PEDIATRICS Vol. 121 No. 1 January 2008, pp. e187-e192 (doi:10.1542/10.1542/peds.2007-0068)
ARTICLE |
Probability of Walking in Children With Cerebral Palsy in Europe
a Queen Silvia Children's Hospital/Sahlgrenska, University of Göteborg, Göteborg, Sweden
b Department of Physiotherapy, Institution of Neuroscience and Physiology, Sahlgrenska Academy, Göteborg University, Göteborg, Sweden
c National Institute of Public Health, Rigshospitalet, Copenhagen, Denmark
d Registre des Handicaps de l'Enfant et Observatoire Périnatal-Technologies de l'Imagerie de la Modélisation et de la Cognition, Pavillion Taillefer, Grenoble, France (RHEOP-TIMC-ThEMAS, CHU) CHU de Grenoble, Grenoble, France
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
OBJECTIVES. The purpose of this work was to describe walking ability in children with cerebral palsy from the Surveillance of Cerebral Palsy in Europe common database through 21 years and to examine the association between walking ability and predicting factors.
PATIENTS AND METHODS. Anonymous data on 10042 children with cerebral palsy born between 1976 and 1996 were gathered from 14 European centers; 9012 patients were eligible for the analyses.
RESULTS. Unaided walking as the primary way of walking at 5 years of age was reported for 54%, walking with assistive devices was reported for 16%, and no walking ability was reported for 30%. The proportion of children who were unable to walk was rather stable over time in all of the centers, with a mean proportion of 28%. Walking ability related significantly to cerebral palsy types, that is, spastic unilateral, spastic bilateral, dyskinetic, and ataxic cerebral palsy, as well as to IQ level, active epilepsy, and severe visual and hearing impairment. Severe cerebral palsy, defined as both the inability to walk and an IQ of <50, was present in 20% of the subjects. Logistic regression revealed that intellectual capacity was the variable most associated with walking ability in all 4 of the cerebral palsy types. The presence of a severe intellectual impairment increased the risk of being unable to walk 56 times if the child had unilateral spastic cerebral palsy type and 9 times if the child had bilateral spastic cerebral palsy type.
CONCLUSIONS. The collaboration Surveillance of Cerebral Palsy in Europe provides a powerful means of monitoring trends in cerebral palsy and its functional consequences. The proportion of nonwalking in children with cerebral palsy seems to be rather stable over years and across centers despite the changes that have occurred in neonatal care across Europe. As is well known and also shown in this study, walking ability varied strongly with cerebral palsy type. Additional impairments, as well as the presence of epilepsy, correlated significantly with walking ability and, thus, the walking ability can be an indicator of total disability load.
Key Words: children • epidemiology • cerebral palsy • gross motor function • impairments
Abbreviations: CP—cerebral palsy • SCPE—Surveillance of Cerebral Palsy in Europe • BW—birth weight • GA—gestational age • GMFCS—Gross Motor Function Classification System
Parents of children with cerebral palsy (CP) always want professionals to predict whether the child will walk. In 1998, a collaborative network of CP registers and surveys in 14 centers in 8 countries across Europe was formed, called Surveillance of Cerebral Palsy in Europe (SCPE).1 The aim was to develop a common database of children with CP to monitor trends, to provide information for service planning, and to provide a framework for collaborative research. In 2003, 2 more centers were included (Table 1). Data on children with CP have been gathered from the birth years 1976–1996, now composed of >10000 children.1, 2 Consensus on a definition and description of the different CP types has been reached and a reference and training manual for standardized assessment has been developed.3 Children with CP often have accompanying problems, which have been acknowledged recently,4 but the motor problem is still the core characteristics in CP. In the SCPE publication of 2002, walking ability and types of CP were described (including spastic unilateral, spastic bilateral, dyskinetic, and ataxic).2 On average, 50% of the children were able to walk, 20% were walking with aids, and 30% were unable to walk. Practically all of the children with unilateral spastic CP and ataxic CP were able to walk. One of 5 children with CP was found to have a severe intellectual deficit and was unable to walk. The aim of this report was to describe walking ability in children with CP from the SCPE database through 21 years and to examine the association between walking ability and predicting factors.
|
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
CP was defined according to the SCPE as a group of disorders, that is, it is an umbrella term; it is permanent but not unchanging; it involves a disorder of movement and/or posture and of motor function; it is a result of a nonprogressive interference, lesion, or abnormality; and this interference, lesion, or abnormality is in the developing or immature brain. Case subjects were included if they fulfilled the clinical criteria after their fourth birthday. Anonymous data on children with CP were sent to the coordinating center in Grenoble (Table 1). The standardized data set formed by the SCPE was used, which included a total of 50 items. The items used in this study were the center, birth year, CP type, walking ability, intellectual impairment, epilepsy, visual impairment, hearing impairment, birth weight (BW), and gestational age (GA). CP type was divided into spastic unilateral, spastic bilateral, dyskinetic, and ataxic as defined in the SCPE. Walking ability was the primary way of walking at 5 years of age graded as (1) unaided walking, (2) walking with aids, or (3) unable to walk. Intellectual impairment was graded as an (1) IQ of
85 or normal schooling, (2) IQ of 84 to 50, or (3) IQ of <50. Epilepsy was graded as (1) no active epilepsy or (2) active epilepsy (seizures the last year or antiepileptic treatment). Visual impairment was graded as (1) no severe visual impairment or (2) severe visual impairment (0.3 visual acuteness on the better eye, after correction). Hearing impairment was graded as (1) no severe hearing impairment or (2) severe hearing impairment (loss of >70 dB). For the logistic regression, all of the included variables were graded binary as the presence or absence of each indicator. The database was composed of 10042 subjects born between 1976 and 1996. Centers 10 and 14 composed of 347 subjects had not reported all 4 of the CP types and, thus, were excluded from this analysis. Additional exclusions were 323 subjects with unknown CP type and 360 subjects with missing information on walking ability, leaving 9012 eligible for analyses.
Statistical analyses were performed with StatView software 5.0 (SAS Institute, Cary, NC). The 
2 test was used for contingency tables with Bonferroni correction for paired comparisons. The Spearman rank correlation test was used for regression analyses. A P value of 
.005 was set for statistical significance. This threshold of the P value (ie, .005) was chosen to avoid nonrelevant significance of statistical results because of the large size of our population data. Logistic regression analyses were performed to identify variables associated with variations in walking ability.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Unilateral spastic CP was present in 2874 children (32%), bilateral spastic CP in 5144 (57%), dyskinetic CP in 613 (7%), and ataxic CP in 381 (4%). Unaided walking was reported for 4823 children (54%), walking with assistive devices in 1478 (16%), and no walking ability in 2711 (30%). Walking ability related significantly to CP type (P < .0001). In the unilateral spastic group, only 3% were nonwalking; in the ataxic group, 10% were nonwalking; in the bilateral spastic group, 43% were nonwalking; and in the dyskinetic group, 59% were nonwalking (Table 2).
|
The mean proportion of children unable to walk was 28%. There were significant differences across centers of this proportion (P < .0001) mainly because of different distributions of CP type (Table 3). In 4 centers (C02, C03, C09, and C11), the trend over time was not reflecting a stable rate. When excluding these centers, and after taking into account by adjustment the difference between the 10 remaining centers, the prevalence rate of children with CP who were unable to walk was stable over time during the whole studied period (1976–1996; Fig 1).
|
|
Intellectual capacity was reported for 7866 (87%) of the 9012 children. IQ was <50 in 2254 children (29%), between 50 and 84 in 2062 children (26%), and >85 in 3550 children (45%; Table 2). Walking ability was significantly related to IQ levels (r = 0.55; P < .0001; Table 4). Among children with an IQ of >85, 8% were unable to walk as compared with 71% of children with an IQ of <50 (Table 2).
|
Severe CP, defined as both the inability to walk and an IQ of <50, was present in 1607 of the 7866 children where intellectual capacity was known. Severe CP was most common in dyskinetic CP occurring in 200 (40%) of 496 subjects, followed by 1340 (30%) of 4489 subjects with bilateral spastic CP, 14 (4%) of 304 subjects with ataxic CP, and 53 (2%) of 2510 subjects with unilateral spastic CP.
Information on epilepsy was reported for 7563 of the 9012 children (84%). Active epilepsy was present in 1118 children (15%; Table 2). Walking ability was significantly related to epilepsy (r = 0.26; P < .0001; Table 4). Sixty percent of children with active epilepsy were unable to walk, as compared with 25% among those not having active epilepsy.
Visual capacity was reported for 8393 (93%) of the 9012 children. Severe visual impairment was present in 1009 children (12%; Table 2). Walking ability was significantly related to the presence of severe visual impairment (r = 0.37; P < .0001; Table 4). Of children with severe visual impairment, 79% were unable to walk, as compared with 23% among those not having visual impairment.
Hearing capacity was reported for 8141 (90%) of the 9012 children. Severe hearing impairment (loss of >70 dB) was reported for 214 children (3%; Table 2). Walking ability was significantly related to severe hearing impairment (r = 0.09; P < .0001; Table 4). Fifty-four percent of children with severe hearing impairment were unable to walk as compared with 29% among those not having hearing impairment.
BW was reported for 8477 (94%)of the 9012 children. A BW of <1000 g was present in 445 (5%), 1000 to 1499 g in 1200 (14%), 1500 to 2499 in 2095 (25%), and 2500 g in 4737 (56%; Table 5). Although significant, the associations between walking ability and the BW group were very weak (r = –0.05; P < .0001; Table 4). No distinct trend could be seen.
|
GA was reported for 8427 (94%) of the 9012 children. A GA of <28 weeks was present in 514 children (6%), 28 to 31 weeks in 1489 (18%), 32 to 36 weeks in 1621 (19%), and
36 weeks in 4803 (57%; Table 5). Walking ability varied significantly with GA (r = –0.03; P < .0001; Table 4). Unaided walking in the children with a GA of >36 weeks was significantly more frequent and walking with aids less frequent than in the other GA groups (P < .0001). Among the children unable to walk, there was no significant difference between the groups (P = .13; Table 5). Inability to walk was present in 2068 children with known IQ levels and GA groups (Fig 2). Severe CP (inability to walk and IQ of <50) was predominantly found in the bilateral spastic group, composed of 83% (1226 of 1471) as compared with 13% in the dyskinetic CP and 3% and 1% in the unilateral spastic and ataxic groups, respectively.
|
Logistic regression analysis on the association between walking ability and 5 independent variables (IQ of <50, severe visual impairment, active epilepsy, GA of <34 weeks, and BW of <2500 g) was performed on a total of 5872 children where complete information on all of the variables was available. Severe hearing impairment was omitted, because it rather decreased the R2. The results by CP types are shown in Table 6. Intellectual capacity was the variable most associated with walking ability in all 4 of the CP types. The presence of a severe intellectual impairment increased the risk of being unable to walk 56 times if the child had a unilateral spastic CP type, whereas this increase was 9 times if the child had a bilateral spastic CP type. Severe visual impairment and active epilepsy remained associated variables in the unilateral, bilateral spastic, and dyskinetic CP types. GA and BW were not significantly associated variables, but their inclusion in the analyses increased the R2 marginally.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This collaborative work from SCPE provides a powerful means of monitoring trends in CP and its functional consequences. Although there was not an operational definition of walking ability between centers, data were collected during a period of constant consensus discussions, and the data set was revised from the birth year period 1990–1996. According to the revised data set, walking ability was categorized into 7 different levels, that is, unaided, without restrictions, without assistive devices, with mobility devices, unable to walk even with aids, self-mobility with limitations, and self-mobility severely limited. It was not possible to transpose all of these data from the old to the new version, and detailed information was lost in the need for simplification of alternative answers. In this study, walking ability was, therefore, grouped into 3 levels.
More than half of the children were able to walk without aids, another 16% needed assistive or mobility devices, and one third were unable to walk. These findings are compatible with the review by Koman et al,5 who reported that >50% of children with CP in the 8- to 17-year age band could walk without aid and 25% were unable to walk, and with that of Himmelmann et al,6 who reported 30% who were unable to walk in a population-based study of children 4 to 8 years with CP.
Walking ability may alter as the children get older. The present study was a cross-sectional study, and the children were included if they fulfilled the clinical criteria for CP after their fourth birthday at the time for data collection. Some children can achieve independent walking at the age of 7 to 8 years, an age where the peak motor performance occurs in children with CP according to the motor growth curves created by Rosenbaum et al.7 Some children will lose their walking ability as they grow. In a Norwegian study of adults with CP, 70% of the hemiplegic group and 43% of the diplegic group self-reported the walking ability to be improved or unchanged (Jahnsen et al8). Deterioration in walking ability in adulthood is common in the whole group because of pain, fatigue, and lack of adapted physical activity (Jahnsen et al8). Andersson et al reported unchanged walking ability in half of the people with hemiplegia and in one third of the participants with diplegia at the age of 20 to 58 years compared with the childhood capacity (Andersson and Mattsson9).
As is well known and also shown in the present study, walking ability varied strongly with CP type. Unilateral and ataxic CP types have the best and dyskinetic and bilateral CP types the least walking ability. In addition, intellectual, visual, and hearing impairments, as well as the presence of epilepsy, correlated significantly with walking ability, and, thus, walking ability can be considered an indicator of total disability load. This has been reported also from the register in western Sweden.6, 10 Severe CP was present in 20%, predominantly in children with dyskinetic and bilateral spastic CP. Severe CP in the bilateral spastic group born at term coincides well with the subtype of tetraplegic CP used in other studies.6
A correlation of –0.05 or 0.09 may be statistically significant without explaining the variance. To get more information on factors predictive of walking ability, a logistic regression analysis was performed separately for the different CP types. Intellectual impairment was the most prominent explanatory factor of walking ability in all of the CP types. Within this cross-sectional study, we were able to show associations but not able to suggest that intellectual impairment was responsible for the incapacity to walk. The intellectual impairment could have its origin at the same time and for the same reason as the brain lesion responsible for the motor deficit.
Epilepsy and visual impairment were additional associated factors in spastic and dyskinetic CP. GA and BW were not statistically associated with walking ability but increased the R2 marginally. Himmelmann et al6 have shown that the proportion of children with CP and with the mildest motor impairment increased significantly with GA.
The proportion of nonwalking in children with CP seems to be rather stable over time and across 10 of the 14 centers despite the changes that have occurred in neonatal care across Europe. Differences between the centers in the distribution of walking ability were mainly because of differences in the distribution of CP types. Although this study was performed in well-defined populations across Europe, local differences may well occur. The differences may also be because of problems in the validity of the classifications. At the time of data collection, the Gross Motor Function Classification System (GMFCS),11 a valid, reliable, and widely used classification of gross motor function, was not introduced into the SCPE data set. The reported levels could not reliable be transposed into 5 GMFCS levels, because GMFCS is a classification of gross motor function and not just walking ability, and the data needed for the GMFCS were not available in the data set. Prognosis of walking has been described according to the 5 GMFCS levels.7 In the future, a more detailed picture of the motor function and the probability of walking in children with CP can be presented using the decision tree,1 the GMFCS and the Bimanual Fine Motor Function classification,6 or the Manual Ability Classification System.12
|
ACKNOWLEDGMENTS |
|---|
This work was supported by grants from the European Commission (Research DG-QLG5-2001-30133 and DGSANCO-2003-3131).
|
FOOTNOTES |
|---|
Accepted Jun 21, 2007.
Address correspondence to Eva Beckung, PT, PhD, Department of Physiotherapy, Sahlgrenska University Hospital, Hus A4, SE 431 80 Mölndal, Sweden. E-mail: eva.beckung{at}vgregion.se
The authors have indicated they have no financial relationships relevant to this article to disclose.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
- Surveillance of Cerebral Palsy in Europe (SCPE). A collaboration of cerebral palsy registers. Dev Med Child Neurol. 2000;42 :816 –824[CrossRef][Web of Science][Medline]
- Surveillance of Cerebral Palsy in Europe (SCPE). Prevalence and characteristics of children with cerebral palsy in Europe. Dev Med Child Neurol. 2002;44 :633 –640[CrossRef][Web of Science][Medline]
- Mann K. Surveillance of Cerebral Palsy in Europe, Reference and Training Manual CD. University Children's Hospital; 2006
- Rosenbaum P, Paneth N, Leviton A, et al. A report: the definition and classification of cerebral palsy April 2006 [published correction appears in Dev Med Child Neurol. 2007;49(suppl):480]. Dev Med Child Neurol. 2007;109 :8 –14
- Koman A, Paterson Smith B, Shilt J. Cerebral palsy. Lancet. 2004;363 :1619 –1631[CrossRef][Web of Science][Medline]
- Himmelmann K, Beckung E, Hagberg G, Uvebrant P. Gross and fine motor function and accompanying impairments in cerebral palsy. Dev Med Child Neurol. 2006;48 :417 –423[CrossRef][Web of Science][Medline]
- Rosenbaum P, Walter S, Hanna S, et al. Prognosis for gross motor function in cerebral palsy: creation of motor developmental curves.
JAMA. 2002;288
:1357
–1363
[Abstract/Free Full Text] - Jahnsen R, Villien L, Egeland T, Stanghelle JK, Holm I. Locomotion skills in persons with cerebral palsy.
Clin Rehabil. 2004;18
:309
–316
[Abstract/Free Full Text] - Andersson C, Mattsson E. Adults with cerebral palsy: a survey describing problems, needs, and resources, with special emphasis on locomotion. Dev Med Child Neurol. 2001;43 :76 –82[CrossRef][Web of Science][Medline]
- Beckung E, Hagberg G. Neuroimpairments, activity limitations and participation restrictions in children with cerebral palsy. Dev Med Child Neurol. 2002;44 :309 –316[CrossRef][Web of Science][Medline]
- Palisano R, Rosenbaum P, Walter S, Russell D, 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[Web of Science][Medline]
- Eliasson A-C, Rösblad B, Krumlinde Sundholm L, Beckung E, Arner M, Rosenbaum P. The Manual Ability Classification System for Children With Cerebral Palsy: scale development and evidence of validity and reliability. Dev Med Child Neurol. 2006;48 :549 –554[CrossRef][Web of Science][Medline]
PEDIATRICS (ISSN 1098-4275). ©2008 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
