RCTs

Randomized Controlled Trials

WMD

The Weighted Mean Difference

SMD

The Standard Mean Difference

ASIA-LEMS

Lower Extremity Motor Function Score of the American Spinal Injury Association Motor Score

BBS

Berg Balance Scale Score

MBI

The Modified Barthel Index score

MAS

The Modified Ashworth Score

SCI

Spinal Cord Injury

SET

Suspension Exercise Training

TSCI

Traumatic Spinal Cord Injury

ASIA

American Spinal Injury Association

CI

Confidence Interval

MRI

Magnetic Resonance Imaging

CT

Computed Tomography

GRADE

Grading of Recommendations, Assessment, Development, and Evaluation

Author contributions

L.Y.T. and P.J.G contributed to the design. L.Y.T. wrote the manuscript and conducted the statistical analysis. P.J.G. and W.M.M.checked the manuscript data and made further revisions to the manuscript. Z.L.W. provided financial support. All authors have read and approved the content of the manuscript.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Financial benefits to the authors

None.

Competing interests

The authors declare no competing interests.