Are We Giving Kids a Hand? Wrist and Hand Splints in the Management of Cerebral Palsy

By Phil Stevens, MEd, CPO, FAAOP

In 2008, the International Society for Prosthetics and Orthotics (ISPO) convened a consensus conference in Oxford, England, called Recent Developments in Healthcare for Cerebral Palsy: Implications and Opportunities for Orthotics. I participated in the event by conducting a systematic literature review on the orthotic management of the hips, trunk, spine, and upper limbs.1 Unfortunately, there was not much research about wrist and hand splints to review. There was a small series of early case studies and anecdotal reports published in the 1980s and early 1990s; however, from 1994 to 2008, the time frame that I had been assigned to examine, there was only a single published study, and it came out just months before the conference was held.

Recent years have seen a rapid expansion of the available literature on the topic, and this article will provide a better answer to the question that could only be marginally answered during the 2008 ISPO review: Are there established benefits associated with the bracing of the wrist and/or hand with static and/or dynamic splints in children with cerebral palsy (CP)?

Phrased more broadly, patients with CP often present with limitations in their affected extremities including thumb adduction (or thumb-in-palm deformity), ulnar deviation at the wrist, and wrist and finger flexion. To what extent and in which populations do wrist and hand splints appear to aid in functional ability?

Defining the Treatment Population

CP is an umbrella diagnosis, encompassing a broad range and spectrum of disorders. Clinicians are increasingly aware of the Gross Motor Function Classification System (GMFCS), which is now widely used to classify individuals based on their mobility limitations. As a profession, we are generally less aware of the Manual Ability Classification Scale (MACS), a similar classification scale used to delineate the functional abilities of the upper limb.2

As with the GMFCS, the MACS consists of five functional levels. Both classification scales begin with the level of least disability, with subsequent levels representing gradual increases in disability. The narrative descriptions of each MACS level are shown in Figure 1.

Figure 1

Static versus Dynamic Splinting: Should Some Wrist Motion Be Preserved?

In the first of these recent publications, researchers examined the question of whether or not limited wrist motion should be preserved in splinting approaches by grading the performance of their subjects in three conditions: 1) no intervention; 2) static splinting (fixing the wrist at 15-20 degrees of wrist extension); and 3) dynamic splinting (a spiral design, also set at 15-20 degrees of wrist extension but permitting about 30 degrees of resisted wrist flexion and extension).3 Both splint types were low-temperature thermoplastic designs more commonly identified with splints that are fabricated on-site by an occupational therapist.

The study subjects were children with right (n=5) or left (n=5) hemiplegia and with MACS level 3 or 4 ability. Thus, they were children with more involved dexterity limitations who could handle objects but only with some difficulty. Further, even with easily manipulated objects, they often required adaptations to the manual tasks. In addition to testing their grip and pinch strength in each of the three conditions outlined earlier, researchers administered an adapted Nine-Hole Peg Test (NHPT) in which the children were timed as they removed nine one-inch-diameter wooden dowels from a five-by-five-inch pegboard.

As the sole measure of dexterity, the NHPT seemed to suggest functional enhancements with either of the splinting conditions. For the subjects with right hemiplegia, the average times to complete the NHPT decreased by roughly 20 percent in both splint designs compared to their unbraced performance times. A comparable improvement was observed for subjects with left hemiplegia in the dynamic splinting condition. Relative improvements were also measured with the static splint for subjects with left hemiplegia, but these were less than those recorded with the dynamic splint.

In contrast to the dexterity assessment, measurements of grip and lateral pinch (or key grip) strength were variable, depending upon the affected side and type of splinting. Irrespective of the presence or type of hand splint, the relative strengths of both prehension patterns were substantially lower for subjects with left hemiplegia than those observed in subjects with deficits on their right sides. Among patients with right hemiplegia, the strongest measures of grip strength were obtained with the dynamic wrist splint (10.8 lb.), and the unbraced condition (7.9 lb.), followed by the static wrist splint (5.4 lb.).

Also among the subjects with right hemiplegia, with respect to the lateral pinch task, both splinting approaches appeared to compromise strength. The average lateral pinch in the unbraced condition (6.1 lb.) exceeded that observed in both the dynamic splint (5.3 lb.) and static splint (4.1 lb.) conditions. However, considerable variability existed in both strength measures, with standard deviations among the five subjects with right hemiplegia often exceeding the mean strength values in the various conditions.

In summary, among subjects with MACS 3 or 4 function, both splinting approaches across the wrist appeared to benefit functional dexterity. In contrast, with respect to grip strength, the dynamic splinting approach appeared to facilitate a stronger grip than either the rigid splint or unbraced condition. Key grip appeared compromised by both splinting approaches.

Is Splinting More Beneficial in Patients with Less Disability?

In a more recent work, bioengineers from Brazil performed a similar trial with a cohort of 32 subjects.4 While the mean age of this cohort was comparable to that of the previous study (8½ years old), the children in the Brazilian study had less severe hemiplegia. By limiting their analysis to MACS level 1 and 2 patients, the authors were dealing with subjects capable of successfully handling objects, either easily or with a reduced quality and/or speed of achievement.

Their intervention, consistent with the Benik-style splint that is familiar to many pediatric orthotists, was a neoprene wrist hand orthosis (WHO) in which the wrist was held in relative extension by a volar thermoplastic stay and the thumb was positioned out of the hand as the proximal joints of the thumb were supported by a neoprene stay. As with the prior study, a dynamometer was used to grade grip strength. In contrast to the 20 percent improvement experienced with the wrist splinting in the prior study, the more able-bodied subjects experienced an average improvement of 50 percent with the neoprene WHO.4

As with the observations with the timed peg test in the earlier study, the subjects in this investigation consistently demonstrated improved dexterity in their braced conditions by reducing the average time needed to pick up small, common objects (13 percent reduction), stack checkers (18 percent reduction), pick up large, light objects (16 percent reduction), and pick up large, heavy objects (24 percent reduction).4 Using an elaborate image acquisition technique with passive reflective markers, the authors report expanded range of motion of about 10 degrees at the trapeziometacarpal joint at the base of the thumb in both flexion/extension and abduction/adduction during the performance of tripod grasps, cylindrical grips, and lateral pinches.4

Just as the subjects with more debilitating hemiparesis appeared to experience benefits in dexterity and strength with an appropriate wrist orthosis, subjects with less severe hemiplegia experienced similar, if not greater, benefits in grip strength, functional dexterity, and range of motion with a neoprene-style WHO.

Engaging the Compromised Hand

In a third investigation, the emphasis was placed on the extent to which the affected hand, or "assisting hand," engaged in the activity or task rather than on the speed with which an activity was performed. As with the previous studies, a cohort of subjects with hemiplegia (n=25) with an average age of 8 years was assembled. The manual abilities of the cohort were variable. The most predominant MACS level was 2 (n=12), with MACS level 1 (n=6) and MACS level 3 (n=7) also represented. The intervention was a canvas WHO with a thenar support and aluminum strut crossing the volar surface of the wrist to position it in relative extension.5

The assessment measure used in this study was the Assisting Hand Assessment (AHA). The AHA is described as "a test kit with selected toys that stimulate spontaneous use of both hands." During the assessment, the child is videotaped across his or her performance with 22 items and subsequently rated on each one. Scores for each item can range from one (no use of the affected hand) to four (efficient use of the affected hand). A summary AHA score is then obtained.

The AHA was administered three times in a classic A-B-A study design where A represents the unbraced condition and B represents the in-brace condition. In this study, there was no acclimation period; the second administration of the AHA occurred immediately after the brace was fit.

Improvements in the AHA score with the WHO were observed across all functional levels. In roughly half of the subjects, these gains were substantial enough to declare clinically significant improvements in function. Further, upon removal of the orthosis, AHA scores returned to those observed in the first unbraced assessment. Summarizing their observations, the authors report, "With a brace, children were able to grasp objects of different shapes without interaction of the dominant hand.... Children were also able to open their hand and hold objects, such as a bottle or marbles, more easily when wearing a brace. Their grip was more powerful, their hand stabilized the objects, and more interacting between both hands was possible."5

The authors went on to hypothesize two possible reasons for the improvements. Of these, the more intuitive explanation was that the support provided by the brace as it crosses the wrist joint and the base of the thumb enables more dexterity. An alternative explanation that warrants some consideration is that the children became more aware of their assisting hands because of the tactile sensation and, therefore, used them more.

This article added to the findings of its predecessors by documenting improved quality of engagement from the assisting hand in the presence of a WHO that stabilizes the wrist joint and base of the thumb.

What about Bracing for Younger Patients?

The final paper reviewed is distinct from the others in both its target treatment cohort and the details of its intervention. In contrast to the earlier studies, all of which investigated hand and wrist splinting in subjects with a mean age of 8½ years, this study examined a fairly small cohort of subjects with an average age of 3½ years (n=7). Further, while the interventions in the previous studies were WHOs, this study utilized a thumb opponens splint (the McKie splint), which does not extend proximal to the wrist joint. The functional abilities of the subjects were reported as MACS level 2 or 3.6

Also unique to this study was the way the researchers evaluated the efficacy of the intervention. Rather than having a uniform task that was timed or evaluated for quality, the parents and investigators chose an individual functional goal for each child and then rated his or her performance of that goal over time with the intervention. These tasks had to be bimanual and ultimately included such things as drinking from a large cup, grasping a large toy car, buttering bread, pulling the top off of a felt-tip pen, and donning trousers and socks.6

Progress toward improvement in each child's task was ultimately determined by the subjective assessments of the investigators and the parents of each child. Improvements with the orthoses were observed in six of the seven subjects. In four of these six children, the improvements persisted once the use of the hand splint was discontinued. The two remaining subjects demonstrated improvements only when wearing the splint. While it would be reasonable to suppose that the functional improvements observed in these children were the result of the hand splint, given the young ages of the study contingent, they may have also resulted from a learning effect that occurred with daily repetitions of the activity, simple maturation, or both.

Despite these limitations, the study did observe that all parents reported good compliance with the use of the splint and that their children seemed to be more aware of their hands when using the splints.

Are We Giving Kids a Hand?

In contrast to 2008 when there was little literature to provide a substantiated answer to this question, these recent publications allow for a more informed discussion. There is emerging evidence to suggest that across the spectrum of manual function in children with hemiplegia, the use of a WHO that positions the wrist in relative flexion and holds the thumb in opposition appears to enhance manual dexterity, increase grip strength, and encourage spontaneous use of the affected extremity. The effect of an opponens splint in toddler-age children is less defined, but high compliance rates have been reported. To that end, the evidence suggests that the practice of providing appropriate wrist and hand splints to these patients is warranted and may lead to enhanced functionality.

Phil Stevens, MEd, CPO, FAAOP, is in clinical practice with Hanger Clinic, Salt Lake City, Utah. He can be reached at


  1. Stevens, P. 2009. A systematic review of the use of orthoses in the management of patients with cerebral palsy: Hip, trunk, spine and upper limb. Recent Developments in Healthcare for Cerebral Palsy: Implications and Opportunities for Orthotics, ed. C. Morris and D. Condie, 205-34. Copenhagen: International Society for Prosthetics and Orthotics.
  2. Eliasson, A. C., L. Krumlinde-Sundholm, B. Rösblad, E. Beckung, M. Arner, A. M. Ohrvall, and P. Rosenbaum. 2006. The Manual Ability Classification System (MACS) for children with cerebral palsy: Scale development and evidence of validity and reliability. Developmental Medicine and Child Neurology 48 (7):549-54.
  3. Burtner, P. A., J. L. Poole, T. Torres, A. M. Medora, R. Abeyta, J. Keene, and C. Qualls. 2008. Effect of wrist hand splints on grip, pinch, manual dexterity, and muscle activation in children with spastic hemiplegia: A preliminary study. Journal of Hand Therapy 21 (1):36-43.
  4. Barroso, P. N., S. D. Vecchio, Y. R. Xavier, M. Sesselmann, P. A. Arauìjo, and M. Pinotti. 2011. Improvement of hand function in children with cerebral palsy via an orthosis that provides wrist extension and thumb abduction. Clinical Biomechanics 26 (9):937-43.
  5. Louwers, A., A. Meester-Delver, K. Folmer, F. Nollet, and A. Beelen. 2011. Immediate effect of a wrist and thumb brace on bimanual activities in children with hemiplegic cerebral palsy. Developmental Medicine and Child Neurology 53 (4):321-6.
  6. Ten Berge, S. R., A. M. Boonstra, P. U. Dijkstra, M. Hadders-Algra, N. Haga, and C. G. Maathuis. 2012. A systematic evaluation of the effect of thumb opponens splints on hand function in children with unilateral spastic cerebral palsy. Clinical Rehabilitation 26 (4):362-71.