Sage Journals: Discover world-class research

Abstract

BACKGROUND:

Spinal cord injury (SCI) is a disabling condition characterized by multilevel skeletal muscle impairment and rapid cortical and trabecular bone loss. Rehabilitation is a cornerstone of the long-term management of patients with SCI; however, the optimal rehabilitation strategy for improving bone health has not been fully characterized.

OBJECTIVE:

To characterize the current evidence supporting different rehabilitation interventions improving bone health in patients with SCI.

METHODS:

On November 17 ${}^{\text{th}}$ , 2022, five databases (PubMed, Scopus, Web of Science, Cochrane, and PEDro) were systematically searched for randomized controlled trials (RCTs) assessing SCI patients undergoing rehabilitation interventions. The primary outcomes were bone macroscopical effects. Secondary outcomes were changes in bone metabolisms and functional outcomes.

RESULTS:

Out of 499 records, 11 RCTs met the eligibility criteria and were included. Electrical stimulation combined with physical exercise was assessed by 5 studies, standing intervention was assessed by 3 studies, vibration was assessed by 1 study, ultrasound therapy was assessed by 1 study, and electroacupuncture combined with a pulsed magnetic field was assessed by 1 study. The rehabilitation intervention was administered combined with pharmacological treatment (3 studies) or alone (8 studies). Positive effects in terms of BMD were reported by 3 studies. The quality assessment revealed some concerns in 9 out of 11 studies, in accordance with the Cochrane Risk of Bias assessment – version 2.

CONCLUSION:

Our data suggest that multicomponent interventions including rehabilitation might be considered a suitable option to improve bone health management in SCI patients. Further studies are mandatory to characterize the optimal combination of non-pharmacological interventions reducing bone loss and improving the risk of fractures in patients with SCI.

Keywords

Osteoporosis spinal cord injury rehabilitation bone health physiotherapy

1. Introduction

Spinal cord injury (SCI) is a disabling neurological condition characterized by the loss of both motor and sensory function and visceral dysfunction [1, 2, 3, 4]. The prevalence of SCI is highly heterogeneous among different countries, ranging from 13.0 per million to 163.4 per million people, with a significant proportion of traumatic spinal cord injuries affecting young adults [5, 6, 7]. Regrettably, SCI is related to extremely high healthcare and assistance costs, while social and personal costs cannot be quantified [5, 6, 7].

After spinal cord injury, bone loss is a common complication related to skeletal muscle system disuse and impaired mechanical stimuli [3, 8]. More in detail, osteoporosis generally affects the skeletal system caudal to the spinal cord damage, with bone loss occurring most rapidly and significantly in the distal femur and proximal tibia [9, 10]. Within two to three years of SCI, individuals show a 50–100% trabecular bone mineral density (BMD) reduction, and 40–80% lower cortical bone mass [11, 12, 13]. As a result, SCI patients have a 20- to 100-fold higher fracture risk compared to the general population [14]. Therefore, it has been proposed that the comprehensive management of bone loss should be performed in patients with SCI, including both pharmacological and non-pharmacological approaches [15].

In this context, growing evidence highlighted the crucial role of a comprehensive rehabilitation program targeting both physical and psychosocial impairment of SCI patients [8, 16, 17, 18, 19, 20]. To date, several studies reported positive effects of physical exercise and physical activity on bone health in several disabling conditions [21]. On the other hand, rehabilitation might effectively target not only bone health but also balance and the risk of falls [22].

Despite these considerations, the optimal rehabilitation strategies to prevent bone loss in people with SCI have yet to be fully characterized [23, 24, 25, 26]. Furthermore, to our knowledge, no previous systematic review including only randomized controlled trials (RCTs) assessed the effects of specific rehabilitation protocols on bone health of SCI patients.

Therefore, the objective of this systematic review was to provide evidence supporting different rehabilitation strategies for improving bone health in patients with SCI. Moreover, we aimed at characterizing the effects of specific rehabilitation prescriptions in bone health in order to pave the way to an evidence-based approach preventing bone loss in people with SCI.

2. Methods

2.1 Registration

This systematic review of RCTs follows the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) statement [27]. A preliminary search has been performed in the international prospective register of systematic reviews (PROSPERO) for comparable review protocols in progress, without finding similar studies. The systematic review was then submitted to PROSPERO and accepted on 27 ${}^{\text{th}}$ November 2022 with registration number CRD42022376430 (available at https://www.crd.york.ac.uk/prospero).

2.2 Search strategy

The literature search has been performed from onset up to November 17 ${}^{\text{th}}$ , 2022. Five databases (PubMed/Medline, PEDro, Web of Science, Cochrane Central Register of Controlled Trials - CENTRAL, and Scopus) were systematically searched independently by two investigators. The keywords “Spinal cord injury”, “Bone Loss”, “Osteoporosis”, “Management”, “Bone mineral density” were used, in addition the MeSH terms available on PubMed and the Boolean operators AND and OR were used. Supplementary Table 1 reports the full search strategy for each database.

2.3 Selection criteria

In accordance with the PICO model [28], we considered eligible all randomized controlled trials (RCT) satisfying the following criteria:

•
P) Participants: adult (aged more than 18 years) male or female patients with spinal cord injury, without restriction in terms of time of injury.
•
I) Intervention: all rehabilitation interventions administered as exclusive or integrated therapeutic intervention.
•
C) Comparator: any comparator.
•
O) Outcome: the primary outcomes were bone macroscopical effects, assessed by: i. Bone Mineral Density (BMD); ii. Bone Mineral Content (BMC); iii. Trabecular Bone Thickness (TBTh) and Trabecular Bone Separation (TBSp); iv. bone strength indices [Compressive Strength Index (CSI), Cortical Thickness Index (CTI), Cortical Bone Volume (CBV), Buckling Ratio (BR), and Torsional Strength Index (TSI)]; v. quantitative bone ultrasound (qUS) parameters; vi. Cross-Sectional Area (CSA) of red and yellow bone marrow and cortical and trabecular bone. The secondary outcomes were: i. changes in body composition; ii. change in bone biomarkers; iii. hormonal and metabolic changes; iv. changes in physical functioning and physical performance.

Only RCTs published in peer-reviewed International Journal were included. No publication date restriction was applied.

The exclusion criteria were: i. participants with pregnancy, clinical instability (defined as hemodynamic alterations, respiratory events, abnormal laboratory values, reduced level of consciousness or temperature alterations), or palliation; ii. studies involving animals; iii. doctorate theses, conference proceedings, and reviews of the literature; iv. not RCT studies; v. language other than English.

Firstly, the articles resulting from the database-specific search strings were examined for duplication removal. Subsequently, the titles and abstracts were screened by two investigators that independently excluded records that did not meet the inclusion criteria. Lastly, the selected articles were examined in full text, and the articles included in the review were thus extracted. Any conflicting record along this chain was discussed between the two investigators, and in case of unresolved disagreement, a third reviewer was involved to reach consensus.
2.4 Data extraction and synthesis

All data were extracted by two independent reviewers through Excel database. Any difference was solved by discussion between the two reviewers or by consulting a third reviewer.

The data extracted were: i. authors, journal, publication year, and nationality; ii. study design; iii. sample characteristics [number of participants, mean age and age range, gender, time of injury, lesion level and grade, baseline body mass index (BMI) and osteoporosis grade]; iv. comparator characteristics; v. intervention characteristics [type, duration, frequency, and intensity of treatment in the experimental group and in the control group(s)]; vi. primary and secondary outcomes; vii. duration of the study (with possible follow-up); viii study results.

Data were independently synthesized by two reviewers. A qualitative analysis of the evidence was performed, given that the heterogeneity of the study samples, the type of intervention and the outcome measures did not allow a quantitative analysis of the data.

A subgroup analysis was performed based on participant characteristics, time from SCI, type of intervention, and outcome assessed.

2.5 Quality assessment and risk of bias

The qualitative analysis of the studies included in the review was carried out using the PEDro scale [29], which consists of 11 items. In particular, the first item assesses the external validity (or applicability) of the study although is not considered for the purposes of the total score; items 2–9 evaluate the internal validity of the study, while items 10–11 evaluate the interpretability of the results based on the statistical information. Each criterion has “yes” and “no” response options, 1 point for each item whose answer is “yes”, 0 points for items whose answer is “no”. The maximum score obtainable is 10/10. Scores from 9 to 10 were considered “excellent”, 6 to 8 were considered “good”, 4 to 5 were considered “fair”, and $<$ 4 were considered “poor” [29].

The risk of bias assessment of the studies included in the review was performed following the Cochrane Risk of Bias assessment – version 2 (RoB 2) [30], a validated tool for assessing the risk of bias in RCTs. It includes five different domains, each subdivided in criteria that are judged individually (low, high, unclear risk of bias). The overall judgment of a domain derives from the overall judgment of the various criteria. Studies were considered with low risk of bias if all domains evaluated had low risk of bias. Studies in which at least one domain presented unclear risk, were considered studies with some critical issues. Lastly, studies were considered with high risk of bias if at least one domain had a high risk of bias [30].

3. Results

Figure 1.

PRISMA 2020 flow diagram.

3.1 Study characteristics

Overall, a total of 499 records resulted from the 5 databases assessed, while no other records were identified from other sources. After duplication removal, a total of 363 studies were assessed for eligibility and screened for title and abstract. As a result, 319 records were excluded, and 44 studies were sought for retrieval and then screened in full text. Lastly, 11 studies [31, 32, 33, 34, 35, 36, 37, 38, 38, 40, 41] were included in the present systematic review. Figure 1 shows the PRISMA flow diagram, reporting further details about the article selection process. Supplementary Table 2 summarizes the reason for exclusion of the remaining 33 studies assessed full-text for eligibility.

The following studies were included in the present systematic review: Afshari et al. [31], Arija-Blázquez et al. [32], Ben et al. [33], Chen et al. [34], Edwards et al. [35], Groah et al. [36], Holman et al. [37], Morse et al. [38], Shackleton et al. [39], Warden et al. [40], Xu et al. [41].

The studies included were published between2001 [40] and 2022 [31, 39] and were conducted in America (USA $n=$ 3 [31, 36, 37, 38]; collaboration between USA and Canada $n=$ 1 [35]), Europe (Spain $n=$ 1 [32]), Africa (South Africa $n=$ 1 [39]) and Asia (China $n=$ 2 [34, 41], Australia $n=$ 2 [33, 40]).

The present review included a total of 301 patients (250 males, and 51 females). The patients were characterized by a mean age ranging between 23.9 $\pm$ 7.3 years [40] and 47.6 $\pm$ 16.3 years [35]. BMI was reported only by four studies [31, 32, 37, 38], and ranged between 23.94 $\pm$ 3.62 kg/m2 [31] and 26.5 $\pm$ 4.9 kg/m2 [32]. Both post-acute and chronic SCI were assessed, with time since injury varied between 35.9 $\pm$ 16.9 days [36] and 21.1 $\pm$ 13.4 years [35]. More in detail, post-acute subjects were enrolled in five studies: Arija-Blázquez et al. [32], Ben et al. [33], Groah et al. [36], Warden et al. [40], and Xu et al. [41]. Chronic subjects were assessed in six studies: Afshari et al. [31], Chen et al. [34], Edwards et al. [35], Holman et al. [37], Morse et al. [38], and Shackleton et al. [39].

Patients with cervical SCI were assessed in five studies [31, 35, 37, 39, 40]; patients with thoracic SCI were evaluated in eight studies [31, 32, 34, 35, 37, 39, 40, 41], and patients with lumbar SCI was assessed in two studies [35, 41]. Two studies did not clarify the level of injury [33, 38]. Asia Impairment Scale (AIS) was the most used tool used to characterize the functional impairment of SCI ( $n=$ 6 [31, 35, 36, 37, 39, 40]).

Table 1
Literature search strategy on different databases

PubMed: (“spinal cord injury” [MeSH Terms] OR “spinal cord trauma” [All Fields] OR “spinal cord injuries” [All Fields] OR “spinal cord transection” [All Fields] OR “spinal cord laceration” [All Fields]) AND (“osteoporosis” [MeSH Terms] OR “osteoporosis” [All Fields] OR “bone loss” [All Fields]) AND (“management” [All Fields] OR “treatment” [All Fields] OR “therapy” [All Fields] OR “therapies” [All Fields] OR “treatments” [All Fields] OR “prevention” [All Fields]) AND (“bone density” [MeSH Terms] OR “bone mineral density” [All Fields] OR “bone mineral content” [All Fields] OR “risk of fracture” [All Fields] OR “risk of fractures” [All Fields]).

Physiotherapy Evidence Database (PEDro): spinal cord injur* osteoporosis

Web of Science: (spinal cord injury) AND (osteoporosis OR bone loss) AND (management OR treatment OR prevention) AND (bone mineral density OR risk of fracture)

Cochrane Central Register of Controlled Trials (CENTRAL) Advanced Search: ID Search #1 MeSH descriptor: [Spinal Cord Injuries] explode all trees #2 MeSH descriptor: [Osteoporosis] explode all trees #3 MeSH descriptor: [Therapeutics] explode all trees #4 MeSH descriptor: [Bone Density] explode all trees #5 #1 AND (#2 AND #3) AND #4

Scopus: TITLE-ABS-KEY: spinal cord injury AND ((osteoporosis OR osteoporosis) AND (treatment OR management OR therapy OR prevention)) AND (bone mineral density OR bmd OR risk of fracture)

Table 1 summarizes in detail the sample characteristics of both intervention groups and comparator groups of each study included in the present review.

3.2 Intervention and control characteristics

Heterogeneous rehabilitation modalities were assessed by the RCTs included. In particular, rehabilitation interventions currently proposed were:

•
Functional Electrical Stimulation (FES) combined with physical exercise was assessed in five studies [31, 32, 36, 37, 38]. Afshari et al. [31] positioned electrodes on the quadriceps and hamstrings muscles and delivered a stimulus to produce full knee flexion-extension during arms rowing training. The group compared this intervention against conventional physical rehabilitation arm and rowing training arm [31]. Similarly, Morse et al. [38] assessed the effects of FES and rowing training combined with Zoledronic acid 5 mg/100mL [38]. Arija-Blázquez et al. [32] performed bilateral FES, placing three electrodes on quadriceps muscle to produce contraction, comparing it to sham FES [32]. Groah et al. [36] performed FES bilaterally on the quadriceps muscle combined with non-specified individualized inpatient rehabilitation program, comparing it to the same rehabilitation program without FES [36]. Holman et al. [37] used FES to perform open kinetic chain neuromuscular electrical stimulation-evoked resistance training, with electrodes applied on distal and proximal thigh. The intervention group (IG) was also treated with transdermal testosterone patches, compared with transdermal testosterone patches alone [37].
•
Standing modalities were studied in threeRCTs [33, 34, 39]. More in detail, Ben et al. [33] tested weight-bearing and stretch achieved through a block under the experimental foot combined with a tilt table standing. The control was the opposite foot, without weight-bearing [33]. Chen et al. [34] assessed an intelligent standing mobile robot (XZ-Droid Home) with which the patient performed squatting and standing up, and tested it against a regular standing frame. Both IG and control group (CG) underwent rehabilitation, comprising management of pain and spasticity, preventing injury progression, managing secondary complications, guiding strength training, mat mobility and transfer, and personalized rehabilitation education [34]. Lastly, Shackleton et al. [39] used robotic locomotor training to achieve exoskeleton walking in IG. This training modality was compared to activity-based training, which focused on prehabilitation, muscle recruitment, posture and joint stability, resistance and endurance training, pregait, and gait training [39].
•
Vibration therapy was studied by Edwards et al. [35]. The group used a vibration platform over which the patient placed both feet flat while remaining seated. This IG was compared to the administration of teriparatide 20 $\mu$ g/day and sham vibration, and a group with both pharmacological and non-pharmacological interventions [35].
•
Ultrasound (US) therapy was assessed by Warden et al. [40]. The authors applied active coupled US-heads on medial and lateral surfaces of the calcaneus, comparing it with sham treatment on contralateral calcaneus [40].
•
Electroacupuncture combined with a pulsed magnetic field was assessed by Xu et al. [41]. More in detail, electroacupuncture was performed by treating points GB 34 and GB 39 on both sides. This treatment was combined with XT-20 0 0B Osteoporosis Treatment Apparatus (pulsed magnetic field below the level of spinal cord injury), Reneed tablet, and rehabilitation treatment (standing, limb air pressure, limb function training) [41].

Rehabilitation was combined with pharmacological and nutritional intervention as it follows. Morse et al. [38] treated IG patients with a one-time infusion of Zoledronic acid 5 mg/100mL solution [38]. Holman et al. [37] used transdermal testosterone patches in both IG and CG. In the study by Edwards et al. [35], control groups received teriparatide 20 $\mu$ g/day [35]. Moreover, all subjects received 1000 mg/day of calcium carbonate and 1000 IU/day of cholecalciferol. Lastly, in the study by Xu et al. [41] SCI patients received Reneed tablet 500 mg daily in both IG and CG [41]. All interventions and control protocols characteristics are shown in detail in Table 1.
3.3 Primary outcomes

BMD was evaluated in eight studies [31, 32, 33, 34, 35, 36, 39, 41]. Afshari et al. [31] assessed leg, pelvis and total BMD, and reported a significantly lower decrease in pelvis BMD ( $p=$ 0.028), a positive effect that was reported also after the CGs crossover (BMD Pelvis, $p<$ 0.05). Moreover, the group reported a significant effect in preventing total BMD decrease ( $p=$ 0.039) [31]. Arija-Blázquez et al. [32] assessed leg, femoral neck, trochanteric, intertrochanteric, ward’s triangle, whole hip and lumbar area BMD, without reporting any differences between IG and CG ( $p=$ NS) [32]. Ben et al. [33] evaluated total proximal femur BMD (IG vs CG: 0.857 $\pm$ 0.131 vs 0.848 $\pm$ 0.142 g/cm2, $p=$ NR) and reported that the effect of the intervention was reducing the bone loss by 0.5% ( $-$ 1.8% to 2.9%) ( $p=$ NR) compared to the control group [33]. Edwards et al. [35] evaluated areal BMD at spine, hip, femoral neck, proximal and distal femur, and proximal tibia, and reported a significant difference in spine BMD in the CGs (treated with teriparatide, $p=$ 0.09). Assessing volumetric BMD (vBMD) for trabecular bone of femur and tibia epiphyseal, metaphyseal, and diaphyseal regions, the authors reported significant differences over time in the IG at the trabecular vBMD femoral metaphysis ( $p=$ 0.005). Moreover, significant differences were underlined in CGs at the trabecular vBMD femoral epiphysis and metaphysis ( $p=$ 0.08 and 0.005, respectively) [35]. Chen et al. [34] evaluated BMD and assessed T score and Z score, which demonstrated a significant difference between IG and CG ( $p=$ 0.04 and $p=$ 0.03, respectively). Groah et al. [36] assessed lumbar, hip, femoral and tibial BMD at 6 weeks and at 3 months after intervention without reporting any significant difference in intragroup and between group analysis ( $p=$ NS) [36]. Shackleton et al. [39] reported a significant decrease in CG at the level of hip and femoral neck BMD ( $p<$ 0.05) [39]. Lastly, Xu et al. [41] did not find any significant differences in the intragroup and intergroup analysis in femoral neck, greater trochanter and Ward’s triangle BMD ( $p=$ NS) [41].

BMC was assessed in three studies [31, 35, 40]. Afshari et al. [31] did not find significant differences in leg, pelvis, and total BMC ( $p=$ NS). Edwards et al. [35] evaluated BMC for cortical bone of femur and tibia at epiphyseal, metaphyseal, and diaphyseal regions, and found significant differences over time in the IG at the cortical BMC femoral metaphysis and diaphysis ( $p=$ 0.001 and 0.054 respectively). Moreover, significant differences over time were found in the CGs at the cortical BMC femoral metaphysis and diaphysis ( $p=$ 0.001 and 0.054 respectively) [35]. Warden et al. [40] found a significant worsening over time ( $p<$ 0.001); however, no significant differences were found in between groups analysis ( $p=$ NS) [40].

Trabecular bone thickness and trabecular bone separation parameters were assessed by Holman et al. [37] for distal femur and proximal tibia, though the authors did not perform statistical inference and multiple comparisons of the results [37].

Bone strength indices were assessed in two studies [35, 38]. Edwards et al. [35] computed CSI and TSI for epiphyseal, metaphyseal, and diaphyseal regions of femur and tibia, without reporting significant differences ( $p=$ NS). The authors also assessed BV for cortical bone of femur and tibia at epiphyseal, metaphyseal, and diaphyseal regions, without reporting any differences [35]. Morse et al. [38] assessed CBV, CTI and BR at the proximal tibia and distal femur level and reported significant differences between IG and CG (all $p<$ 0.05) in all but distal femur CBV ( $p=$ 0.05) [38].

Bone ultrasound attenuation (BUA) and speed of sound (SOS) were evaluated in one study [40]. Although these parameters had a significant worsening over time ( $p<$ 0.001), no significant differences were found in the between-groups analysis ( $p=$ NS) [40].

Table 2
Main characteristics of the studies included in the systematic review

Table 2, continued

Holman et al. [37] assessed the yellow and red bone marrow CSA and cortical bone CSA at proximal, middle, and distal femoral levels, though the authors did not perform statistical inference and multiple comparisons of the results [37]. Further details are shown in Table 2.

3.4 Secondary outcomes

Changes in bone biomarkers were assessed in four studies [32, 35, 36, 41].

•
Serum alkaline phosphatase (ALP) was analyzed in the study by Xu et al. [41], reporting a significant difference in the intragroup analysis in both groups and in the intergroup analysis ( $p<$ 0.05) [41].
•
Procollagen type I C-peptide (PICP) was analysed in one study [41]. Xu et al. reported a significant difference in the intragroup analysis in both groups ( $p<$ 0.05) [41].
•
Bone gamma-carboxyglutamic acid containing protein (BGP) was analyzed in one study [41], reporting a significant difference in the intragroup analysis and in the intergroup analysis ( $p<$ 0.05) [41].
•
Osteocalcin was evaluated in two studies [32, 36], without reporting significant results ( $p=$ NS) [32, 36].
•
Urinary N-terminal telopeptide (N-TX) was evaluated in one study [36], at 6 weeks and at 3 months after intervention without reporting significant differences in intragroup and between group analysis ( $p=$ NS) [36].
•
Collagen type 1 cross-linked C-telopeptide (CTX) was analyzed in two studies [32, 35], which did not report significant changes following intervention [32, 35].
•
The ratio Osteocancin/CTX was calculated in one study [32], that did not report significant changes [32].
•
Type 1 procollagen amino-terminal propeptide(P1NP) was analyzed in one study [35], which did not report significant changes following intervention [35].
•
Bone-specific alkaline phosphatase (BSAP) was analyzed in one study [35], which did not report significant changes following intervention [35].

Lastly, changes in physical functioning and physical performance were assessed by Afshari et al. [31], reporting significant improvement in the IG ( $p<$ 0.001) in terms of exercise capacity, without reporting significant effects in CGs [31]. See Table 2 for further details.

Figure 2.
Cochrane Risk of Bias assessment – version 2 (RoB 2).

Table 3
Main findings of the studies included in the systematic review

Table 3, continued

Table 4
Qualitative assessment of the studies included using the PEDro scale

1. Eligibility criteria 2. Random allocation 3. Concealed allocation 4. Baseline comparability 5. Blind subjects 6. Blind therapists 7. Blind assessors 8. Adequate follow-up 9. Intention- to-treat analysis 10. Between- group comparisons 11. Point estimates and variability Total

Afshari et al. 2022 1 1 0 0 0 0 0 0 0 1 1 3

Arija-Blázquez et al. 2014 1 1 0 1 1 0 0 1 0 1 1 6

Ben et al. 2005 1 1 1 1 0 0 1 1 1 1 1 8

Chen et al. 2020 1 1 0 1 0 0 1 1 0 1 1 6

Edwards et al. 2018 1 1 1 1 1 1 0 1 0 1 1 8

Groah et al. 2010 0 1 0 1 0 0 1 0 0 1 1 5

Holman et al. 2020 1 1 0 1 0 0 0 1 1 1 1 6

Morse et al. 2019 1 1 0 0 1 0 0 0 0 1 0 3

Shackleton et al. 2022 0 1 0 1 0 0 0 1 0 1 1 5

Warden et al. 2001 1 1 0 1 1 0 1 0 0 1 1 6

Xu et al. 2020 1 1 0 1 0 0 0 1 0 1 1 5

Note: Eligibility criteria item does not contribute to total score. 0 $=$ NO, 1 $=$ YES. Scores from 9 to 10 are considered “excellent”, 6 to 8 are considered “good”, 4 to 5 are considered “fair”, and $<$ 4 are considered “poor”.

3.5 Quality assessment

In accordance with the PEDro scale [29], two studies resulted in poor quality [31, 38], three studies were characterized as fair [36, 39, 41], while the remaining six studies resulted in good quality scores [32, 33, 34, 35, 37, 40]. Table 3 shows in detail the results of the PEDro scale for each item.

The risk of bias assessment was performed according to the Cochrane Risk of Bias assessment – version 2 (RoB 2) [30]. Some concerns emerged from this analysis in domain 2, specifically “Deviations from the intended interventions”, due to lack of an appropriate analysis estimating the effect of assignment to intervention. This resulted in total outcome of “some concerns” for 9 studies out of 11 [31, 32, 34, 35, 36, 38, 39, 40, 41]. Further details about the risk of bias assessment of each study are shown in Fig. 2.

4. Discussion

Rehabilitation is currently considered a cornerstone in the complex framework of care of SCI patients, with growing research suggesting several therapeutic strategies to improve bone health in these patients [3, 8, 16]. However, bone loss still remains a major issue in SCI patients and the optimal rehabilitation approach to prevent this burdensome condition has not been standardized yet.

Thus, this systematic review of RCTs aimed at summarizing the current evidence supporting the different therapeutic strategies to improve bone health in SCI patients trying to characterize the specific biological effects of a comprehensive rehabilitation approach to these patients.

Interestingly, our findings identified several therapeutic interventions to improve bone health in SCI patients, including FES, physical exercise, vibration therapy, standing, ultrasound, and electroacupuncture combined with pulsed magnetic field. Taken together, the results of the present systematic review underlined that several instrumental therapies were assessed to improve bone health of SCI patients, but with conflicting results and only 3 studies suggested positive effects in terms of BMD improvement at lower limb level [31, 33, 34].

While the effects of physical therapies on bone health are still controversial in the current literature [42, 43, 44], physical exercise is considered a non-pharmacological intervention supported by several guidelines in both osteoporosis prevention and management, due to its multitarget effects on the whole musculoskeletal system, improving both BMD and reducing the risk of falls [45, 46, 47]. Despite these considerations, few studies [31, 32, 36, 37, 38] assessed the effects of physical exercise on SCI patients. However, people with SCI could be affected by detrimental musculoskeletal impairment, with crucial implications in biological responses to physical exercise [48, 49]. Indeed, as part of normal bone metabolism in adult individuals, bone tissue undergoes a continuous process of resorption and formation; when this turnover is at equilibrium, it results in unchanged bone mass, while when there is unbalance, it results in bone changes (growth, aging) and pathology. More in detail, osteoclasts initiate bone resorption, osteoblasts are responsible for bone formation, while osteocytes are the cells dedicated at maintaining bone tissue [50]. On the other hand, as early as 1892, it was determined that mechanical loads in live beings can influence bone architecture. This concept is known as Wolff’s law, which was further explored till defining that loading on load-bearing bones determines the majority of their strength in adult life [51]. In this context, voluntary load bearing is lacking in SCI patients’ life. In this context, it has been suggested that spasticity might have a protective role in bone tissue in patients with neurological disability [52, 53], highlighting the close link between bone and muscle tissues due to their cross-regulation promoted by mechanical forces, in accordance with the mechanostat theory theorized by Frost [54]. On the other hand, it has been proposed that FES treatment might have positive effects on patients affected by spasticity [55]. Despite these considerations, our review underlined that no studies assessed spasticity combined with rehabilitation interventions to manage bone health in SCI patients and more in general the impact of spasticity on bone in these patients is far from being fully characterized.

On the other hand, muscle contraction has been targeted by different rehabilitative programs, including FES and vibration therapy, aiming at increasing mechanical stimuli on the bone and optimizing the protective role of muscle contraction on bone loss in these patients [31, 32, 35, 36, 37, 38].

In accordance with the International Osteoporosis Foundation, non-pharmacological intervention should be considered the first-line therapy in osteoporosis management [56]. However, in patients with a higher risk of fracture, a pharmacological approach is necessary to increase BMD and reduce the risk of fractures [57, 58]. In line with these findings, several studies included in the present review assessed the role of rehabilitation combined with both antiresorptive drugs (zoledronate [38]) and anabolic therapies (teriparatide [35] and testosterone [37]). The recent study by Dionyssiotis et al. [15] reported that level 1 supports intravenous zoledronic acid for preventing sublesional decline, while few studies are currently available about the effects of anabolic treatments for osteoporosis management in this specific population [15].

Lastly, a specific nutritional approach should be part of a comprehensive rehabilitation intervention, given the large consensus about the role of micronutrients in the management of bone health [57]. Despite this evidence, it was surprising to notice that just one study [35] assessed the effects of a combined intervention including rehabilitation and vitamin D $+$ calcium supplementation in SCI patients. However, the most recent guidelines suggested a vitamin D supplementation targeting $>$ 40 ng/ml vitamin D serum levels in patients with SCI in order to optimize extraskeletal effects [15]. Therefore, further studies are needed to characterize the positive effects of rehabilitation combined with specific nutritional intervention in these patients, aiming at optimizing the synergic effects of non-pharmacological approaches targeting the multimodal bone metabolism modifications induced by SCI [4, 8].

Interestingly, when consulting existing literature, previous reviews assessed the effects of rehabilitation interventions on bone health management of SCI patients. In particular, the review by Sutor et al. [10] assessed the effects of physical exercise and physical activity, providing interesting perspectives about the pathophysiology and molecular mechanisms underpinning physical activity positive effects on bone health of SCI patients. However, several studies with heterogenous designs were included and the authors did not apply a systematic approach. Similarly, the review by Chandrasekaran et al. [59] assessed the effects of FES in patients with SCI. Remarkably, the study highlighted the positive impact of FES in terms of muscle weight, muscle cross-sectional area, and physical functioning. However, the authors did not focus only on bone tissue modifications and the literature was assessed without a systematic approach.

Therefore, to the best of our knowledge, the present study is the first systematic review of RCTs assessing the effectiveness of different rehabilitation strategies to manage bone loss in SCI patients. Altogether, our data highlighted a large gap of knowledge about the potential role of rehabilitation interventions on bone health of SCI patients. On the other hand, a multitarget approach might further implement the synergisms between non-pharmacological and pharmacological interventions, with positive effects not only on bone health but also on functional outcomes of patients with SCI.

Despite these considerations, we are aware that this study has several limitations. In particular, the high heterogeneity of the study interventions limits to draw a quantitative synthesis of the study results. However, this is the first study addressing in a systematic way the effects of rehabilitation interventions on bone health management of SCI patients. In addition, this is the first systematic review assessing the effects of different rehabilitation modalities in this field, despite currently available literature does not allow to compare the effectiveness of different approaches. On the other hand, it should be noted that most of the studies included in the present systematic review showed some concerns in risk of bias assessment, highlighting the need for good quality studies assessing rehabilitation intervention in preventing bone loss of SCI patients.

Lastly, the etiology of SCI lesion, its level, and completeness might severely affect the rehabilitation outcomes, with potential sought of bias in the data synthesis. Therefore, further stratification might improve knowledge about the optimal therapeutic treatment tailored to patient’s characteristics.

5. Conclusions

Altogether, findings of this systematic review underlined that several rehabilitation interventions have been studied to improve bone health in patients with SCI, with promising results reported for physical exercise combined with FES and vibrations. However, several questions are still open about the synergisms of different rehabilitation strategies and the role of a specific rehabilitation approach combined with pharmacological management should still be characterized. Further good-quality studies are needed to improve knowledge about the optimal rehabilitation approach in SCI, improving not only functional outcomes but also bone health of these frail patients.

Ethical approval

Not applicable.

Funding

The authors report no funding.

Informed consent

Not applicable.

Author contributions

Conceptualization, L.L., A.d.S., and M.I.; methodology, L.L., A.d.S., and M.I.; investigation, L.L., A.F., and A.T.; writing – original draft preparation, L.L. and A.F.; writing – review and editing, A.d.S, and M.I.; figures, S.M.; visualization, C.C., S.M, and A.A.; supervision, A.d.S. and M.I. All authors have read and agreed to the published version of the manuscript.

Supplementary data

The supplementary files are available to download from http://dx.doi.org/10.3233//BMR-230006.

Footnotes

Acknowledgments

The authors want to acknowledge Enrico Cavallo for his support of this work.

Conflict of interest

The authors declare that they have no conflict of interest.

References

Battistuzzo

Zoghi

Galea

. Effects of training on upper limb function after cervical spinal cord injury: a systematic review. Clinical Rehabilitation. 2015; 29(1): 3-13.

Stampas

Tansey

. Spinal cord injury medicine and rehabilitation. Seminars in Neurology. 2014; 34(5): 524-33.

Invernizzi

de Sire

Renò

Cisari

Runza

Baricich

, et al. Spinal Cord Injury as a Model of Bone-Muscle Interactions: Therapeutic Implications From in vitro and in vivo Studies. Front Endocrinol (Lausanne). 2020; 11: 204.

Invernizzi

de Sire

Fusco

. Rethinking the clinical management of volumetric muscle loss in patients with spinal cord injury: Synergy among nutritional supplementation, pharmacotherapy, and rehabilitation. Curr Opin Pharmacol. 2021; 57: 132-9.

Pickett

Campos-Benitez

Keller

Duggal

. Epidemiology of traumatic spinal cord injury in Canada. Spine. 2006; 31(7): 799-805.

Winslow

Rozovsky

. Effect of spinal cord injury on the respiratory system. American Journal of Physical Medicine & Rehabilitation. 2003; 82(10): 803-14.

DeVivo

Krause

Lammertse

. Recent trends in mortality and causes of death among persons with spinal cord injury. Archives of Physical Medicine and Rehabilitation. 1999; 80(11): 1411-9.

Invernizzi

, Sire

Carda

Venetis

Renò

Cisari

, et al. Bone Muscle Crosstalk in Spinal Cord Injuries: Pathophysiology and Implications for Patients’ Quality of Life. Current Osteoporosis Reports. 2020; 18(4): 422-31.

Zehnder

Lüthi

Michel

Knecht

Perrelet

Neto

, et al. Long-term changes in bone metabolism, bone mineral density, quantitative ultrasound parameters, and fracture incidence after spinal cord injury: a cross-sectional observational study in 100 paraplegic men. Osteoporosis international: a journal established as result of cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA. 2004; 15(3): 180-9.

10.

Sutor

Kura

Mattingly

Otzel

Yarrow

. The Effects of Exercise and Activity-Based Physical Therapy on Bone after Spinal Cord Injury. International Journal of Molecular Sciences. 2022; 23(2): 608.

11.

Eser

Frotzler

Zehnder

Wick

Knecht

Denoth

, et al. Relationship between the duration of paralysis and bone structure: a pQCT study of spinal cord injured individuals. Bone. 2004; 34(5): 869-80.

12.

Frotzler

Berger

Knecht

Eser

. Bone steady-state is established at reduced bone strength after spinal cord injury: a longitudinal study using peripheral quantitative computed tomography (pQCT). Bone. 2008; 43(3): 549-55.

13.

Edwards

Simonian

Troy

Schnitzer

. Reduction in torsional stiffness and strength at the proximal tibia as a function of time since spinal cord injury. Journal of Bone and Mineral Research. 2015; 30(8): 1422-30.

14.

Frisbie

. Fractures after myelopathy: the risk quantified. The Journal of Spinal Cord Medicine. 1997; 20(1): 66-9.

15.

Dionyssiotis

Kalke

Frotzler

Moosburger

Trovas

Kaskani

, et al. S1 Guidelines on Bone Impairment in Spinal Cord Injury. J Clin Densitom. 2021; 24(3): 490-501.

16.

Morone

, Sire

, Cinnera

Paci

Perrero

Invernizzi

, et al. Upper Limb Robotic Rehabilitation for Patients with Cervical Spinal Cord Injury: A Comprehensive Review. Brain Sciences. 2021; 11(12): 1630.

17.

Invernizzi

Carda

Rizzi

Grana

Squarzanti

Cisari

, et al. Evaluation of serum myostatin and sclerostin levels in chronic spinal cord injured patients. Spinal Cord. 2015; 53(8): 615-20.

18.

Solomon

Dhakal

Halpin

Hariharan

O’Connor

Allsop

, et al. Telerehabilitation for individuals with spinal cord injury in low-and middle-income countries: a systematic review of the literature. Spinal Cord. 2022; 60(5): 395-403.

19.

Schultz

Mona

Cameron

. Mental Health and Spinal Cord Injury: Clinical Considerations for Rehabilitation Providers. Current Physical Medicine and Rehabilitation Reports. 2022.

20.

de Melo-Neto

de Campos Gomes

de Morais

Tognola

. Spinal cord injury in elderly patients admitted to a tertiary hospital. J Back Musculoskelet Rehabil. 2017; 30(4): 929-36.

21.

de Sire

Lippi

Aprile

Calafiore

Folli

D’Abrosca

, et al. Pharmacological, Nutritional, and Rehabilitative Interventions to Improve the Complex Management of Osteoporosis in Patients with Chronic Obstructive Pulmonary Disease: A Narrative Review. Journal of Personalized Medicine. 2022; 12(10): 1626.

22.

Yeung

SSY

Reijnierse

Pham

Trappenburg

Lim

Meskers

CGM

, et al. Sarcopenia and its association with falls and fractures in older adults: A systematic review and meta-analysis. Journal of Cachexia, Sarcopenia and Muscle. 2019; 10(3): 485-500.

23.

Lippi

de Sire

D’Abrosca

Polla

Marotta

Castello

, et al. Efficacy of Physiotherapy Interventions on Weaning in Mechanically Ventilated Critically Ill Patients: A Systematic Review and Meta-Analysis. Front Med (Lausanne). 2022; 9: 889218.

24.

Bernardes Neto

SCG

Torres-Castro

Lima

Resqueti

Fregonezi

GAF

. Weaning from mechanical ventilation in people with neuromuscular disease: a systematic review. BMJ Open. 2021; 11(9): e047449.

25.

Yuan

Chao

Beck

Sinderby

Xie

, et al. Neurally adjusted ventilatory assist as a weaning mode for adults with invasive mechanical ventilation: a systematic review and meta-analysis. Critical Care. 2021; 25(1).

26.

Leboff

Greenspan

Insogna

Lewiecki

Saag

Singer

, et al. The clinician’s guide to prevention and treatment of osteoporosis. Osteoporosis International. 2022; 33(10): 2049-102.

27.

Page

McKenzie

Bossuyt

Boutron

Hoffmann

Mulrow

, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021; 372: n71.

28.

Huang

Lin

Demner-Fushman

. Evaluation of PICO as a knowledge representation for clinical questions. AMIA Annu Symp Proc. 2006; 359-63.

29.

Cashin

McAuley

. Clinimetrics: Physiotherapy Evidence Database (PEDro) Scale. J Physiother. 2020; 66(1): 59.

30.

Sterne

JAC

Savović

Page

Elbers

Blencowe

Boutron

, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ. 2019; 366: l4898.

31.

Afshari

Ozturk

Yates

Picard

Taylor

. Effect of hybrid FES exercise on body composition during the sub-acute phase of spinal cord injury. Plos One. 2022; 17(1).

32.

Arija-Blázquez

Ceruelo-Abajo

Díaz-Merino

Godino-Durán

Martínez-Dhier

Martin

, et al. Effects of electromyostimulation on muscle and bone in men with acute traumatic spinal cord injury: A randomized clinical trial. J Spinal Cord Med. 2014; 37(3): 299-309.

33.

Ben

Harvey

Denis

Glinsky

Goehl

Chee

, et al. Does 12 weeks of regular standing prevent loss of ankle mobility and bone mineral density in people with recent spinal cord injuries? Australian Journal of Physiotherapy. 2005; 51(4): 251-6.

34.

Chen

Jin

Yao

Wang

Ouyang

, et al. Influence of the intelligent standing mobile robot on lower extremity physiology of complete spinal cord injury patients. Medicine in Novel Technology and Devices 2020 Sep; 7: 100045.

35.

Edwards

Simonian

Haider

Anschel

Chen

Gordon

, et al. Effects of Teriparatide and Vibration on Bone Mass and Bone Strength in People with Bone Loss and Spinal Cord Injury: A Randomized, Controlled Trial. J Bone Miner Res. 2018; 33(10): 1729-40.

36.

Groah

Lichy

Libin

Ljungberg

. Intensive electrical stimulation attenuates femoral bone loss in acute spinal cord injury. PM R. 2010; 2(12): 1080-7.

37.

Holman

Chang

Ghatas

Saha

Zhang

Khan

, et al. Bone and non-contractile soft tissue changes following open kinetic chain resistance training and testosterone treatment in spinal cord injury: an exploratory study. Osteoporos Int. 2021; 32(7): 1321-32.

38.

Morse

Troy

Fang

Nguyen

Battaglino

Goldstein

, et al. Combination Therapy With Zoledronic Acid and FES-Row Training Mitigates Bone Loss in Paralyzed Legs: Results of a Randomized Comparative Clinical Trial. JBMR Plus. 2019; 3(5).

39.

Shackleton

Evans

West

Derman

Albertus

. Robotic Walking to Mitigate Bone Mineral Density Decline and Adverse Body Composition in Individuals With Incomplete Spinal Cord Injury A Pilot Randomized Clinical Trial. American Journal of Physical Medicine & Rehabilitation. 2022; 101(10): 931-6.

40.

Warden

Bennell

Matthews

Brown

McMeeken

Wark

. Efficacy of low-intensity pulsed ultrasound in the prevention of osteoporosis following spinal cord injury. Bone. 2001; 29(5): 431-6.

41.

Xiang

Pan

Tan

, et al. Effect of electroacupuncture combined with osteoporosis treatment apparatus on bone metabolism in patients with spinal cord injury. World Journal of Acupuncture-Moxibustion. 2020; 30(4): 251-5.

42.

Lang

Gong

Wang

Dong

. Pulse Electromagnetic Field for Treating Postmenopausal Osteoporosis: A Systematic Review and Meta-Analysis of Randomized Controlled Trials. Bioelectromagnetics. 2022; 43(6): 381-93.

43.

Zhu

Wang

Huang

Song

Gan

, et al. Pulsed Electromagnetic Fields May Be Effective for the Management of Primary Osteoporosis: A Systematic Review and Meta-Analysis. IEEE Transactions on Neural Systems and Rehabilitation Engineering. 2022; 30: 321-8.

44.

de Oliveira

RDJ

de Oliveira

Santos-Filho

Sá-Caputo

Bernardo-Filho

. Effectiveness of whole-body vibration on bone mineral density in postmenopausal women: a systematic review and meta-analysis of randomized controlled trials. Osteoporosis international: a journal established as result of cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA. 2022.

45.

Moayyeri

. The association between physical activity and osteoporotic fractures: a review of the evidence and implications for future research. Ann Epidemiol. 2008; 18(11): 827-35.

46.

de Kam

Smulders

Weerdesteyn

Smits-Engelsman

. Exercise interventions to reduce fall-related fractures and their risk factors in individuals with low bone density: a systematic review of randomized controlled trials. Osteoporosis International: A Journal Established as Result of Cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA. 2009; 20(12): 2111-25.

47.

Suominen

. Muscle training for bone strength. Aging Clin Exp Res. 2006; 18(2): 85-93.

48.

Basso

Hansen

. Biological basis of exercise-based treatments: spinal cord injury. PM&R. 2011; 3(6): S73-S7.

49.

Neefkes-Zonneveld

Bakkum

Bishop

van Tulder

Janssen

. Effect of long-term physical activity and acute exercise on markers of systemic inflammation in persons with chronic spinal cord injury: a systematic review. Archives of Physical Medicine and Rehabilitation. 2015; 96(1): 30-42.

50.

Seibel

. Biochemical markers of bone turnover: part I: biochemistry and variability. Clin Biochem Rev. 2005; 26(4): 97-122.

51.

Frost

. A 2003 update of bone physiology and Wolff’s Law for clinicians. Angle Orthod. 2004; 74(1): 3-15.

52.

Garland

Stewart

Adkins

Rosen

Liotta

, et al. Osteoporosis after spinal cord injury. Journal of Orthopaedic Research. 1992; 10(3): 371-8.

53.

Eser

Frotzler

Zehnder

Schiessl

Denoth

. Assessment of anthropometric, systemic, and lifestyle factors influencing bone status in the legs of spinal cord injured individuals. Osteoporosis International. 2005; 16(1): 26-34.

54.

Frost

. From Wolff’s law to the Utah paradigm: insights about bone physiology and its clinical applications. The Anatomical Record: An Official Publication of the American Association of Anatomists. 2001; 262(4): 398-419.

55.

Sivaramakrishnan

Solomon

Manikandan

. Comparison of transcutaneous electrical nerve stimulation (TENS) and functional electrical stimulation (FES) for spasticity in spinal cord injury-A pilot randomized cross-over trial. The Journal of Spinal Cord Medicine. 2018; 41(4): 397-406.

56.

International Osteoporosis Foundation (IOF) exercise recommendations. iof bone health; [internet]. available from: https://www.iofbonehealth.org/exercise-recommendations.

57.

Kanis

Cooper

Rizzoli

Reginster

J-Y

. Review of the guideline of the American College of Physicians on the treatment of osteoporosis. Osteoporosis International. 2018; 29(7): 1505-10.

58.

Rossini

Adami

Bertoldo

Diacinti

Gatti

Giannini

, et al. Guidelines for the diagnosis, prevention and management of osteoporosis. Reumatismo. 2016; 68(1): 1-39.

59.

Chandrasekaran

Davis

Bersch

Goldberg

Gorgey

. Electrical stimulation and denervated muscles after spinal cord injury. Neural Regen Res. 2020; 15(8): 1397-407.

	1. Eligibility criteria	2. Random allocation	3. Concealed allocation	4. Baseline comparability	5. Blind subjects	6. Blind therapists	7. Blind assessors	8. Adequate follow-up	9. Intention- to-treat analysis	10. Between- group comparisons	11. Point estimates and variability	Total
Afshari et al. 2022	1	1	0	0	0	0	0	0	0	1	1	3
Arija-Blázquez et al. 2014	1	1	0	1	1	0	0	1	0	1	1	6
Ben et al. 2005	1	1	1	1	0	0	1	1	1	1	1	8
Chen et al. 2020	1	1	0	1	0	0	1	1	0	1	1	6
Edwards et al. 2018	1	1	1	1	1	1	0	1	0	1	1	8
Groah et al. 2010	0	1	0	1	0	0	1	0	0	1	1	5
Holman et al. 2020	1	1	0	1	0	0	0	1	1	1	1	6
Morse et al. 2019	1	1	0	0	1	0	0	0	0	1	0	3
Shackleton et al. 2022	0	1	0	1	0	0	0	1	0	1	1	5
Warden et al. 2001	1	1	0	1	1	0	1	0	0	1	1	6
Xu et al. 2020	1	1	0	1	0	0	0	1	0	1	1	5

The impact of rehabilitation in bone loss management of patients with spinal cord injury: A systematic review

Abstract

BACKGROUND:

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSION:

Keywords

1. Introduction

2. Methods

2.1 Registration

2.2 Search strategy

2.3 Selection criteria

2.5 Quality assessment and risk of bias

3. Results

Table 1 Literature search strategy on different databases

Table 2 Main characteristics of the studies included in the systematic review

4. Discussion

5. Conclusions

Ethical approval

Funding

Informed consent

Author contributions

Supplementary data

Footnotes

Acknowledgments

Conflict of interest

References

Table 1
Literature search strategy on different databases

Table 2
Main characteristics of the studies included in the systematic review