INTRODUCTION AND PURPOSE
“Cerebral Palsy (CP) describes a group of permanent disorders of the development of movement and posture, causing activity limitations that are attributed to non-progressive disturbances that occurred in the developing fetal or infant brain. The motor disorders of CP are often accompanied by disturbances of sensation, perception, cognition, communication, and behaviour and as well as by epilepsy, and by secondary musculoskeletal problems.”1 (p9)
Despite the best efforts of rehabilitation professionals from many disciplines, a great deal remains unknown about the clinical course of CP.2 However, a classification system based on clinical judgments known as the Gross Motor Function Classification System (GMFCS) was developed to assist with the prognosis for motor function in children with CP.2 The classification system is an ordinal scale that differentiates children by age and level of motor function: level I (walks without limitations), level II (walks with limitations), level III (walks using a hand-held mobility device), level IV (self-mobility with limitations and/or may use powered mobility), and level V (transported in a manual wheelchair).2,3 The GMFCS has high interrater reliability (κ = 0.75),2,4 especially when used with children older than 2 years. The GMFCS has stable test-retest reliability (G = 0.93),4 indicating that a child's GMFCS level, in general, will remain consistent over time. This allows service providers to assist parents in understanding the future method of mobility for their children.
The variety of body systems affected by CP puts children at high risk for primary and secondary impairments and associated health conditions that may require day interventions or hospitalizations for medical or surgical procedures during their developmental years.5 Primary impairments may include hypertonicity, spasticity, poor balance, poor force production, and impaired coordination. Secondary impairments may include muscle contractures, bony deformities, muscle imbalances, dysplasias, ligamentous laxity, and low endurance5,6 as well as health conditions including impairments of vision, learning and understanding, speaking and communicating, emotional regulation, oral motor function, digestion, and epilepsy.7 A literature review was conducted using the key words such as cerebral palsy, CP, Gross Motor Function Classification System, GMFCS, surgical procedures, medical procedures, medical techniques, surgical techniques, children, and child. Although articles were found addressing the outcomes of hospitalizations or medical and surgical procedures experienced by children with CP, there were no studies addressing the prevalence of procedures across GMFCS level or geographical region.
Two primary objectives of this study were to (1) determine the variation in medical and surgical procedures experienced by children with CP by functional ability using the GMFCS and (2) determine the variation in medical and surgical procedures experienced by young children with CP by geographical region. A secondary objective was to explore the relationships between both the frequency of the most prevalent procedure and the total number of medical and surgical procedures among young children and (1) change in motor function over a 1-year period (ie, change score over the 1-year interval) and (2) motor function at the end of the 1-year period (ie, full motor score at the end point). This information will help service providers and parents understand the range of procedures that other young children with CP might experience and provide insight to procedures that are associated with motor function.
Participants were selected from the Movement and Participation in Life Activities of Young Children (Move & PLAY) study database of 429 children between the ages of 18 months and 5 years.8 Children between the ages of 18 months and 5 years were chosen for data collection as this marks a time of relatively rapid motor growth.8 Children included had a diagnosis or a suspected diagnosis of CP. We included children with a suspected diagnosis because, at some sites, the practice was to defer diagnosis until the children were school-aged. One of our coinvestigators, Dr Peter Rosenbaum, suggested the strategy of including children if (1) they were assigned a GMFCS level, (2) therapists identified a distribution of involvement, and (3) therapists identified children who had movement patterns consistent with a diagnosis of CP. This strategy is similar to that of other studies conducted through CanChild, Centre for Childhood Disability Research.9 Children and their parents must have been able to speak English, French, or Spanish. This sample was representative of the distribution of GMFCS levels from population-based studies.7 The children and their families were recruited from a number of sites across Canada and the United States, including 8 sites in Canada (from Newfoundland, Nova Scotia, Ontario, Manitoba, Saskatchewan, and British Columbia) as well as regions around Philadelphia (Pennsylvania), Atlanta (Georgia), Oklahoma City (Oklahoma), and Seattle (Washington). Recruitment was conducted by regional coordinators in the United States and a project manager in Canada. Written informed consent was obtained from the participants' caregivers prior to data collection. The Move & PLAY study was approved by the Research Ethics Board at the University of Western Ontario and all academic institutions involved, as well as several clinical sites (n = 21 approvals).8
The sample for this study consisted of 386 children with CP from the Move & PLAY database for whom data were collected at all study time points. Three participants were excluded because of a combination of inconsistent classification of GMFCS levels and distribution of involvement over the course of the study, as well as lack of a definitive diagnosis of CP and lack of therapist judgment that the child looked as though he or she had CP. Demographic characteristics of participants are contained in Table 1 and included 216 boys and 170 girls with an average age of 38 months (SD = 11.5) at the study onset. Families who remained in the study through the 1-year period had significantly higher incomes than families who withdrew or were lost to follow-up; all other child and family demographics did not differ between these groups.
Data came from a detailed questionnaire developed by the Move & PLAY investigators, which was completed by each participant's caregiver covering all aspects of the children's health, including hospitalizations or medical and surgical procedures. Specifically, at the beginning of the 1-year study, parents were asked whether their children had experienced any neurological surgeries (ie, shunts, other brain surgeries, dorsal rhizotomy, or “other”) or surgeries on muscles or tendons or bones. They had several opportunities to report on “other” surgeries and hospitalizations (including the reason for hospitalization). They were also asked about spasticity management (other than dorsal rhizotomy) such as Botulinumtoxin A (Botox), baclofen, casting, or “other.” Finally, they were asked about their children's use of medical technology, including a gastrostomy tube, a tracheotomy tube, or a ventilator. For medical and surgical procedures, parents were asked whether they were conducted “within the last year” or “longer than a year ago.” At the end of the study, parents were asked about any surgeries, hospitalizations, or procedures for spasticity management over the year since they were first involved in the study. Please see the Appendix for the questions asked at these 2 time points.
The GMFCS was used to classify the functional abilities of the children. Selection of this classification system is supported by strong reliability and validity,2,3 as well as significant uptake in the research and clinical communities.10 The Gross Motor Function Measure-66-Basal and Ceiling approach (GMFM-66-B&C)11 was used as the measure of motor abilities. Each item is measured by observation of the best performance of a maximum of 3 trials and scored on a 4-point ordinal scale from 0 (does not initiate) to 3 (completes). The Gross Motor Ability Estimator software12 was used to calculate a GMFM-66 score from as few as 15 items completed between 3 sequential scores of “3” and 3 sequential scores of “0” when items were placed in order of difficulty. Concurrent validity with the full GMFM-6612 was excellent (intraclass correlation coefficient = 0.987; 95% CI 0.972–0.994), as was test-retest reliability (0.994, 95% CI 0.987–0.997).11 An accurate estimate of the GMFM-66 can be obtained in 15 to 20 minutes.
Data Collection Procedures
The detailed questionnaire was completed at the study onset and updates were recorded 1 year after the first data collection point. Each child was classified into a GMFCS level on the basis of the therapist observation on 2 occasions over the study (ie, at both the beginning and the end of a 1-year study); the classification at the final data collection point was used for study purposes because reliability improves as children get older.2,4 A kappa statistic of 0.95 (P < .001) was computed between GMFCS levels on these 2 occasions for the 386 children in the study, indicating a high test-retest reliability. In the case of discrepancies, 5 and 10 children were classified to a level indicating greater and fewer motor challenges, respectively, over the year. The GMFM-66-B&C approach was also administered by trained and reliable raters, who obtained greater than 80% item agreement with the criterion test, at the beginning and the end of the 1-year study.
Several strategies were implemented to assist with data management. First, medical and surgical procedures and reasons for hospitalizations were classified under subheadings through consensus among all authors (Table 2). Second, data collected from the periods “within the last year,” “longer than a year ago,” and over the year of the study were pooled to reflect procedures experienced by young children. Frequency counts were completed on the subheadings generated through consensus across all data collection points. The following variables were excluded from analysis because of low prevalence: dorsal rhizotomy (n = 1), neurological surgeries other than brain surgery (n = 3), oxymetry (n = 1), pulse oximetry suction (n = 1), and oxygen/C-PAP (n = 1). Each data collection site was classified into an East, West, or Central region, depending on its location (note that this decision was made after ascertaining nonsignificant differences by country). The East region consisted of Halifax (Nova Scotia), St John's (Newfoundland), Atlanta (Georgia), and Philadelphia (Pennsylvania); the West region consisted of Seattle (Washington), and Vancouver and Victoria, British Columbia; and the Central region consisted of Oklahoma City (Oklahoma), and Hamilton (Peterborough) and Toronto (Ontario), Winnipeg (Manitoba), and Regina (Saskatchewan).8
The total number of hospitalizations and medical and surgical procedures in young children, collected from all of the parent-completed health questionnaires combined, as well as GMFM scores collected at the beginning and the end of the study, were first summarized descriptively. They were then analyzed using 1-way analyses of variance (ANOVAs) using SPSS version 17.0. The relationships between the frequencies of the most prevalent procedure and the total procedures and (1) difference in GMFM score over the period of a year and (2) GMFM scores at the end of the study were explored using the Pearson product moment correlation. A α level of < .10 was used to determine significance in this exploratory study; values were adjusted using the Bonferroni correction for the 20 comparisons, for an adjusted α level of .005. Post hoc Tukey's analyses were completed to further investigate subgroup differences.
Summary data for the medical and surgical procedures, in order from most to least prevalent, classified by GMFCS level, are presented in Table 3. The most prevalent procedure was the use of Botox injection. Although there were response options for tracheotomy and being ventilated, no one responded to these options. The average number of total medical and surgical procedures for all children in the study was 2.5 (SD = 2.0). The average GMFM score at the beginning of the study was 45.4 (SD = 20.6) and at the end of the study was 50.8 (SD = 21.5) for the sample as a whole. The average difference in GMFM scores over a period of a year was 5.4 (SD = 5.3), with an effect size of 0.26, which is considered “small.”13 Effect sizes for children in GMFCS levels I and II and III and IV were 0.60 and 0.75, respectively, indicating that when the variability of scores among the whole group was controlled, that effect sizes were “medium” and “large” for these 2 groups of children who had greater capacity to gain functional motor abilities than children in GMFCS level V. (Note: Effect size is calculated by dividing the difference in GMFM scores over the period of 1 year by the pooled standard deviations of scores at the beginning and the end of the study.) For subsequent analyses, GMFCS levels I and II were grouped because they represent children who have the potential to ambulate without gait aides, whereas children in levels III and IV will ambulate with gait aides. In contrast, children in GMFCS level V are nonambulatory; this latter category was not used in the subsequent analysis about relationships to motor function.
Variation in Medical and Surgical Procedures by Functional Ability
Table 4 contains the ANOVA results for the 7 significant contrasts including the results of Tukey's post hoc testing. Figures 1 to 6 contain the box plots of these significant procedures for Botox, gastrostomy tube, gastrointestinal tract, respiratory system, eyes, and shunt, respectively. The 1-way ANOVA for all medical and surgical procedures was significant (F = 13.86, df = 4, P < .001), with children in level V experiencing significantly more procedures than children in levels I, II, and III and children in level IV experiencing more than children in level I, as seen in Figure 7.
Variation in Medical and Surgical Procedures by Region
A significant difference across regions was found only for musculotendinous procedures (F = 15.76, df = 2, P < .001), with post hoc testing, revealing that East was significantly different from both West and Central. As seen in Figure 8, although statistical significance was obtained, effectively this meant that 2 children had 2 musculotendinous procedures in the East in contrast to none in the Central and West regions. The vast majority of these young children in all regions had no musculotendinous procedures, with frequencies of 4, 2, and 3 for 1 procedure for East, Central, and West regions, respectively.
Relationships to Motor Function
The most prevalent procedure was the use of Botox injections. The correlations between the number of both Botox injections and total procedures and GMFM scores are contained in Table 5 for the sample as a whole and GMFCS levels I and II versus III and IV. The only relationship to reach statistical significance for the number of Botox injections was GMFM scores at the end of the study for the total sample. The relationship was negative, and only 1.4% of the variation in motor function was explained by the number of injections. Similarly, no significant correlations were obtained between the total number of procedures and change in GMFM score. Two significant associations were obtained for the GMFM score at the end of the study: for the total sample (explaining about 15% of the variation in motor score, with children having received more procedures obtaining lower scores) and for children in levels I and II (explaining about 5% of the variance in motor score).
In this descriptive study, we examined the variation in medical and surgical procedures experienced by 386 children with CP by GMFCS level and geographical region. To the best of our knowledge, this is the first study to address this topic. Regardless of functional ability level, children in this study experienced a median of 2 or 3 procedures; however, the variability ranged from 0 to 8, 6, 9, 7, and 10 procedures for children in GMFCS levels I, II, III, IV, and V, respectively (Figure 7).
We determined that procedures involving gastrostomy tubes, the gastrointestinal tract, and the respiratory system were most prevalent in children with GMFCS level V. According to Palisano and colleagues,3 with regard to gross motor function, children in level I are most able to perform daily activities and children in level V are most limited. Therefore, a significant difference in hospitalizations or medical and surgical procedures would be expected between levels I and V because there is a large variance in functional ability between these levels. Eye procedures were of greatest prevalence in children classified as levels III and IV. Children in level III were most likely to undergo procedures involving the insertion or revision of a shunt. Although statistically significant, the magnitude of the difference among regions was not clinically relevant (ie, the largest difference was that 2 children in the East had 2 musculoskeletal procedures, in contrast to none in Central and West regions).
Our research indicated that Botox is the most prevalent procedure, with 40% of participants across all GMFCS levels having had at least 1 Botox procedure over the course of the study. Children classified as level IV had a significantly higher rate of Botox procedures than children in other levels.
In our study, the number of Botox injections was not meaningfully associated with either change in GMFM scores over a 1-year period or GMFM scores at the study's end point. Consistent with most of the literature in the field, a study by Flett and colleagues14 determined statistically significant improvements in spasticity following Botox injections within groups, when compared with baseline; however, they did not find a significant difference between groups in terms of motor function, as measured by the GMFM.12 It should be noted that spasticity is only one of many motor disorders associated with CP, and it is not realistic to think that addressing only 1 factor might have a significant effect on function; function is a complex phenomenon.15 Furthermore, a comprehensive systematic review concluded that most of the previous trials cover short observation periods ranging from 6 to 28 weeks only. Therefore, they are insufficient in capturing long-term effects of repeated Botox injections such as potential optimization of motor function and the prevention of contractures and secondary pain.16 In contrast to negligible results relating to Botox, the total number of procedures experienced by young children was negatively associated with motor function, a result that is not unexpected.
The nonsignificant results of associations between frequency of botox injections and the number of procedures and change in GMFM are not explained by the effect sizes obtained in this study. The minimal detectable change of the GMFM-66-B&C has been estimated to be 3.63,11 and we obtained a larger change score in this study (ie, 5.4). We believe that difficulties in explaining or predicting change relate to the fact that developmental change is nonlinear and complex17; in retrospect, linear statistical methods such as the Pearson product moment correlation are not compatible with the nonlinearities of change processes.
This study has several limitations. The first limitation is that a firm diagnosis of CP was not made for each participant. Second, these results cannot be generalized to all young children with CP. Although our initial sample was representative of children across GMFCS levels, we did lose some families through attrition. Families who either withdrew from the study or were lost to follow-up had mothers who were younger and had lower levels of education. Furthermore, not all children in the sample had reached 6 years of age at the end of the study and so the frequencies we report are an underestimate of what children might experience to age 6 years. Another limitation is the lack of a control group, which restricts readers from interpreting the meaning of the results relating to the unique health experiences of raising a child with CP compared with a child without CP. Finally, there is the possibility for a parental recall bias due to the data on medical and surgical procedures being collected retrospectively throughout these young children's early lives. Nonetheless, findings supporting parental recall of the injuries sustained by their young children mitigate this limitation.18 This study provides parents and health care providers with information regarding common procedures experienced by young children with CP and how one common procedure (ie, Botox) and the total number of procedures are related to early motor function. Regardless of functional ability level, parents can expect 2 to 3 medical or surgical procedures during early childhood. The significant variability across children of all GMFCS levels highlights the importance of individualized care for each child with a diagnosis of CP. The results serve to provide an initial step for researchers to look further into the prevalence of these procedures and subsequent long-term consequences. Health care providers are also able to use these results to allocate resources toward children at particular GMFCS levels.
The GMFCS classification of a child with CP affects the whole family. The family shares concerns and priorities, while health care providers deliver material to make informed decisions about services and supports for the best outcomes.19 It is also the responsibility of health care providers to assist with preparing the family for key periods in their children's lives.19 This article serves to provide families with information regarding common medical and surgical procedures across all GMFCS levels. Family-centered care is the reference standard for pediatric rehabilitation care, and this study fits with this model20 by providing information important to families of young children with CP.
This study is the first to describe variations in medical and surgical procedures for children with CP across GMFCS levels. On average, parents can expect 2 to 3 procedures for all young children with CP. The most prevalent procedures include Botox injections; ear, nose, and throat procedures; gastrostomy tube insertion and/or revisions; eye procedures; gastrointestinal tract procedures; shunt insertions and/or revisions; and respiratory system procedures, all of which differed by GMFCS level. These procedures did not differ by geographical region. Finally, although Botox injections are known to reduce spasticity and potentially control contractures, its long-term positive influence on motor function is not supported by this study. The total number of procedures does negatively affect motor function. This information is useful for therapists and parents when planning comprehensive services for young children with CP by assisting parents in understanding potential future medical and surgical interventions based on the GMFCS level of their children. Future research should elaborate on these preliminary data by following children with CP through school age and adolescence.
1. Rosenbaum P, Paneth N, Leviton A, Goldstein M, Bax M. A report: the definition and classification of cerebral palsy
. Dev Med Child Neurol. 2007;49:8–14.
2. Palisano R, Rosenbaum P, Walter S, Russell S, Wood E, Galuppi B. Development and reliability of a system to classify gross motor function in children with cerebral palsy
. Dev Med Child Neurol. 1997;39:214–223.
3. Palisano R, Rosenbaum P, Bartlett D, Livingston M. Gross Motor Function Classification System expanded and revised. Dev Med Child Neurol. 2008;50:744–750.
4. Wood E, Rosenbaum P. The Gross Motor Function Classification System for cerebral palsy
: a study of reliability and stability over time. Dev Med Child Neurol. 2000;42:292–296.
5. Gajdosik CG, Cicirello N. Secondary conditions of the musculoskeletal system in adolescents and adults with cerebral palsy
. Phys Occup Ther Pediatr. 2001;21(4):49–68.
6. Kondratek M, McCollum H, Garland A. Long-term physical therapy management following a single-event multiple level surgery. Pediatr Phys Ther. 2010;22:427–438.
7. Wong C, Bartlett DJ, Chiarello LA, Chang HJ, Stoskopf B. Comparison of the prevalence and impact of health problems of preschool children with or without cerebral palsy
[published online ahead of print March 28, 2011]. Child Care Health Dev. DOI: 10.1111/j.1365-2214.2011.01233.x.
8. Bartlett D, Chiarello L, Westcott McCoy S, et al. The Move & PLAY study: an example of comprehensive rehabilitation outcomes research. Phys Ther. 2010;90:1660–1672.
9. Rosenbaum PL, Walter SD, Hanna SE, et al. Prognosis for gross motor function in cerebral palsy
: creation of motor development curves. JAMA. 2002;288:1357–1363.
10. Gray L, Ng H, Bartlett DJ. The Gross Motor Function Classification System: an update on impact and clinical utility. Pediatr Phys Ther. 2010;22:315–320.
11. Brunton LK, Bartlett DJ. Validity and reliability of two abbreviated versions of the Gross Motor Function Measure. Phys Ther. 2011;91:577–588.
12. Russell DJ, Rosenbaum PL, Avery LM, Lane M. Gross Motor Function Measure (GMFM-66 and GMFM-88) User's Manual. London, UK: Mac Keith Press; 2002.
13. Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Hillsdale, NJ: Lawrence Erlbaum Associates; 1988.
14. Flett PJ, Stern LM, Waddy H. Botulinum toxin A versus fixed cast stretching for dynamic calf tightness in cerebral palsy
. J Paediatr Child Health. 1999;35:71–77.
15. Wright FV, Rosenbaum PL, Goldsmith CH, Law M, Fehlings DL. How do changes in body functions and structures, activity, and participation relate in children with cerebral palsy
? Dev Med Child Neurol. 2008;50:283–289.
16. Lukban MB, Rosales RL, Dressler D. Effectiveness of botulinum toxin A for upper and lower limb spasticity in children with cerebral palsy
: a summary of evidence. J Neural Transm. 2009;116:319–331.
17. Thelen E, Smith LB. A Dynamic Systems Approach to the Development of Cognition and Action. Cambridge, MA: MIT Press; 1994.
18. Cummings P, Rivara FP, Thompson RS, Reid RJ. Ability of parents to recall the injuries of their young children. Inj Prev. 2005;11:43–47.
19. King G, Tucker MA, Baldwin P, Lowry K, LaPorta J, Martens L. A life needs model of pediatric service delivery: services to support community participation and quality of life for children and youth with disabilities. Phys Occup Ther Pediatr. 2002;22:53–77.
20. Palisano RJ, Almarsi N, Chiarello L, Orlin MN, Bagley A, Maggs J. Family needs of parents of children and youth with cerebral palsy
. Child Care Health Dev. 2010;36:85–92.
Questions Asked About Medical or Surgical Procedures or Hospitalizations
At the beginning of the 1-year study, parents were asked the following questions. Those responding “yes” were asked to indicate if the procedure was “within the last year” or “longer than a year ago.”
- Has your child had neurological surgeries?
- Brain surgery, other than shunt?
- Dorsal Rhizotomy (spinal surgery to reduce spasticity)?
- Other neurological surgery? (if yes, what kind?)
- Has your child had orthopedic surgeries and procedures? “Orthopedic” means surgery on bones, muscles, and/or tendons.
- Surgery on muscles and/or tendons?
- Surgery on bones?
- Has your child had any other surgeries? (If yes, what kind? 2 opportunities provided)
- Other than for surgery, has your child had other hospitalizations in the past year? (If yes, provide the reasons for hospitalization. 2 opportunities provided)
- Has your child had any procedures for spasticity management-–other than dorsal rhizotomy?
- Has your child ever had a baclofen pump? (If yes, is it still in use?)
- Has your child ever had Botox injections?
- Has your child had any other procedures for spasticity management?
- Does your child use any medical technology, such as a G-Tube, tracheostomy tube, or ventilator? (check all that apply)
One year later, parents were asked:
- Has your child had any surgeries since the first study visit, which was in [insert month and year]? (If yes, what kind of surgery? 2 opportunities provided).
- Other than for surgeries, has this child had other hospitalizations since the first study visit? (If yes, provide the reasons for the hospitalizations. 2 opportunities provided).
- Has this child had any procedures for spasticity management since the first study visit? Some examples of spasticity management include oral baclofen or Botox injections. (If yes, specify the procedure(s). 2 opportunities provided).
Keywords:© 2012 Lippincott Williams & Wilkins, Inc.
cerebral palsy; geographic factors; medical procedures; physical activity; surgical procedures