Assessing functional differences in gross motor skills in children with cerebral palsy who use an ambulatory aid or orthoses: can the GMFM-88 help? Dianne J Russell* MSc, Associate Professor, School of Rehabilitation Science, and Co-investigator, CanChild Centre for Childhood Disability Research, McMaster University, Hamilton, Ontario, Canada; J W Gorter MD PhD, Paediatric Physical Medicine and Rehabilitation, Rehabilitation Centre De Hoogstraat, and University Medical Centre, Partner of NetChild, Network for Childhood Disability Research, Utrecht, the Netherlands. *Correspondence to first author at CanChild Centre for Childhood Disability Research, Rm 408, IAHS, McMaster University, 1400 Main Street W, Hamilton, Ontario, L8S 1C7, Canada. E-mail:
[email protected]
The purpose of this study was to determine whether the Gross Motor Function Measure (GMFM-88) is sensitive to withinchild changes in function as a result of children who use an ambulatory aid or orthoses in comparison with unaided or barefoot function. Data from 257 children (140 males, 117 females) with cerebral palsy (CP) were analyzed from a 5-year longitudinal study. The children’s age ranged from 2 to 15 years (mean 7y 4mo; SD 2y 11mo), and type of CP, included spastic (n=206 [80.2%]), dyskinetic (n=13 [5.1%]), ataxic (n=9 [3.5%]), hypotonic or mixed (n=27 [10.5%]), and those missing (n=2 [0.8%]), and in motor ability (Gross Motor Function Classification System [GMFCS] levels I, n=40, II, n=34, III, n=93, IV, n=76, and V, n=14). Paired t-tests between barefoot and aided assessments showed significantly higher GMFM-88 total scores overall (n=257) for aided assessment and for three subgroups (ankle–foot orthoses only, ambulatory aid only, ankle–foot orthoses plus ambulatory aid), providing evidence that the GMFM-88 is sensitive to functional changes as a result of using an aid and/or orthoses. With regard to changes within the orthoses-only group, significant changes varied by GMFCS level. The GMFM-88 might provide a clinically useful tool to help in understanding the impact of ambulatory aids and orthoses on gross motor skills of children with CP. See end of paper for list of abbreviations.
462
Developmental Medicine & Child Neurology 2005, 47: 462–467
Ambulatory aids and orthoses are frequently prescribed for children with cerebral palsy (CP), either separately or in combination. Ambulatory aids such as walkers, crutches, and canes are intended to increase stability by broadening the child’s base of support to enable standing and walking. The reasons for lower limb orthotic prescription may include the following four goals: first, the correction and/or prevention of deformity; second, the provision of a base of support; third, the facilitation of training in skills; and fourth, the improvement of gait efficiency (Condie and Meadows 1995). Morris (2002a) highlights where these treatment goals for the use of orthoses fit within the International Classification of Functioning, Disability and Health framework (ICF; World Health Organization 2001). He proposes that the first goal is related to changing the impairment of the body structure, whereas the other three relate more to the activity dimension of the ICF. Evaluation of the effectiveness of ambulatory aids and orthoses needs to be relevant to the goal of treatment. If, for example, the primary goal is to improve a child’s activity level, then impairment-based measures of muscle activity and joint range of motion (ROM), although useful in understanding the effect on the underlying impaired body function or structure, will not provide the necessary information on the activity level. It is for this reason that it is important to be clear about what aspect of a person’s situation is being assessed. Two recent reviews of the efficacy of lower-limb orthoses for children with neurological impairments describe the limitations of the evidence that orthoses achieve their treatment objectives (Morris 2002b, Teplicky et al. 2002). The evidence from the studies reviewed showed that there was an improvement in walking patterns (stride length, step length, and velocity) when children wore ankle–foot orthoses (AFOs) compared with when they walked barefoot; however, this difference was less discernible when the children wore shoes. Methods of evaluating the impact of aids or orthoses have been primarily by clinical observation of the quality of movement, particularly gait. Standardized assessments have included measures of passive ROM with the use of goniometers, or active ROM and postural control and balance with the use of three-dimensional clinical gait analysis, force plate information, and oxygen cost during activity. However, the use of a fully equipped gait laboratory is expensive and time-consuming and this prohibits its widespread use as a feasible assessment for orthotic interventions. Passive ROM is relatively inexpensive and quick to assess but it is known to be unreliable even with trained assessors (McDowell et al. 2000). Moreover, when the primary goal is to improve a child’s activities, impairment-based measures such as ROM are not necessarily appropriate or helpful. In these instances we are more concerned with the effect of the orthoses on everyday activities such as playing on the floor or climbing stairs. What is required therefore are standardized assessments of motor function activities that are relatively inexpensive, clinically feasible, have strong psycho- metric properties of reliability and validity, and are sensitive to changes in standing and walking abilities. The Gross Motor Function Measure (GMFM) is a measure designed specifically for children with CP to evaluate change in gross motor function (Russell et al. 1989). The 88-item GMFM (now referred to as the GMFM-88) is a standardized assessment with demonstrated reliability, validity, and responsiveness
to change over time (Russell et al. 2002). What is less clear is the evidence for its usefulness in evaluating the impact of aids and/or orthoses. Although there is a section of the GMFM that allows one to assess and score the function of a child wearing AFOs or using ambulatory aids, the criteria for when to apply the aid are not standardized and the psychometric properties of the GMFM-88 for use with aids and orthoses have not been investigated. Nevertheless, the GMFM has been used in four studies evaluating the effectiveness of AFOs. Evans et al. (1994) used the GMFM and the Gross Motor Performance Measure (GMPM) in a randomized control trial to evaluate the effects of lower extremity orthoses on 34 preambulatory children with spastic CP. Although both groups improved significantly over 4 months, they found no significant difference between the orthoses group (orthoses plus physiotherapy) and the control group (physiotherapy only) on gross motor function as measured by the GMFM-88 dimensions D (Standing) and E (Walking, Running, and Jumping). However, there was no comparison of aided versus barefoot function. Ferdjallah et al. (2000) looked at the relationship between postural stability, gait, and GMFM goal scores in a study of the effectiveness of two types of AFO (hinged and supramalleolar) in a sample of five children with spastic diplegic CP. They concluded that GMFM goal total scores increased when children were assessed with orthoses in comparison with barefoot assessment and recommended a combination of measures to evaluate the effects of AFOs, including measures of postural stability, gait, and GMFM assessment. However, the authors did not provide information on which dimensions of the GMFM were used to calculate the goal total scores, nor did they provide statistical analyses to support their conclusions. In a study of 10 children with spastic diplegic CP who used hinged AFOs, Maltais et al. (2001) found a significant decrease in oxygen uptake when children used their hinged AFO compared with barefoot walking but no change in GMFM goal total scores (Standing; and Walking, Running, and Jumping dimensions). Buckon et al. (2001) looked at the effects of three different AFOs (hinged, posterior leaf spring, and solid) on 30 children with hemiplegic CP. They found that children could walk faster with certain types of AFO without increasing the energy cost of walking. There was no significant difference in GMFM scores (Standing; and Walking, Running, and Jumping dimensions) between the different AFOs or between AFO use and walking barefoot. In summary, there is conflicting evidence on the impact of orthoses on gross motor function for children with CP using the GMFM-88. This could be because changes seen in gait parameters with the use of AFOs do not translate into changes in other gross motor abilities; the GMFM-88 is not sensitive to the functional changes brought about by the use of AFOs, or the small sample sizes of the studies increased the likelihood of a type II error (failure to reject the null hypothesis even when it is true). The purpose of this paper is to use data from a large study, in which many of the participants had both aided and unaided GMFM-88 assessments, to determine whether the GMFM88 is sensitive to clinically important changes in function as a result of children using a walking aid or orthoses in comparison with unaided or barefoot function. In addition, recommendations for the use of the GMFM-88 in assessing walking aids and orthoses are discussed.
Method PARTICIPANTS
Children who participated in the Ontario Motor Growth (OMG) study (n=657) were eligible for participation in this study. The OMG study is a population-based longitudinal study of motor development for children with CP (Rosenbaum et al. 2002). The method of obtaining the randomized, stratified sample of children from the province of Ontario is reported in detail by the authors. Children were between the ages of 1 and 13 years at entry to the OMG study and were evaluated over a 4year period (1996 to 2000). Children were assessed with the GMFM-88 and Gross Motor Function Classification System (GMFCS; Palisano et al. 1997) every 6 months if they were under 5 years old and yearly for children 5 to 16 years old. Because children were stratified by severity they represented the full spectrum of functional abilities as measured by the GMFCS. All records were reviewed to identify a sample of children who had at least one GMFM-88 assessment tested with and without an aid and/or orthoses. Our primary purpose was to compare the impact of aids and orthoses (between unaided and aided function) on within-child gross motor abilities. If a child had more than one GMFM assessment with an aided score, then the child’s first assessment with an aided score was used for the analyses. Children’s assessments were excluded if there was no information on which type of ambulatory aid was used during the assessment. MEASURES
Severity of functional abilities was judged by the clinical therapist with the use of the five-level GMFCS (Palisano et al. 1997). The GMFCS was developed specifically for children with CP and has been shown to have good reliability and validity (Palisano et al. 1997, 2000; Wood and Rosenbaum 2000). The GMFCS is an ordinal-level classification system used to describe a child’s abilities and limitations in gross motor function. Descriptions are provided for different age groups: less than 2 years, 2 to less than 4 years, 4 years to less than 6 years, and 6 to 12 years old. Distinctions between levels are based on functional limitations, the need for assistive mobility devices (walkers, crutches, canes, or wheeled mobility), and, to a smaller extent, the quality of movement (Palisano et al. 1997). The GMFM-88 is a performance-based measure of gross motor function, with 88 items ranging in difficulty from activities in lying and rolling to more complex activities such as hopping, jumping, and stair climbing. The GMFM-88 was administered to the children by therapists trained in its use and tested to ensure an adequate level of competence by using a criterion test videotape (Russell et al. 1994). In accordance with the administration and scoring guidelines, therapists were instructed first to test all 88 items without aids/orthoses or shoes. If the child typically used an ambulatory aid or orthoses, the therapist would redo items in the dimensions of the GMFM in which the aid/orthoses would generally be used (primarily in dimensions D [Standing] and E [Walking, Running, and Jumping]). Once an aid or orthoses was used, therapists were required to complete all items within that dimension. DATA ANALYSIS
Statistical analyses included descriptive, summary statistics and paired t-tests with 95% confidence intervals to compare aided and unaided assessments. Two-tailed tests with a significance level of α
