Research in Developmental Disabilities 32 (2011) 2909–2915

Contents lists available at ScienceDirect

Research in Developmental Disabilities

Classification of topographical pattern of spasticity in cerebral palsy:

A registry perspective
Susan M. Reid a,b,d,*, John B. Carlin a,c, Dinah S. Reddihough a,b,d
a
Department of Paediatrics, University of Melbourne, Parkville, Victoria 3052, Australia
b
Developmental Disability Research, Murdoch Childrens Research Institute, Flemington Road, Parkville 3052, Australia
c
Clinical Epidemiology and Biostatistics Unit, Murdoch Childrens Research Institute, Flemington Road, Parkville 3052, Australia
d
Developmental Medicine, Royal Children’s Hospital, Flemington Road, Parkville 3052, Australia

A R T I C L E I N F O A B S T R A C T

Article history: This study used data from a population-based cerebral palsy (CP) registry and systematic
Received 20 April 2011 review to assess the amount of heterogeneity between registries in topographical patterns
Accepted 5 May 2011 when dichotomised into unilateral (USCP) and bilateral spastic CP (BSCP), and whether the
Available online 31 May 2011 terms diplegia and quadriplegia provide useful additional epidemiological information.
From the Victorian CP Register, 2956 individuals (1658 males, 1298 females), born 1970–
Keywords: 2003, with spastic CP were identified. The proportions with each topographical pattern
Cerebral palsy were analysed overall and by gestational age. Binary logistic regression analysis was used
Classification
to assess temporal trends. For the review, data were systematically collected on
Population surveillance
topographical patterns from 27 registries. Estimates of heterogeneity were obtained,
Quadriplegia
Registries
overall and by region, reporting period and definition of quadriplegia. Among individuals
born <32 weeks, 48% had diplegia, whereas the proportion for children born 32 weeks
was 24% (p < 0.001). Evidence was weak for a temporal trend in the relative proportions of
USCP and BSCP (p = 0.038), but much clearer for an increase in the proportion of spastic
diplegia relative to quadriplegia (p < 0.001). The review revealed wide variations across
studies in the proportion of diplegia (range 34–90%) and BSCP (range 51–86%). These
findings argue against a topographical classification based solely on laterality.
ß 2011 Elsevier Ltd. All rights reserved.

1. Introduction

To describe spastic motor types in cerebral palsy (CP), three commonly occurring topographical patterns have been
used – quadriplegia or tetraplegia, diplegia, and hemiplegia. Some topographical classifications have additional terms,
such as monoplegia and triplegia to describe the involvement of one or three limbs respectively, right and left to
describe the side of the hemiplegia, and an assessment of asymmetry (Stanley, Blair, & Alberman, 2000). Motor
problems affecting the head and trunk are less often included. Although there has been growing awareness of the lack of
precision and consistency in the use of these descriptors, only one study tested their reliability, finding that there was
only 50% percent agreement on topographical pattern among a group of six experienced neurodevelopmental clinicians
(Blair & Stanley, 1985). There have been recent debates about the advantages and disadvantages of abandoning the
terms diplegia and quadriplegia altogether (Colver & Sethumadhavan, 2003; Hurvitz & Brown, 2010; Rosenbaum,

* Corresponding author at: Developmental Medicine, Royal Children’s Hospital, Flemington Road, Parkville, Victoria 3052, Australia.
Tel.: +61 393454807; fax: +61 393457686.
E-mail address: sue.reid@mcri.edu.au (S.M. Reid).

0891-4222/$ – see front matter ß 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ridd.2011.05.012
2910 S.M. Reid et al. / Research in Developmental Disabilities 32 (2011) 2909–2915

Gorter, Palisano, & Morris, 2010; Shevell, 2010; Shevell, Dagenais, & Hall, 2009). Some of the impetus for these debates
has come from research initiated by European CP registries as part of the collaborative process for forming the
Surveillance of Cerebral Palsy in Europe (SCPE). To allow data to be harmonized, SCPE adopted a simpler approach
whereby the pattern of spasticity is solely described as bilateral or unilateral (Surveillance of Cerebral Palsy in Europe
(SCPE), 2000). The basis for classification is a question about whether or not the limbs on both sides of the body are
involved. A study of the validity and reliability of the SCPE guidelines using case vignettes failed to achieve good
agreement between raters on the type and laterality of motor involvement (Gainsborough, Surman, Maestri, Colver, &
Cans, 2008). This was assumed to be largely due to individual or local differences in the interpretation of clinical
information and suggests a need for more clearly defined criteria to reliably distinguish between topographical patterns.
The 2004 International Workshop on Definition and Classification of Cerebral Palsy recommended that the trunk,
each limb and oropharynx be described individually in terms of any impairments of movement or posture, and that the
terms diplegia and quadriplegia should either be well defined or not used at all until more precise terminology evolves
and gains acceptance (Rosenbaum, Paneth, Leviton, Goldstein, & Bax, 2007). The members favoured the approach taken
by SCPE, but acknowledged that even the distinction between unilateral and bilateral spasticity can still be blurred since
many children with primarily unilateral CP also have some degree of contralateral motor involvement, and some
children with bilateral involvement have appreciable asymmetry across sides. Some commentators have suggested that
motor type and the Gross Motor Function Classification System (GMFCS) (Palisano et al., 1997) alone are sufficient to
describe CP (Hurvitz & Brown, 2010), an argument supported by claims that the GMFCS has a greater ability to explain
differences in motor development than topographical pattern or motor type (Gorter et al., 2004). On the other hand,
arguments have been offered in favour of retention of the old classification system. One line of reasoning has been that
CP subtype based on a standard neurological examination is often assigned at an earlier age than GMFCS and therefore
has some usefulness in informing the experienced clinician about possible aetiologies, pathologies, prognosis, and
management (Shevell et al., 2009b). A further argument in favour of retaining a topographical classification came from
the Functional Assessment Research Group in the United States who reported that classification of topographic pattern
into diplegia and hemiplegia differentiated between independently ambulant children classified at the same GMFCS
level in various aspects of motor function, activity, participation, and quality of life (Damiano et al., 2006).
This study aimed to further examine the issue of classification by topographical pattern of spasticity from the CP registry
perspective and to answer two broad questions. How much heterogeneity exists between CP registries in the relative
proportions of unilateral and bilateral spastic CP and, for individuals with bilateral spastic CP, is the additional information
provided by a classification that discriminates on the basis of the degree of upper limb involvement useful for improving our
understanding of CP epidemiology? Data from Victoria, Australia, and data obtained from a systematic review of the
international literature were used. The specific aims of the Victorian study were to describe the proportion of CP cases with
each topographical pattern of spasticity in a large population sample of children with CP born in Victoria between 1970 and
2003, to explore any association with gestational age, and to investigate trends over time in the proportion of CP cases with
each topographical pattern. The aim of the systematic review was to assess the degree of heterogeneity for topographical
patterns reported by population-based CP registries in developed countries, and identify regional, definitional or temporal
factors that affect the degree of heterogeneity.

2. Method

2.1. Victorian study

Established in the mid-1980s, the Victorian Cerebral Palsy Register (VCPR) registers all CP cases born or resident in
the Australian state of Victoria from 1970. On average, the VCPR ascertains 100 new CP cases per year from a
denominator population of approximately 65 000 live births, an estimated birth prevalence of 1.8 per 1000 live births.
Since its inception, the VCPR has identified new cases through active surveillance of inpatient and outpatient lists at the
two tertiary paediatric hospitals in Victoria, through direct referral from clinicians, and by self-referral as a result of
promotion of the Register through service providers, brochures, and websites. To be registered with the VCPR,
individuals are required to have a disorder of movement and/or posture that is the result of a non-progressive
abnormality in the developing brain up to the age of two years. Once identified, data pertaining to each individual are
collected from hospital medical records, developmental paediatricians, and families. Birth data are collected from the
Victorian Perinatal Data Collection Unit. The diagnosis and clinical details are confirmed at the age of five years from the
same sources. Ongoing reconfirmation of diagnosis is undertaken on a regular basis and entries are deleted if individuals
no longer have neurological signs or impairments or if they are subsequently found to have a progressive disorder.
For the analysis, monoplegia was combined with hemiplegia, and triplegia with quadriplegia. A category of bilateral
spastic CP was formed by grouping diplegia with quadriplegia. The proportion of CP cases with each topographical
pattern of spasticity was calculated and the proportions were tabulated by grouped gestational age at delivery. To assess
trends over time, the data were assessed visually, and binary logistic regression analyses were used to assess the average
change in the odds of unilateral versus bilateral spastic CP for each year of birth, and in the odds of a diplegic versus a
quadriplegic pattern of bilateral spastic CP. Statistical analysis was performed in Stata 11 (StataCorp, 2009, College
Station, Texas, USA).
 S.M. Reid et al. / Research in Developmental Disabilities 32 (2011) 2909–2915 2911

Identification
356 articles identified 4 additional articles identified
through MEDLINE search through other sources

360 articles after duplicates removed

Screening

360 articles screened 330 articles excluded

30 articles assessed 9 articles excluded:

Eligibility

for eligibility duplicate registry (8)

>20% missing data (1)

21 articles included in
qualitative synthesis
Included

27 studies included in
qualititative synthesis

Fig. 1. Flow diagram of the process used to select articles and studies.

2.2. Systematic review

Searches were performed through MEDLINE for full text articles published in English between 1980 and 2010. The search
term used was ‘cerebral palsy/epidemiology’ as a major subject heading. Titles and abstracts were first screened for evidence
that the data originated from a population based registry in a developed country, included all types of CP, and contained
information on CP subtypes. Manuscripts were then screened for data on topographical patterns of spasticity. Published
registry reports and bibliographies of published papers and review articles were searched for additional material. A flow
diagram of the process used to select studies is shown in Fig. 1.
Using a standardised form, information was collected by the first author on each registry and each publication, including
the classification system and definitions used to describe topographical patterns. The most recent and/or largest dataset was
chosen where multiple publications were available, or, where serial reports were obtained, data were amalgamated to cover
a greater time period. To minimise bias, a CP prevalence of at least 1.5 cases per 1000 births and no more than 20 percent
missing data were set a priori as requirements for inclusion.
For studies fulfilling the inclusion criteria, data on the total number of cases, the number with a spastic motor type, and
the number with each topographical pattern of spasticity were retrieved. The frequencies of unilateral and bilateral CP were
calculated as a proportion of all spastic CP, and the frequencies of diplegia and quadriplegia were calculated as a proportion
of bilateral spastic CP, excluding missing data. Where applicable, triplegia was combined with quadriplegia and monoplegia
with hemiplegia.

Table 1
Topographical pattern of spasticity as a proportion of all spastic CP by gestational age group, Victoria, 1970–2003.

Gestational age group Topographical pattern of spasticity Total

Hemiplegia Diplegia Quadriplegia

20–27 weeks, n (%) 71 (28.7) 110 (44.5) 66 (26.7) 247 (100.0)

28–31 weeks, n (%) 99 (22.5) 222 (50.3) 120 (27.2) 441 (100.0)
32–36 weeks, n (%) 143 (30.2) 166 (35.0) 165 (34.8) 474 (100.0)
37+ weeks, n (%) 622 (38.4) 343 (21.0) 666 (40.8) 1631 (100.0)
Unknown gestation, n (%) 73 (44.8) 28 (17.2) 62 (38.0) 163 (100.0)
All gestations, n (%) 1008 (34.1) 869 (29.4) 1079 (36.5) 2956 (100.0)
2912 S.M. Reid et al. / Research in Developmental Disabilities 32 (2011) 2909–2915

The standard error (SE) and 95% confidence intervals around each proportion were calculated for each study. A measure of
the heterogeneity of each proportion was obtained using a quantity called I-squared that estimates the percentage of total
variation across studies due to heterogeneity rather than chance alone. Values for I-squared lie between 0% and 100% with a
value of 0% indicating no observed heterogeneity and larger values indicating increasing heterogeneity (Higgins, Thompson,
Deeks, & Altman, 2003). Subgroup analyses and assessments of heterogeneity were performed for the proportion of CP cases
with each topographical pattern by geographic region, the middle year of the reporting period, and the definition used for
quadriplegia.

3. Results

3.1. Victorian study

For the Victorian study, 3353 individuals with CP born in Victoria between 1970 and 2003 and registered prior to June
2009 were identified from the VCPR, excluding individuals with a clear cause for their CP after day 28 of life. Of these, 3262
(97%) had data on motor type, and 2956 (91%) were classified as having a predominantly spastic motor type. Table 1 displays
the proportions of each topographical pattern for the total Victorian cohort (1658 males, 1298 females) and stratified by
gestational age at delivery. Among individuals with a spastic motor type who were born very preterm (<32 weeks), 25% had
a hemiplegic pattern, 48% a diplegic pattern, and 27% a quadriplegic pattern of spasticity, whereas the proportions for
children born later than 32 weeks’ gestation were 36%, 24%, and 39% respectively (p < 0.001).
Topographical patterns as a proportion of all spastic CP by decade of birth are tabulated in Table 2, and by year of birth in
Fig. 2, expressed as three-year moving averages. There was weak evidence that the proportion of unilateral relative to
bilateral spastic CP increased over time (logistic regression odds ratio per year of birth 1.008; 95% CI 1.000, 1.017, p = 0.038).
Within the group with bilateral spastic CP, there was much stronger evidence for an increase in the proportion of spastic
diplegia relative to quadriplegia over time (odds ratio 1.028; 95% CI 1.018, 1.037; p < 0.001).

3.2. Reviewed studies

Data on population proportions of topographical patterns of spasticity were obtained from 27 individual CP registries,
representing eight geographic regions (Table 3) (Andersen et al., 2008; Bottos, Granato, Allibrio, Gioachin, & Puato, 1999;
Hagberg, Hagberg, Beckung, & Uvebrant, 2001; Hagberg, Hagberg, & Olow, 1993; Hagberg, Hagberg, Olow, & von Wendt,
1996; Himmelmann, Hagberg, Beckung, Hagberg, & Uvebrant, 2005; Kavcic & Perat, 1998; MacGillivray & Campbell, 1995;
Mongan, Dunne, O’Nuallain, & Gaffney, 2006; Parkes, Dolk, Hill, & Pattenden, 2001; Peek et al., 2008; Pharoah, Platt, & Cooke,
1996; Riikonen, Raumavirta, Sinivuori, & Seppala, 1989; Shevell, Dagenais, & Hall, 2009; Sigurdardottir et al., 2008; Surman
et al., 2008; Surveillance of Cerebral Palsy in Europe (SCPE), 2002; Topp, Uldall, & Greisen, 2001; Watson, Blair, & Stanley,
2006; Westbom, Hagglund, & Nordmark, 2007; Yeargin-Allsopp et al., 2008). Nine studies reported solely on bilateral and
unilateral spastic CP while 18 reported separate data for hemiplegia, diplegia and quadriplegia. Two different definitions for
quadriplegia emerged. Four registries used the Swedish classification that defined quadriplegia as massive total motor
disability with four limbs severely involved, the upper limbs as much as lower. Five others defined quadriplegia as four limb

Table 2
Topographic pattern of spasticity as a proportion of all spastic CP cases in Victoria by decade of birth and gestational age at delivery.

Decade of birth

1970–1979 1980–1989 1990–1999 2000–2003

Spastic hemiplegic CP, n (%) 242 (33.6) 275 (32.1) 343 (34.8) 148 (37.9)
20–27 weeks 7 (1.0) 25 (2.9) 27 (2.7) 12 (3.1)
28–31 weeks 14 (1.9) 29 (3.4) 38 (3.9) 18 (4.6)
32–36 weeks 29 (4.0) 40 (4.7) 48 (4.9) 26 (6.6)
37+ weeks 144 (20.0) 163 (19.0) 225 (22.8) 90 (23.0)
Unknown gestation 48 (6.7) 18 (2.1) 5 (0.5) 2 (0.5)

Spastic diplegic CP, n (%) 180 (25.0) 236 (27.5) 318 (35.2) 135 (34.5)
20–27 weeks 5 (0.7) 18 (2.1) 65 (6.6) 22 (5.6)
28–31 weeks 34 (4.7) 70 (8.2) 79 (8.0) 39 (10.0)
32–36 weeks 52 (7.2) 50 (5.8) 41 (4.1) 23 (5.9)
37+ weeks 70 (9.7) 94 (11.0) 129 (13.1) 50 (12.8)
Unknown gestation 19 (2.6) 4 (0.5) 4 (0.4) 1 (0.2)

Spastic quadriplegic CP, n (%) 299 (41.5) 346 (40.4) 326 (33.0) 108 (27.6)
20–27 weeks 3 (0.4) 15 (1.7) 39 (3.9) 9 (2.3)
28–31 weeks 23 (3.2) 46 (5.4) 43 (4.4) 8 (2.0)
32–36 weeks 51 (7.1) 70 (8.2) 28 (2.8) 16 (4.1)
37+ weeks 176 (24.4) 199 (23.2) 216 (21.9) 75 (19.2)
Unknown gestation 46 (6.4) 16 (1.9) 0 (0.0) 0 (0.0)
All spastic CP, n (%) 721 (100.0) 857 (100.0) 987 (100.0) 391 (100.0)
 S.M. Reid et al. / Research in Developmental Disabilities 32 (2011) 2909–2915 2913

100

Percentage of all spastic CP

80

60

40

20 Hemiplegia
Diplegia
Quadriplegia
0
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
Fig. 2. Spastic hemiplegia, diplegia and quadriplegia as a proportion of all spastic CP, expressed as 3-year moving averages, Victoria, 1970–2003.

involvement with the upper limbs at least or more affected than the lower limbs. In addition, some European registers
employed an ataxic diplegia category which in publications was sometimes included with spastic diplegia, ataxia or ‘other’.
As a proportion of bilateral spastic CP, the range of estimates for quadriplegia across 18 studies was 10–66% (Table 3). The
proportions varied substantially between geographic regions (range 19–54%) and the degree of heterogeneity within each
region was high for Australia, United Kingdom and Northern Europe and moderate for North America. When the registries
were grouped according to the definition used for spastic quadriplegia, quadriplegia as a proportion of bilateral spastic CP for
those using the Swedish definition ranged from 13% to 33% compared with a range of 30–62% for registries where
quadriplegia was defined as the upper limbs at least as affected as the lower limbs. There were no significant differences in
the proportions or degree of heterogeneity based on the middle year of reporting period.

Table 3
Proportions of topographical patterns of spasticity from 27 CP registry studies.

CP registry Region Birth years Number Number spastic Spastic cases BSCPa as Quadriplegia
included CP cases CP cases as % all CP % spastic CP as % BSCP

Victoria, Australia Australia 1970–2003 3262 2956 90.6 65.9 55.4

South Australia Australia 1993–2002 374 355 94.9 63.1 44.2
Western Australia Australia 1970–1999 1549 1277 82.4 57.3 30.1
Mersey region, England United Kingdom 1966–1989 1599 1466 91.7 64.2 61.7
Oxford region, England United Kingdom 1984–2002 1219 1086 89.1 60.1 66.0
Avon, England United Kingdom 1969–1988 489 386 78.9 54.1 33.5
North of England United Kingdom 1980–1990 222 208 93.7 61.1 –
Scotland United Kingdom 1984–1990 683 545 79.8 66.4 –
Northern Ireland United Kingdom 1981–1993 724 694 95.9 58.4 37.5
West of Ireland Ireland 1990–1999 74 62 83.8 61.3 –
Southern Ireland Ireland 1980–1990 151 124 82.1 50.8 –
Netherlands Western Europe 1980–1988 82 80 97.6 58.8 –
Isere, France Western Europe 1980–1989 220 194 88.2 62.4 –
Haute-Garonne, France Western Europe 1980–1985 83 68 81.9 75.0 –
Denmark Northern Europe 1987–1999 297 253 85.2 76.3 18.1
Finland Northern Europe 1968–1982 112 84 75.0 57.1 10.4
Norway Northern Europe 1996–1998 273 239 87.5 60.2 –
Iceland Northern Europe 1985–2000 135 114 84.4 70.2 38.4
Western Sweden Northern Europe 1983–1998 772 649 84.1 58.9 14.1
Southern Sweden Northern Europe 1990–1997 341 253 74.2 58.9 12.8
Slovenia South East Europe 1981–1990 746 651 87.3 67.4 36.4
North East Italy Southern Europe 1965–1989 578 500 86.5 65.9 33.2
Central Italy Southern Europe 1980–1990 50 44 88.0 86.4 –
Wisconsin, USA North America 1994–1998 98 92 93.9 63.4 52.5
Alabama, USA North America 1994–1998 114 108 94.7 64.2 44.3
Georgia, USA North America 1994–1998 129 120 93.0 60.8 64.4
Quebec, Canada North America 1999–2002 243 218 89.7 62.8 62.0
Range, % 51–86 10–66
I-squared, % 80.0 98.4
Number of studies, n 27 18
a
Bilateral spastic cerebral palsy.
2914 S.M. Reid et al. / Research in Developmental Disabilities 32 (2011) 2909–2915

Summarising data from 27 studies on bilateral spastic CP as a proportion of all spastic CP, wide variation was seen in the
proportions relative to a unilateral spastic pattern, with a range of 51–86% (Table 3). Heterogeneity was moderately high
both between and within geographic regions, with the exception of North America where the four values ranged by less than
4% (60.8–64.2).

4. Discussion

This research supports the findings of previous studies in demonstrating that substantial heterogeneity exists between
population based registries in the proportions of bilateral spastic CP cases classified as either diplegia or quadriplegia.
Although our comparison of studies according to two different definitions for quadriplegia suggests that definitions may
explain some of the variation, heterogeneity remained high even when the same definition was used. The wide range of
proportions reported in this and other reviews is unlikely to reflect real differences in topographical patterns of spasticity,
rather it is probable that differing individual or local interpretations of the same definition account for much of the variation
(Colver & Sethumadhavan, 2003). This is not surprising, since guidance about differentiating between varying degrees of
upper limb involvement is missing from most definitions. Part of the confusion may also lie in the fact that, historically, the
terms have been used to describe clinical syndromes (Colver & Sethumadhavan, 2003). Syndromic definitions assume
certain causal pathways, imaging findings, severity levels or associations with intellectual or other impairments and there
are some individuals who do not neatly fit a particular pattern. Factors such as completeness of ascertainment and the
inclusion of postneonatally acquired or outborn cases may also have affected the proportions obtained from the reviewed
studies.
A widely favoured solution has been to combine the terms diplegia and quadriplegia into a single category of bilateral
spastic CP. Although this may be the best compromise to facilitate pooling of data from registries that previously used
different definitions, there is little evidence to suggest that this classification is more reliable. The distinction still relies on
subjective judgement about how much asymmetry is permitted in bilateral patterns. In this study, heterogeneity between
individual registries was still moderately high for the proportion of bilateral relative to unilateral spastic CP (range 51–86%).
In our opinion this is still an unacceptable degree of heterogeneity. Although Colver and Sethumadhavan reported almost the
same range of proportions for bilateral spastic CP (range 51–77%), their interpretation was that the variation between
countries had largely disappeared (Colver & Sethumadhavan, 2003). Our finding of minimal heterogeneity between the four
registries from North America, three of which are part of the Autism and Developmental Disabilities Monitoring (ADDM)
Network and use the same methodology and training for the staff, possibly suggest that reliability could be vastly improved if
consistent definitions and methodologies are used by individual registries.
Our premise in undertaking this study was that if associations are found between the degree of upper limb involvement
and aetiological, pathological, or other clinical factors, it would be important to persevere in developing a reliable
classification that describes the degree of upper limb involvement as well as laterality. To this end, we were able to confirm a
specific association between spastic diplegia and preterm birth, suggesting that topographical pattern has a part to play in
differentiating between important causal pathways to CP. We also showed that the relative proportion of cases classified as
diplegia increased in Victoria over the study period, a change that was less clearly seen for bilateral spastic CP. Most of this
increase occurred during the 1990s and 2000s and may reflect an increase in the number and proportion of children with CP
who were born preterm (paper in preparation). Since birth prior to 32 weeks’ gestation, is associated with spastic diplegia,
one would expect that the proportion of CP cases with spastic diplegia would increase in such circumstances. However, this
is not the whole explanation, as there was also an increase in the relative proportions of spastic diplegia and spastic
hemiplegia, and a corresponding decrease in the proportion of spastic quadriplegia, for children born at term. Data from
Western Australia also show a marked decline in the proportion of CP cases with spastic quadriplegia in the second half of the
1990s (personal communication).
There are valid arguments from both clinical and epidemiological perspectives to suggest that a topographical
classification is a useful and necessary adjunct to other classifications currently in common use. GMFCS on its own does not
distinguish children with CP for all characteristics of interest to clinicians and researchers, including imaging findings
(Robinson et al., 2008), aetiological factors, and comorbidities (Shevell et al., 2009a). We agree with the recommendations
from the 2004 Bethesda Workshop that the trunk, each limb and oropharynx be described individually in terms of
impairments of movement or posture to allow for separate descriptions of upper and lower limb involvement and an
assessment of laterality, irrespective of motor type and severity.

5. Conclusions

CP registries require consistency of methods, definitions and classification systems to facilitate investigations of trends,
aetiological associations, and comparisons with other geographical regions. To this end, categories for the classification of
topographical patterns need to be mutually exclusive and exhaustive (Rosenbaum et al., 2007), clinicians examining children
must make accurate and reproducible observations that are reported systematically, and coders, if used, must apply
guidelines and interpret recorded observations accurately (Gainsborough et al., 2008). For this to happen, we need a
common and detailed proforma for the recording of the results of the clinical examination and assessments of functional
performance and capacity, as well as clear guidelines for classification based on this information.
 S.M. Reid et al. / Research in Developmental Disabilities 32 (2011) 2909–2915 2915

Acknowledgements

The authors are grateful to Anna Lanigan and Tess Lionti who assisted with data collection for this study and the Victorian
Perinatal Data Collection Unit which provided birth data. We also acknowledge the support of the Victorian Medical
Insurance Agency Ltd, the name behind Professional Services Australia, and the Victorian Department of Health who
provided funding for the Victorian Cerebral Palsy Register. The first author’s doctoral research is supported by a scholarship
(2009–11) from the Australian National Health and Medical Research Council.

References

Andersen, G. L., Irgens, L. M., Haagaas, I., Skranes, J. S., Meberg, A. E., & Vik, T. (2008). Cerebral palsy in Norway: Prevalence, subtypes and severity. European Journal
of Paediatric Neurology, 12, 4–13.
Blair, E., & Stanley, F. (1985). Interobserver agreement in the classification of cerebral palsy. Developmental Medicine and Child Neurology, 27, 615–622.
Bottos, M., Granato, T., Allibrio, G., Gioachin, C., & Puato, M. L. (1999). Prevalence of cerebral palsy in north-east Italy from 1965 to 1989. Developmental Medicine
and Child Neurology, 41, 26–39.
Colver, A. F., & Sethumadhavan, T. (2003). The term diplegia should be abandoned. Archives of Disease in Childhood, 88, 286–290.
Damiano, D., Abel, M., Romness, M., Oeffinger, D., Tylkowski, C., Gorton, G., et al. (2006). Comparing functional profiles of children with hemiplegic and diplegic
cerebral palsy in GMFCS Levels I and II: Are separate classifications needed? Developmental Medicine and Child Neurology, 48, 797–803.
Gainsborough, M., Surman, G., Maestri, G., Colver, A., & Cans, C. (2008). Validity and reliability of the guidelines of the Surveillance of Cerebral Palsy in Europe for
the classification of cerebral palsy. Developmental Medicine and Child Neurology, 50, 828–831.
Gorter, J. W., Rosenbaum, P. L., Hanna, S. E., Palisano, R. J., Bartlett, D. J., Russell, D. J., et al. (2004). Limb distribution, motor impairment, and functional
classification of cerebral palsy. Developmental Medicine and Child Neurology, 46, 461–467.
Hagberg, B., Hagberg, G., Beckung, E., & Uvebrant, P. (2001). Changing panorama of cerebral palsy in Sweden. VIII. Prevalence and origin in the birth year 1991–94.
Acta Paediatrica, 90, 271–277.
Hagberg, B., Hagberg, G., & Olow, I. (1993). The changing panorama of cerebral palsy in Sweden. VI. Prevalence and origin during the birth year period 1983-1986.
Acta Paediatrica Scandinavica, 82, 387–393.
Hagberg, B., Hagberg, G., Olow, I., & von Wendt, L. (1996). The changing panorama of cerebral palsy in Sweden. VII. Prevalence and origin in the birth year period
1987–90. Acta Paediatrica, 85, 954–960.
Higgins, J. P. T., Thompson, S. G., Deeks, J. J., & Altman, D. G. (2003). Measuring inconsistency in meta-analyses. British Medical Journal, 327, 557–560.
Himmelmann, K., Hagberg, G., Beckung, E., Hagberg, B., & Uvebrant, P. (2005). The changing panorama of cerebral palsy in Sweden. IX. Prevalence and origin in the
birth-year period 1995 to 1998. Acta Paediatrica, 94, 287–294.
Hurvitz, E. A., & Brown, S. H. (2010). The terms diplegia, quadriplegia, and hemiplegia should be phased out. Developmental Medicine and Child Neurology, 52,
1066–1070.
Kavcic, A., & Perat, M. V. (1998). Prevalence of cerebral palsy in Slovenia: Birth years 1981 to 1990. Developmental Medicine and Child Neurology, 40, 459–
463.
MacGillivray, I., & Campbell, D. M. (1995). The changing pattern of cerebral palsy in Avon. Paediatric and Perinatal Epidemiology, 9, 146–155.
Mongan, D., Dunne, K., O’Nuallain, S., & Gaffney, G. (2006). Prevalence of cerebral palsy in the West of Ireland 1990–1999. Developmental Medicine and Child
Neurology, 48.
Palisano, R., Rosenbaum, P., Walter, S., Russell, D., Wood, E., & Galuppi, B. (1997). Development and reliability of a system to classify gross motor function in
children with cerebral palsy. Developmental Medicine and Child Neurology, 39, 214–223.
Parkes, J., Dolk, H., Hill, N., & Pattenden, S. (2001). Cerebral palsy in Northern Ireland: 1981–93. Paediatric and Perinatal Epidemiology, 15, 278–286.
Peek, A., van Essen, P., Gibson, C., Scott, H., Chan, A., & Haan, E. (2008). 2007 Annual report of the South Australian Cerebral Palsy Register. Adelaide: Children, Youth
and Women’s Health Services.
Pharoah, P. O. D., Platt, M. J., & Cooke, T. (1996). The changing epidemiology of cerebral palsy. Archives of Disease in Childhood, Fetal and Neonatal Edition, 75, 169F–
173F.
Riikonen, R., Raumavirta, S., Sinivuori, E., & Seppala, T. (1989). Changing pattern of cerebral palsy in the southwest region of Finland. Acta Paediatrica Scandinavica,
78, 581–587.
Robinson, M. N., Peake, L. J., Ditchfield, M. R., Reid, S. M., Lanigan, A., & Reddihough, D. S. (2008). Magnetic resonance imaging findings in a population-based cohort
of children with cerebral palsy. Developmental Medicine and Child Neurology, 51, 31–45.
Rosenbaum, P., Gorter, J. W., Palisano, R., & Morris, C. (2010). The relationship of cerebral palsy subtype and functional motor impairment: A population-based
study. Developmental Medicine and Child Neurology, 52, 682–683.
Rosenbaum, P., Paneth, N., Leviton, A., Goldstein, M., & Bax, M. (2007). A report: The definition and classification of cerebral palsy April 2006. Developmental
Medicine and Child Neurology, 49(Suppl.), 8–14.
StataCorp. (2009). Stata Statistical Software: Release 11. College Station, Texas: StataCorpLP.
Shevell, M. I. (2010). The terms diplegia and quadriplegia should not be abandoned. Developmental Medicine and Child Neurology, 52, 508–509.
Shevell, M. I., Dagenais, L., & Hall, N. (2009a). Comorbidities in cerebral palsy and their relationship to neurologic subtype and GMFCS level. Neurology, 72,
290–296.
Shevell, M. I., Dagenais, L., & Hall, N. (2009b). The relationship of cerebral palsy subtype and functional motor impairment: A population-based study.
Developmental Medicine and Child Neurology .
Sigurdardottir, S., Eiriksdottir, A., Gunnarsdottir, E., Meintema, M., Arnadottir, U., & Vik, T. (2008). Cognitive profile in young Icelandic children with cerebral palsy.
Developmental Medicine and Child Neurology, 50, 357–362.
Stanley, F., Blair, E., & Alberman, E. (2000). Cerebral palsies: Epidemiology and causal pathways. London: Mac Keith Press.
Surman, G., Newdick, H., King, A., Gallagher, M., & Kurinczuk, J. J. (2008). 4Child: Four counties database of cerebral palsy, vision loss and hearing loss in children.
Annual Report 2008, including data for births 1984 to 2002. Oxford: National Perinatal Epidemiology Unit.
Surveillance of Cerebral Palsy in Europe (SCPE). (2002). Prevalence and characteristics of children with cerebral palsy in Europe. Developmental Medicine and Child
Neurology, 44, 633–640.
Surveillance of Cerebral Palsy in Europe (SCPE). (2000). Surveillance of cerebral palsy in Europe: A collaboration of cerebral palsy surveys and registers.
Developmental Medicine and Child Neurology, 42, 816–824.
Topp, M., Uldall, P., & Greisen, G. (2001). Cerebral palsy births in eastern Denmark, 1987–90: Implications for neonatal care. Paediatric and Perinatal Epidemiology,
15, 271–277.
Watson, L., Blair, E., & Stanley, F. (2006). Report of the Western Australian Cerebral Palsy Register. West Perth: Telethon Institute for Child Health Research.
Westbom, L., Hagglund, G., & Nordmark, E. (2007). Cerebral palsy in a total population of 4–11 year olds in southern Sweden. Prevalence and distribution
according to different CP classification systems. BMC Pediatrics, 7, 41.
Yeargin-Allsopp, M., Van Naarden Braun, K., Doernberg, N. S., Benedict, R. E., Kirby, R. S., & Durkin, M. S. (2008). Prevalence of cerebral palsy in 8-year old children
in three areas of the United States in 2002: A multisite collaboration. Pediatrics, 121, 547–554.