A Synthesis of Best Evidence for the Restoration of Upper-Extremity Function in People with Tetraplegia

ABSTRACT

Purpose: Because upper-limb function represents overall function for individuals with tetraplegia, the restoration of upper-extremity function is exceedingly important for this population. The purpose of this review was to identify interventions that optimize upper-limb function after tetraplegia based on best available evidence.

Methods: A search of MEDLINE, AMED, and PubMed with the search terms “hand function AND tetraplegia” and “upper limb function AND tetraplegia” found 384 articles. After elimination of duplicates and review of titles and abstracts, 43 studies were found to be applicable. Study quality of all applicable studies was assessed with a modified version of the Scottish Intercollegiate Guidelines Network for Cohort Studies methodology.

Results: The applicable studies were organized into three categories: conventional therapies (CT), electrical stimulation therapies (ES), and surgical interventions (SI). The proportion of papers in each category that presented with sufficient methodological quality to contribute to best evidence was as follows: CT: 0/2; ES: 10/21; SI: 6/20.

Conclusions: ES therapies are beneficial as assistive technologies and as therapeutic intervention in the subacute phase of recovery. SIs are suitable for individuals who meet very specific criteria for tendon-transfer surgery. Further clinical trials are warranted for ES and SI therapies to substantiate prescription of therapeutics.

For individuals with tetraplegia, the upper limbs not only fulfil the same functions for which hands and arms are normally used but also substitute for other parts of the body that may no longer have even partial function (e.g., the lower-limb function of walking is replaced by the use of upper limbs for wheelchair propulsion). Improved upper-limb function after tetraplegia is one of the most significant factors in improving quality of life, according to individuals with tetraplegia.^1,2 Therefore, the more normal the upper-limb function recovered after tetraplegia, the more functional the individual.

Because of the importance of upper-extremity function for individuals with tetraplegia, restoring arm and hand function has been a major focus of their rehabilitation. The methods developed to restore upper-limb function vary in type depending on neurological status, determined by the level of tetraplegia. Tetraplegia can present as many different degrees of neurological deficit, because of the anatomical structure of the spinal cord. The objective of this mixed-methods review was to synthesize the best current evidence available to support interventions targeted toward restoring upper-limb function after tetraplegia and to provide a framework for clinical application. This review differs from chapter 6 of the Spinal Cord Injury Rehabilitation Evidence (SCIRE) 2008 review³ in that the literature is presented to guide clinical decision making for the rehabilitation professional based on the presentation of tetraplegia and the current knowledge available on the recovery profile of spinal-cord injury (SCI).

Tetraplegia

Traumatic SCI includes injuries to the spine and spinal cord, which can occur anywhere along the spinal column (cervical, thoracic, and lumbar) as a result of trauma. Injuries to the thoracic and lumbar spine result in paraplegia, while injuries to the cervical spine result in tetraplegia. The significant difference between these two general classes of injury is that an individual with paraplegia would retain normal upper-limb function, whereas an individual with tetraplegia loses not only the function of the lower body but also function of the upper limbs, the use of which is vital in maintaining independence after SCI. The degree of upper-limb impairment varies depending on the level of injury (see below); an injury to higher levels of the cervical spine will cause greater impairment in the upper limbs, and the degree of impairment is reduced as the level of injury moves from the occiput toward the T1 vertebra.

SCI is defined by the neurological level of injury, which is usually defined by the most caudal level at which sensory and motor function are fully intact according to the International Standards of Neurological Classification of SCI (ISCSCI). Classification of SCI is defined by the ASIA Impairment Scale (AIS) and indicates whether the injury is complete or incomplete. Complete SCI occurs when there is no motor or sensory function preserved in the sacral segments S4–S5 (A); there are three classes of incomplete SCI. An incomplete injury occurs when there is sensory but not motor function preserved below the neurological level and includes the sacral segments S4–S5 (B); when motor function is preserved below the neurological level, and more than half of the key muscles below the neurological level have a muscle grade less than 3 (C); or when motor function is preserved below the neurological level, and at least half of the key muscles below the neurological level have a muscle grade of 3 or more (D).⁴ Based on an international review of SCI epidemiology, SCI is complete in 40% of patients and incomplete in 60%.⁵ Tetraplegia represents approximately two-thirds of the total SCI population.⁶ Within tetraplegia there are categories based on classification (complete and incomplete); the complete class requires further sub-grouping because of differences in sensory and motor preservation according to level. This review considers the neurological state of the tetraplegic sub-groups. A key element that emerges from the literature reviewed is that there is a specific application of interventions to classification and sub-groups within tetraplegia. Figure 1 defines the tetraplegic sub-groups according to neurological integrity.

Figure 1.

Sub-groups of the tetraplegic population, based on nature of injury and sensorimotor capacity.

Traditionally, individuals with complete SCI do not regain any motor or sensory function below the neurological level of injury, whereas individuals with incomplete injuries may spontaneously recover sensory and motor deficits. However, the degree of recovery and the way in which the recovery influences function vary significantly from individual to individual and are not always predictable. Thus, two elements contribute to the heterogeneity of the tetraplegic population: the first is the difference in upper-limb intactness secondary to the neurological level of injury, and the second is the varying degree of recovery. The heterogeneity of the tetraplegic population creates many complexities in conducting research for effective therapeutic interventions to restore upper-limb function.

There are two predominant general rehabilitative approaches for the upper limb in tetraplegia. One is to compensate for functional loss by using the available capacity of the sensorimotor system. The second is directed toward restoring lost sensorimotor capacity. The available sensorimotor capacity, the nature of the injury, and the individual's response to treatment are factors that inform clinical decision making during therapeutic intervention. The research in this review was organized into three categories of interventions that optimize upper-limb function: conventional therapies (CT), upper-limb electrical stimulation therapies (ES), and upper-limb surgical interventions (SI).

Potential for Recovery

The term “recovery” refers to the changing state (improvement) of neurophysiology and/or functional capacity. After tetraplegia, even if an injury is complete, it is common to anticipate some type of recovery. In the case of complete injuries where no neurological change is expected, the anticipated recovery is functional; in the case of incomplete injuries, the anticipated recovery is both neurological and functional. Neurological recovery involves changes in impairment, while functional recovery relates to ability; depending on the severity of injury, either type of recovery, or both, can be expected.

Motor recovery of key muscles does occur in the upper limb after both complete and incomplete cervical SCI. Motor recovery can be predicted by the intact motor capacity in the acute phase after injury.^7–9 It has also been noted that gains in strength of elbow and shoulder muscles can continue to occur up to 15 months following discharge from rehabilitation.¹⁰ In one study, individuals who retained sensation to pinprick in a motor segment with grade 0 power showed an 85% chance of motor recovery to at least grade 3.¹¹ Some of the recovery in strength may be attributable to plasticity within the spinal cord; although the regeneration of nerve fibres is limited in the adult central nervous system, changes in function occurring for several years after injury may depend on the reorganization of circuits spared by the lesion^12,13 and on cortical reorganization.^14–18 Where cortical reorganization does occur after SCI, it is theorized that the sensorimotor cortex plays a critical role in the recovery of function. Therefore, the sensation spared at the time of injury, which is often followed by motor recovery,¹¹ may be partially related to changes that occur in the sensorimotor cortex and reorganization of tracts.

Individuals with tetraplegia improve their ability to function after a period of rehabilitative intervention¹⁹ and often continue to improve once discharged into the community. When improvement of function is documented, both complete and incomplete groups of SCI show improvement.²⁰ This would indicate that an individual unable to progress in neurological status can make progress in functional status through skill acquisition or adaptation.²¹ Specifically, shoulder strength and elbow extension strength show stronger associations with functional independence than do other muscle groups in the upper limb.^21–23

Understanding the neurophysiology,^13,24 structure, and secondary sequelae of SCI enhances the clinician's ability to prescribe interventions. Targeting rehabilitative approaches toward specific muscle groups that have a known potential to recover and have a greater impact on functional independence,^25,26 and targeting the timing of interventions to optimize recovery and function, is one strategy to optimize the efficacy of interventions. Current knowledge of the recovery profile after SCI assists in framing clinical decision making for rehabilitative approaches for tetraplegia by defining the timing and location of recovery.

METHODS AND MATERIALS

A literature search using MEDLINE, AMED, and PubMed was conducted with the search terms hand function, upper limb function, and tetraplegia, using the Boolean operator AND for searches in each database. Table 1 describes the search terms, the combinations used, and the results from each database. Articles were retained for review if the research was specific to humans and reported the efficacy of interventions to restore upper-limb function; the search was limited by the database search filters human, English, and adult research (19+ years). The search identified 384 articles; manual review of titles reduced this number to 84. The articles removed were reviews, studies unrelated to the research question, research related to measurement development, concepts in physiology, and duplicate articles. The remaining 84 articles were reviewed by abstract, and articles were removed that did not describe research studies. The full text of the remaining 72 articles was reviewed by the reviewer (SKR) for final eligibility. Seventeen articles did not specifically report on interventions to restore upper-limb function, and these articles (which related to physiology exploration, cortical changes, and cardiovascular fitness) were excluded from the review; a further 12 articles provided evidence of recovery of motor strength and function post injury but were not related to interventions to restore upper-limb function. The 43 remaining articles were studies reporting on the results of interventions applied to improve upper-limb function for individuals with tetraplegia.

Table 1.

Search Strategy and Results

Database	Terms*	Results
MEDLINE	hand function AND tetraplegia	102
	upper limb function AND tetraplegia	52
AMED	hand function AND tetraplegia	40
	upper limb function AND tetraplegia	0
PubMed	hand function AND tetraplegia	117
	upper limb function AND tetraplegia	73

^*Each term was searched separately, then in combination with other terms, using the Boolean operator AND, as shown in column 2.

Review Process

The incidence of SCI is low (worldwide incidence of approximately 22 persons per million translates to 130,000 new SCIs each year⁶) relative to other pathologies such as breast cancer or stroke. As a result, research on this population is challenging. Researchers are often limited in the quality of research they are able to produce because of small sample sizes and the lack of opportunity to randomize subjects. Therefore, the literature on SCI cannot be reviewed using many of the conventional quality assessments. Five methodologies for quality assessment of literature were considered: Downs and Black (1998);²⁷ Jadad and colleagues (1996);²⁸ Verhagen and colleagues (1998);²⁹ the Physiotherapy Evidence Database (PEDro);³⁰ and the Scottish Inter-Collegiate Guidelines Network (SIGN).³¹ SIGN has established six types of quality assessments for reviewing literature; for this review, the type specific to cohort studies³¹ was chosen because it does not consider randomization of groups to be a dominant factor in the strength of study design. The SIGN checklist for cohort studies was further modified to be more specific to the assessment of SCI literature. No established psychometric properties were available for this methodology. The original SIGN methodology consists of 17 questions. Three questions related to subject selection and two questions related to outcomes and blinding were eliminated because of their poor applicability to the SCI literature. The SIGN checklist for cohort studies rates each question on a nominal six-point scale; this was modified to a dichotomous score (0 or 1), designating whether or not a specific criterion was adequately incorporated, to produce a numeric score out of 12. (see Box 1 for details of the modifications made to the SIGN methodology.) A score of 8/12 was chosen (admittedly arbitrarily) as the cut-off score to determine whether an article could be considered to provide relatively robust findings. A score of 8 would indicate that the article met at least two-thirds of the required criteria, as defined by the quality-assessment tool. Although a score of 8 does not define which qualities are present or absent, an article meeting at least two-thirds of the required criteria could be considered to present a higher grade of research than one with a score below 8, without eliminating the entire pool of literature.

Box 1. Scottish Intercollegiate Guidelines Network (SIGN) Checklist 3 and Modifications^*.

Checklist Item		Original Scoring*	Dichotomous Scoring
1	The study addresses an appropriate and clearly focused question.	6-point scale	Yes=1 No=0
Selection of Subjects
2	The two groups being studied are selected from source populations that are comparable in all respects other than the factor under investigation.†	6-point scale	—
3	The study indicates how many of the people asked to take part did so, in each of the groups being studied.	6-point scale	Yes=1 No=0
4	The likelihood that some eligible subjects might have the outcome at the time of enrolment is assessed and taken into account in the analysis.†	6-point scale	—
5	What percent (Did any) of individuals or clusters recruited into each arm of the study dropped out before the study was completed?	%	Yes=1 No=0
6	Comparison is made between full participants and those lost to follow up, by exposure status.†	6-point scale	—
Assessment
7	The outcomes are clearly defined.	6-point scale	Yes=1 No=0
8	The assessment of outcome is made blind to exposure status.†	6-point scale	—
9	Where blinding was not possible, there is some recognition that knowledge of exposure status could have influenced the assessment of outcome.†	6-point scale	—
10	The measure of assessment of exposure is reliable.	6-point scale	Yes=1 No=0
11	Evidence from other sources is used to demonstrate that the method of outcome assessment is valid and reliable.	6-point scale	Yes=1 No=0
12	Exposure level or prognostic factor is assessed more than once.	6-point scale	Yes=1 No=0
Confounding
13	The main potential confounders are identified and taken into account in the design and analysis.	6-point scale	Yes=1 No=0
Statistical Analysis
14	Have confidence intervals been provided?	6-point scale	Yes=1 No=0
Overall Rating
15	How well was the study done to minimize the risk of bias or confounders, and to establish a causal relationship between exposure and effect?	Code ++, +, or −	Yes=1 No=0
16	Taking into account clinical considerations, your evaluation of the methodology used, and the statistical power of the study, are you certain that the overall effect is due to the exposure being investigated?	Yes / No	Yes=1 No=0
17	Are the results of this study directly applicable to the patient group targeted in this guideline?	Yes / No	Yes=1 No=0

^*Six-level scale: 6=well covered; 5=adequately addressed; 4=poorly addressed; 3=not addressed; 2=not reported; 1=not applicable.

^†Removed from the original checklist.

All papers were reviewed using the modified version of the SIGN checklist for cohort studies (see Box 1)³¹ and were organized into three categories (see Figure 2): conventional therapies (CT) (2), electrical stimulation therapies (ES) (21), and upper-limb surgical interventions (SI) (20).

Figure 2.

Search strategy and selection of research for review.

RESULTS

Figure 2 presents the flowchart with results of the literature search. Of the 43 interventional studies reviewed, 2 were categorized as conventional therapies; neither of these articles presented with a “high grade” of research (≥8/12), and thus neither could be considered stronger evidence to support best practices. In the ES category, 10 of 21 studies presented a higher grade of research (these studies are summarized in Table 2); 6 of the 20 studies categorized as SI provided a higher grade of research (summarized in Table 3).

DISCUSSION

The intervention literature is very specific to the sub-groups within tetraplegia (see Figure 1). Little evidence exists for ES and SI therapies in the C4 complete group, where optimizing upper-limb function is not feasible and therapy is therefore directed toward protecting the flaccid upper limbs³² by splinting, positioning, range-of-motion exercises, and, most recently, functional electrical stimulation (FES) coupled with arm ergometry to prevent contractures, prevent upper-limb muscle atrophy, and improve cardiovascular circulation.³³ The highest prevalence of tetraplegia is at C5/C6, and the majority of evidence available for interventions that optimize upper-limb function is specifically for the C5/C6/C7 complete group. Individuals in the C8/T1 complete sub-group of tetraplegia generally retain enough use of and function in the upper limbs that methods to enhance the function they already have are usually employed. Strengthening, massed practice, and functional education are emphasized. For the population with incomplete SCI, the potential to enhance recovery also increases. Methods targeted to enhance or influence recovery of neurological function, as opposed to the ability to simply perform a task, become more appropriate. The incomplete sub-group of tetraplegia can benefit from any type of upper-limb therapy, depending on the individual's neurological state (recovering or stable).

Clinical Relevance of Selected Literature

Conventional Therapies

One method of conservative management is to position the hand so that the finger flexors will tighten and wrist extension will enhance a functional tenodesis grasp. Research on methods of positioning to encourage this development, known as the “functional hand,” has shown that splinting and positioning the hands at night has only a 50% rate of success in establishing a functional hand.³⁴ These findings are questionable, however. Doll and colleagues grouped all levels of tetraplegia in the same group and did not consider the neurological contributions of spasticity, sensation, and different motor innervation.³⁴ Harvey and colleagues also studied nightly splinting of the flexor pollicis longus muscle in a shortened position to reduce extensibility over a 3-month period, and found that splinting did not reduce extensibility of the flexor pollicis longus.³⁵ Thus there is little evidence to support nocturnal splinting as a method of tightening muscles in the hand and forearm; the factors related to level of lesion were not addressed in these studies. Although developing a tenodesis grasp remains common practice for tetraplegia, supporting evidence of effectiveness is lacking. The Guidelines for Management of the Upper Limb in Tetraplegia of the Paralyzed Veterans of America (PVA) suggest the use of splinting to protect joints, prevent pain syndromes, maintain residual function in the upper limb, and prevent secondary complications of the denervated upper limb.³²

The absence of research evidence supporting conventional therapies does not discredit the current practices of rehabilitation professionals; evidence to support interventions that are currently in use has been established in other neurological populations, such as constraint-induced movement therapy and massed-practice therapy to enhance normal upper-limb function post injury.^36,37 It does, however, emphasize the gap in rehabilitation research with the SCI population, where current and accepted neurological rehabilitation practices have not been studied as rigorously as newer interventions (e.g., somatosensory stimulation [SS] and functional electrical stimulation [FES]).

Electrical Stimulation Therapies (ES)

This category includes functional electrical stimulation (FES) and somatosensory stimulation (SS) as therapeutic interventions and assistive technology devices such as the Freehand System (NeuroControl, Cleveland, OH),^38–40 the Bionic Glove (Neurokinetics, Edmonton, AB),⁴¹ and the NESS Handmaster (Bioness, Valencia, CA).⁴² Approximately half of the available literature in the field of ES relates to the development of devices and therapies for humans. Although half of the articles retrieved were of insufficient methodological quality to meet the threshold requirements, these studies have established significant concepts, key elements for future work, and the initial development of products:

• FES can be administered to strengthen muscles and assist voluntary muscle control.⁴³

• Appropriate screening of study participants is required, because not all patients can benefit from FES systems; clinicians should assess neurological status and muscle response to FES prior to administering interventions.⁴⁴

• The Freehand System, a surgically implanted neuro-prosthetic system that provides stimulation to the muscles of the forearm to enhance grasping capabilities, has been piloted.^38–40

• The NESS Handmaster device, an external FES unit much like a splint that is placed over the muscles of the forearm, has been piloted.⁴²

• The Bionic Glove, an FES device that is placed on the hand/wrist as a splint and provides muscle activation when “on,” assisting with grasping and hand opening, has been piloted; the Bionic Glove is intended for individuals with tetraplegia who have active wrist extension.⁴¹

These articles reporting preliminary work on ES are the precursors to the subsequent, more rigorous trials presented in Table 2. Development of control strategies for FES is one area that remains at the “bench” stage of FES research. ES literature determined to be of sufficient methodological quality and to provide evidence supporting rehabilitative approaches for the enhancement of function in the upper limb are reported in Table 2.

All the studies listed in Table 2 provide evidence for the efficacy of ES, but the studies measure different outcomes, although strength (increase) as one indicator of efficacy and function/activities of daily living (ADL) as a second indicator are consistent across studies. ES can be used in the form of assistive technologies or applied as therapeutic interventions to influence the integrity of the sensory and motor systems.

Assistive technologies such as the NESS Handmaster, the Freehand system, and the Bionic Glove are beneficial for individuals in the C5/C6/C7 complete group, particularly once they are neurologically stable. These technologies are not therapeutic interventions but, rather, assistive devices that improve hand function when in place and active.^45–48 Research on these three devices shows that individuals can complete a higher number of ADL tasks independently with the device in situ and on. The ability to grasp and release, the strength of lateral key and palmar grasp, and the ability to manipulate objects in the hand improve, while the time required to perform a task decreases. All of these factors lead to improved function of the upper limb while the device is on.

Both SS and FES can be applied as interventions to change the state of the muscle. There is significant evidence to support SS as an adjuvant to rehabilitation therapies, particularly massed practice (MP). SS in association with a comprehensive, functionally oriented rehabilitation programme results in improved sensation, motor function, and functional ability and produces cortical changes related to the trained function.^49–51 FES as an intervention during the subacute phase of recovery has proved to be more effective than conventional rehabilitation alone when administered as a therapy in conjunction with functional tasks.^52,53 However, FES applied as stimulation outside of a functional task in the subacute phase shows no superiority over conventional rehabilitation⁵⁴ Popovic and colleagues,⁵² who applied the FES within a functional paradigm, suggested that FES embedded in a functional paradigm accounted for the difference between control and treatment groups, but they were not able to demonstrate a statistically significant difference. Because both Popovic and colleagues⁵² and Kohlmeyer and colleagues⁵⁴ included individuals whose neurological status was changing during the time of the intervention, further studies with larger control and treatment groups are required to prove the effect of the treatment. Kohlmeyer and colleagues applied FES, but not within a functional paradigm, and did not see any effects. The more recent SS studies done in the Field-Fote laboratory^49–51 (which have not been replicated in other labs) substantiate the notion that stimulation therapies applied in conjunction with functionally oriented interventions show the most efficacy. Additional support for a functional paradigm comes from the fact that neuromuscular stimulation coupled with arm ergometry also enhances triceps strength relative to exercise alone.⁵⁵

Thus, stimulation therapies are beneficial not only as assistive technologies for the neurologically stable individual but also as therapeutic interventions for the C5/C6/C7 complete sub-group during the rehabilitation phase of recovery (4 weeks to 18 months post injury).

As the incidence of incomplete SCI increases,⁵ the potential to enhance recovery also increases. The increased prevalence of incomplete injury is due in part to the superior acute management of SCI (EMS care, surgical decompression, and neuro-protection).^56–58 Methods targeted to enhance or influence recovery of neurological function, as opposed to just the ability to perform a task, become more appropriate. Targeting recovery with novel interventions is a new but very promising approach. As Table 2 shows, there is evidence that SS and MP are useful interventions to enhance strength, function, and sensation.^49–51 These interventions also have an effect on the cortical plasticity of the brain and spinal cord. FES therapies applied to the incomplete SCI sub-group can have a training effect for the development of voluntary-muscle strength and control. After a period of training with the FES unit, an individual has increased functional strength without the unit in place.⁵² These ES therapies are safe for humans, and the next steps should involve replication in other research settings and multi-centre randomized controlled trials.

Table 2.

Electrical Stimulation Literature of Sufficient Methodological Quality*

Author (date)	Modified SIGN score	Sample and characteristics	Objectives of study	Study design	Outcome measures	Effect and summary of study results
Beekhuizen and Field-Fote (2008)49	12	Chronic tetraplegia 24 (22 male, 2 female) 24 C5–C7 incomplete: 11 AIS–C, 13 AIS–D Age: 19–70 y	To establish the efficacy of SS and MP	Randomized cohort study—4 groups: 1. Conventional therapy (control) 2. SS 3. MP 4. SS+MP Intervention: 5×/wk for 3 wk	Measures completed pre and post intervention: JHFT WMFT Maximal pinch grip force SWM MEP strength in thenar muscle	Analysis: One-way ANCOVA (ANOVA for SWM) between pre- and post-test differences, then compared between-group differences Effect: MP+SS and SS groups had significant increases in JHFT (F=11.24, p<0.001), WMFT (F=12.52, p<0.001), SWM (F=4.69, p<0.01), and grip strength (F=9.51, p<0.001) in comparison to the control group. MP+SS and MP groups had significant increases in MEP output (F=19.06, p<0.01) relative to control group. Summary: MP+SS are beneficial rehabilitative tools in the restoration of force production and function. SS alone may be more beneficial in the acute phases of recovery.
Hoffman and Field-Fote (2007)50	10	Chronic traumatic tetraplegia 1 (1 male): C6 complete Age: 22 y	To establish the efficacy of SS and MP in increasing the cortical map	Single case study SS+MP Intervention: 5×/wk for 3 wk	Measures completed pre and post intervention: ISNCSCI upper- extremity motor score JHFT SWM Chedoke Hand Inventory Cortical mapping with TMS	Analysis: Differences in each measure noted Effect: 33% less time required to perform the JHFT; Chedoke Hand Inventory Score increased 19%; response to SWM improved by one level. The ability to perform functions post treatment that were not performed prior to treatment was also noted. There was no change in motor threshold for cortical representation, but the cortical map did increase in size. Summary: SS and MP may induce cortical reorganization that is associated with improvement in function.
Popovic (2006)51	10	Sub-acute traumatic tetraplegia 21 (21 male): 10 complete, 11 incomplete Age: 24–70 y	To determine the clinical efficacy of FES therapy in the rehabilitation of grasp function	Randomized controlled trial—4 groups: •Group 1: complete control •Group 2: incomplete control •Group 3: complete intervention •Group 4: incomplete intervention Intervention: 5×/wk for 12 wk	Measures completed pre and post intervention: TR-HFT SCIM FIM	Analysis: Wilcoxon signed rank test for statistically significant differences between comparison groups Effect: Intervention groups showed greater improvements after the intervention than control groups, but the differences were not statistically significant. Summary: FES therapy has the potential to be an effective treatment modality to restore grasp function in tetraplegia.
Beekhuizen (2005)52	10	Chronic traumatic tetraplegia 10 (9 male, 1 female): C5–C7 incomplete, 4 AIS-C, 6 AIS-D Age: 22–63 y	To establish the efficacy of SS and MP on the cortical map and function	Randomized controlled trial—2 groups: 1. MP+SS 2. MP Intervention: 5×/wk for 3 wk	Measures completed pre and post intervention: Pinch and grip strength JHFT WMFT Cortical mapping with TMS	Analysis: One-way ANCOVA between pre- and post-test differences, then compared between-group differences Effect: Pinch grip, WMFT, and JHFT show statistically significant differences in pre- and post-MP+SS groups (p<0.05). No significant difference in MEP amplitude before and after intervention was found. Summary: MP+SS is superior to MP alone and improves pinch strength and arm and hand function to a greater degree than MP alone.
Alon and McBride (2003)45	8	Chronic tetraplegia 7 (7 male): 7 complete C5 and C6 Age: 25–46 y	To test safety and efficacy of the NESS Handmaster	Longitudinal cohort study Same group compared before and after treatment Intervention: 3 wk of training with NESS	Measures completed at baseline, midpoint, and post treatment: Grip strength Fugl-Meyer hand subtest GRT Subjective test	Analysis: ANOVA for statistically significant differences between pre- and post-test Effect: 3 weeks of training in use of the NESS Handmaster elicited significant improvements in grip strength, ability to perform ADL, and grasp and release abilities as well as self- perceived improvement in function. Summary: Function can be optimized with use of the NESS Handmaster for individuals with C5 or C6 tetraplegia.
Taylor et al. (2002)46	11	Chronic traumatic tetraplegia 9 (8 male): 9 complete C5 and C6 Mean age: 38.4 y	To establish the functional impact of the Freehand implanted neuro-prosthetic	Longitudinal cohort study Same group compared before and after implant with 1-year follow-up	Measures completed pre and post intervention: GRT Grip strength ADLs Two-point discrimination	Analysis: Wilcoxon signed rank test for statistically significant differences between pre and post testing Effect: Statistically significant differences were noted in GRT score (average ↑ from 1.4 tasks pre intervention to 5.1 tasks post intervention). Grip strength mean difference (pre/post) for lateral key pinch=11.2N, palmar grasp= 9.5N, and five-finger pinch=10.4N. The number of ADLs completed increased by 3.8 new tasks. Summary: There is a significant impact on functional ability with the implant when on. Both independence and strength improve.
Peckam et al. (2001)47	11	Chronic traumatic tetraplegia 51 (42 male, 9 female): 51 complete C5 and C6 Age: 16–57 y (mean age 32 y)	To establish the efficacy of the Freehand implanted neuro-prosthetic	Longitudinal cohort study Same group compared before and after implant with 3-year follow-up	Measures completed pre and post intervention: GRT ADL Test Dynamometry	Analysis: Cohen's differences between proportions test (applied to proportions of sample) Effect: GRT: 49 participants gained 1 or more tasks; 37 gained 3 or more tasks. ADL Test: 21 participants gained 1 or more tasks; 20 gained 3 or more tasks (only 21 tested). Lateral key pinch: pre=0–1.6N, post= 9.4–15.3N. Palmar grasp: pre=0–1.5N, post=3.3–8.4N. Summary: Number of tasks performed improves and independence and strength increases when implant is on.
Popovic et al. (1999)48	10	Chronic traumatic tetraplegia 12 (12 male): C5 to C7 Age: 18–58 y	To establish the clinical benefits of using the Bionic Glove	Longitudinal cohort study Same group compared before and after implementation of prosthetic with 6-month follow-up	Measures completed pre intervention and 1, 3, and 6 months post intervention: FIM and QIF Upper Extremity Function Test ROM	Analysis: Descriptive statistical comparisons Effect: FIM and QIF scores remained similar before and after 6 months with prosthetic. Upper Extremity Function Test: time to complete tasks reduced by 6 sec, number of tasks completed increased by 1. ROM of fingers increased by 20%. Summary: The Bionic Glove improves independence and functions that are performed by the hand.
Needham- Shropshire et al. (1997)55	11	Chronic tetraplegia (post 1 year) 34 (31 male, 3 female) Age: 18–45 y (mean age 24 y)	To determine the efficacy of neuromuscular stimulation (NMS) coupled with arm ergometry	Randomized controlled trial—3 groups: Group 1: NMS and ergometry×8 wk Group 2: 4 wk NMS-assisted exercise with 4 wk voluntary arm-crank exercise Group 3: voluntary exercise for 8 wk	Triceps MMT	Analysis: ANOVA for statistically significant differences between groups Effect: Groups 1 and 2 showed a statistically significant difference in the number of triceps muscles improved relative to Group 3 at both 4 and 8 wk. Summary: NMS coupled with arm ergometry enhances voluntary triceps muscle strength relative to exercise alone.
Kohlmeyer et al. (1996)54	8	Acute traumatic tetraplegia 45 (40 male, 5 female): C4–5, C5–27, C6–12 Mean age 38.7 y	To evaluate the effectiveness of ES and biofeedback on recovery of tenodesis grasp during acute rehabilitation	Cohort study—4 groups: •Group 1: conventional therapy •Group 2: ES •Group 3: biofeedback, •Group 4: ES+biofeedback Intervention: 5–6 wk, 5×/wk	Measures completed pre and post intervention: MMT of biceps, anterior deltoid, and extensor carpi radialis longus Four self-feeding abilities (rated 0–2)	Analysis: Kruskall–Wallis one-way ANOVA and Spearman correlation coefficients for non-parametric data for between-group differences. Effect: All groups showed improvement in strength and function; however, all groups showed the same amount of improvement (no statistical or clinical difference between groups was noted). Summary: ES and biofeedback are not superior to conventional rehabilitation treatment.

Note: Table applies to C5/C6/C7 complete and incomplete sub-groups of tetraplegia in Figure 1.

AIS=ASIA Impairment Scale; ext=extension; FES=functional electrical stimulation; FIM=Functional Independence Measure; GRT=Grasp and Release Test; ISNCSCI=International Standards of Neurological Classification of Spinal Cord Injury; JHFT=Jebsen Hand Function Test; MEP=motor evoked potential; MMT=manual muscle testing; MP=massed practice; N=Newton of force; QIF=Quadriplegia Index of Function; ROM=range of motion; SCI=spinal-cord injury; SCIM=Spinal Cord Independence Measure; SS=somatosensory stimulation; SWM=Semmes Weinstein Monofilaments; TMS=transmagnetic stimulation; TR-HFT=Toronto Rehab-Hand Function Test; WMFT=Wolf Motor Hand Function Test.

Surgical Interventions for the Upper Limb in Tetraplegia

Surgical intervention for the upper limb in tetraplegia involves transferring the tendon of an active muscle to the insertion of an inactive muscle to reproduce lost movement at a specific joint. As Lamb and Chan explain, “to the patient who has nothing initially, a little becomes a lot.”^59(p.291) Surgical procedures to optimize upper-limb function focus on the mid-cervical group of tetraplegia, because certain components of upper-limb motor and sensory control are required for surgery to be successful.

Approximately two-thirds of the available literature in the field of SI is seminal work documenting the individual experiences of clinicians, with recommendations based on individual clinicians' career work and anecdotal experiences. Although these articles are not of sufficient methodological quality to be included in this review, foundations for the concepts to be studied have been established:

• Outcome and magnitude of functional gain after SI depends on (1) the level of spinal-cord lesion, (2) careful patient selection, (3) thoughtful application of the principles of tendon transfer, (4) absence of severe spasticity, (5) the surgeon's expertise, and (6) postoperative rehabilitation.^60,61

• Meticulous selection of patients for treatment of specific conditions is necessary and is a predictor of therapeutic outcome.⁶²

• Outcomes used to establish efficacy are in question; many studies report no efficacy of surgical interventions, and this result is often due to the choice of measures.⁶³

Early work in the field of SI has only established a surgical framework; more rigorous efficacy trials are required. Table 3 describes those articles that present with sufficient methodological quality and further substantiate the seminal work.

Table 3.

Surgical Intervention Literature with Sufficient Methodological Quality**

Author (date)	Modified SIGN score	Sample and characteristics	Objectives of study	Study design	Outcome measures	Effect and summary of study results
Laffont et al. (2007)64	9	Chronic traumatic tetraplegia 12 patients (10 male, 2 female), 17 hands: C6/C7 Age 17–51 y Procedures included deltoid to triceps transfer, ECRL to FDP, and BR to FDP	To quantify the effect of tendon transfers	Cross-sectional study—3 groups: Group A: no elbow ext with passive tenodesis Group B: elbow ext Group C: elbow ext with finger flexion/ext	Tetra Ball Test to examine grasp and release control of various-sized balls in neutral supination and pronation of forearm	Analysis: ANOVA for statistically significant differences between groups Effect: Failures for grasping decreased from groups A to C (p<0.0001); duration of movement decreased from A to C (p<0.0001) Summary: Group C had superior ability over Groups A and B in terms of grasp and release and quickness of functions.
Johanson et al. (2006)65	8	Chronic traumatic tetraplegia 10 patients (8 male, 2 female), 11 hands: C6/C7 Procedures included BR to FPL 10 y after surgery	BR activation during elbow flexion and thumb abduction after BR to FPL transfer	Cross-sectional study comparing transferred BR activity when used for elbow flexion and thumb abduction	Muscle activation with EMG	Analysis: Wilcoxon signed rank test for between-test conditions Effect: EMG of BR for thumb abduction was 34% and 55% (with elbow stabilized) of maximal BR force for elbow flexion, significantly less (p<0.05). Pinch force was 14N and 20N (with elbow stabilized). Summary: Postoperative difficulty exists in activating BR post transfer for thumb abduction.
Vastamäki (2006)66	9	Chronic traumatic tetraplegia 6 patients (6 male), 10 hands: C5/C6 Mean age 55 y Procedures: deltoid to triceps transfer, BR to FPL	To investigate the long-term (24 y) and short-term (3 y) results after surgery	Longitudinal cohort study—1 group; comparisons at 3 y and 24 y after surgery	•LKP strength and elbow-extension strength.	Analysis: Descriptive statistical comparisons Effect: LKP and elbow ext strength decline at 3 years by 21% and 16% respectively. Functional status is maintained or slightly improved. Subjective information showed half the sample felt improvement and all participants felt they were stable. Summary: Early post-surgical gains are maintained to long-term status despite a loss of strength.
Rothwell et al. (2003)67	10	Chronic traumatic tetraplegia 24 patients, 48 hands: C5/C6 Mean age 42 y Mean time post injury 20 y Procedures included BR to FPL, ECRL to FDP, BR to FDP, PT to FPL	To determine the long-term benefits of reconstructive hand surgery	Longitudinal cohort study Same-group comparison shortly after surgery vs. 12–18 years after surgery	Lamb and Chan Questionnaire: perceived improvement of independence QIF LKP and cylindrical grasp strength	Analysis: Descriptive statistical comparisons Effect: Cylindrical grip: slightly improved (not significant); LKP: slightly diminished QIF compared to functional status by memory. Summary: Individuals have objective long-term improvement in hook grip. Self-perception of function is questionable due to different measures used at baseline and follow-up.
Meiners et al. (2003)68	8	Chronic traumatic tetraplegia 24 patients (21 male, 3 female), 25 hands: C5/C6 Mean age 37.5 y Procedures included BR to FPL, FCR to FPL or EPL, ECRL to FDP, BR to FDP	To examine the efficacy of wrist extension, lateral key pinch, and cylindrical grasp	Longitudinal cohort study Same group before and 9–51 months after tendon transfer	Lamb and Chan Questionnaire: perceived improvement of independence LKP and cylindrical grasp strength	Analysis: Mean of difference between pre and post measures Effect: Subjective questionnaire: 8.7-item increase. Strength increase, cylindrical grasp: 893 g; LKP: 488 g. Summary: Gains in ADLs and strength are made with surgery.
Lo et al. (1998)69	8	Chronic traumatic tetraplegia 9 patients, 12 hands: C6 Age: 23–40 y Procedures: ECRL or B to FDP 4.5 y after injury	To evaluate, subjectively and objectively, improvements after tendon transfers	Retrospective cohort study Compared surgical hand with non-surgical hand	Lamb and Chan Questionnaire Grip and LKP dynamometry MRMT Carroll Upper Extremity Function Test Questionnaire rated pre and post state at time of study	Analysis: Descriptive statistical comparisons of surgical (S) vs. non-surgical (NS) hand Effect: Individuals perceived improvement from memory. Grip strength: 24kg (S) / 21.5kg (NS); LKP strength: 1.2kg (S) / 1.0kg (NS). MRM did not show a difference. Carroll Upper Extremity Test: 72.5 (S), 52.4 (NS). Summary: Objective measures show small differences between surgical and non-surgical hands.

Note: Table applies to the C5/C6/C7 complete sub-group of tetraplegia in Figure 1.

BR=brachioradialis; ext=extension; ECRL=extensor carpi radialis longus; EPL=extensor pollicis longus; FCR=flexor carpi radialis; FDP=flexor digitorum profundus; FIM=Functional Independence Measure; FPL=flexor pollicis longus; GRT=Grasp and Release Test; ISNCSCI=International Standards of Neurological Classification of Spinal Cord Injury; JHFT=Jebsen Hand Function Test; LKP=lateral key pinch; MMT=manual muscle testing; MRMT=Minnesota Rate of Manipulation test; QIF=Quadriplegia Index of Function; ROM=range of motion; SCI=spinal-cord injury.

The studies listed in Table 3 present convincing evidence to support tendon-transfer surgery as an intervention that can be provided once an individual is neurologically stable. Laffont and colleagues make a strong argument that more active joints improve the ability to grasp, release, and manipulate an object.⁶⁴ The ability to grasp and release, in turn, gives individuals the ability to perform more ADL-related tasks independently. Clearly, patients' perception of improvement in upper-limb function after tendon-transfer surgery is very strong.^65–69 It has been reported that lateral key pinch and hook or cylindrical grasp can be improved with tendon-transfer surgery^68,69 but that strength can decrease slightly in the long term,^66,67 likely as a result of use patterns over time. However, the diminished strength does not decrease the independence gained from the procedure, as ADL ability is almost always improved, according to the patient, and long-term follow-up shows stability of functional status.^66,67

Despite the reported benefits of surgery, more rigorous research is required to establish its effectiveness. Existing studies, in general, are not strong enough to substantiate the benefits of any one type of surgery, as most studies include and group different surgical approaches. Further, the studies do not have consistent baseline and follow-up data, and the ADL assessment approach has not been consistent across all studies. Currently, there is a desire to standardize the measurements used by SCI researchers around the world. Because sample sizes are small, most studies use both limbs in the data analysis, often using one limb as a control or treating the two arms as separate entities. Finally, many of the studies are retrospective in nature and do not include adequate baseline measurements. A prospective study for each type of surgery should be conducted separately, with appropriate baseline and follow-up measures.

Despite the evidence for functional gains made after tendon transfer, there is a marked under-use of surgical procedures to enhance function for individuals with tetraplegia. The reasons for this are unknown, but the implication is that the postoperative course of rehabilitation and access to appropriate health benefits are barriers to individuals' agreeing to undergo these procedures.^70,71 Although some researchers feel that there is adequate research to support the use of tendon-transfer surgery, the evidence for tendon transfer should be more rigorous.

One element not addressed in the surgical literature is that the loss of function incurred by the tendon transfer is not measured. Surgical intervention to provide a movement that the individual did not previously have will enhance strength and function at a specific joint; however, these gains need to be compared to the risks of surgery, the rehab course, the potential complications, and what is lost by taking away the function of an active muscle. Only then can the procedures be evaluated for true gain. A full utility study may be warranted.

In summary, the availability of FES methods as assistive devices and interventions is increasing. Some research by the clinician is required, as devices and methods of treatment vary in price, availability (based on location and standards), ease of use, and specificity for tetraplegia. In addition, some of the devices continue to undergo further development by their inventors, and newer versions of these devices are now or may soon be available on the market. Nonetheless, FES methods are currently being implemented as interventions and devices in a number of centres around the world. SIs also continue to be performed, which will enable the field to generate a superior quality of research to support this method.

Limitations

Rigorous research in the field of SCI is limited by the following factors:

• Small sample sizes, large variances within samples, decreased power of studies

• Inability to randomize treatment group versus placebo for ethical reasons

• Difficulty in blinding subjects for many types of interventions because individuals may not want to be part of a non-treatment group and/or because individuals know that they are receiving treatment

• Limited availability of standardized outcome measures appropriate for use as primary outcome measures in studies and lack of consistently used measures across the field

These factors play a significant role in the overall methodological quality of the literature. Since any conventional review methodology would eliminate all of the existing studies, it was necessary to modify an existing methodology to focus on the SCI literature; however, a modified tool does not maintain the established psychometric properties of the original, which is a limitation of this review. Moreover, a systematic review should include evaluation of each article by two reviewers and definition of a consensus for the quality assessment; such a process was not incorporated into the present review, which constitutes another limitation of this work. However, the review's objective was to modify a standard procedure to extract the best available evidence to support interventions that restore upper-limb function.

In those studies that present with sufficient methodological quality (see Tables 2 and 3), many different outcome measures were used. The lack of consistent use of outcome measures across this literature makes comparing results among studies difficult and almost impossible. In fact, comparing different interventions is not an option when different outcomes are used to assess efficacy. Therefore, the results of these studies can be interpreted only individually at this stage in the field.

Future Directions

Although efficacy has been established for some therapeutic interventions, future trials of clinical effectiveness are required. These studies should incorporate the attributes of well-designed randomized controlled trials, be multi-centre in nature, and include stronger statistical rigour than previous research. Larger sample size is a significant factor in substantiating the evidence for research in the SCI field and will likely require cooperation among multiple centres around the world. Measurement approaches must be standardized to enable comparison studies. Furthermore, the measures used must have the sensitivity required to demonstrate responsiveness to small changes. In the design of studies in which interventions are compared, large homogeneous samples are necessary to make true comparisons.⁷² Future studies will need to consider determinants of recovery—such as the specific causes and severity of injury, level of lesion, mechanism of injury, and primary/secondary injury—that influence the recovery profile and play a role in the outcomes of effectiveness studies, in turn influencing the prescription of therapeutic interventions.

CONCLUSIONS

Despite the lack of available scientific evidence for their efficacy, conventional therapies continue to be implemented in the field. As research in the field improves, however, new interventions are beginning to show promise. Electrical stimulation therapies (mainly FES and SS) are beneficial to enhance functional ability when used as assistive technologies and influence recovery when used as therapies in the subacute phase of recovery. Measurement strategies for the field need to be standardized among laboratories and clinical settings before the effectiveness of therapies can be established in randomized controlled trials. Surgical interventions have been shown to have an impact on functional independence for individuals with tetraplegia; however, further supporting evidence of efficacy is required. There has been a recent thrust toward a stronger and focused research approach for upper limb restoration in tetraplegia; the momentum must continue to establish and substantiate best practices, because for an individual with tetraplegia, upper-limb function translates into global function and independence.

KEY MESSAGES

What Is Already Known on This Topic

While there is substantial literature specific to restoration of upper-limb function in individuals with spinal-cord injury (SCI), shortcomings of this research contribute to an insufficiency of evidence to support a change in rehabilitation practices for this population. Although efficacy has been shown for some interventions, well-designed effectiveness studies are required as an immediate next step.

What This Study Adds

This review identifies the elements required for future research to advance the field of SCI rehabilitation for tetraplegia. The key limitations of many of the studies in this field are insufficient sample size and sample selection, lack of standardization of study protocols, poor use of appropriate outcome measures, and failure to establish meaningful clinically important differences. These limitations of the existing literature have hindered the progress of evidence-based practice in this field. If future research is to be successful, stronger efforts to develop more specific outcome measures, larger sample sizes generated by multi-site collaboration, and a better understanding of sample selection are required, and study protocols must be aligned across centres; attention to these requirements will lead to better-designed effectiveness studies. The optimal interventions for improving upper-limb function for individuals with tetraplegia are electrical stimulation therapies, for which there is currently the most convincing evidence to influence rehabilitative approaches.

Physiotherapy Canada 2011; 63(4);474–89; doi:10.3138/ptc.2009-46