15146_layout.fm

Prevention of dislocation of the hip in
children with cerebral palsy
THE FIRST TEN YEARS OF A POPULATION-BASED PREVENTION PROGRAMME G. Hägglund,
In 1994, a register for cerebral palsy and a health-care programme were started in southern
S. Andersson,
Sweden with the aim of preventing dislocation of the hip in children with cerebral palsy. It
H. Düppe,
involved all children with cerebral palsy born in 1992 or later.
H. Lauge-Pedersen,
None of the 206 affected children born between 1992 and 1997 has developed a
E. Nordmark,
dislocation following the introduction of the prevention programme. Another 48 children
L. Westbom
moved into the area and none developed any further dislocation. Of the 251 children with
cerebral palsy, aged between five and 11 years, living in the area on January 1, 2003, only
two had a dislocated hip. One boy had moved into the area at age of nine with a dislocation
and a girl whose parents chose not to participate in the programme developed bilateral
dislocation. One boy, whose condition was considered to be too poor for preventative
surgery, developed a painful dislocation of the hip at the age of five years and died three
years later.
Eight of 103 children in a control group, consisting of all children with cerebral palsy
living in the area between 1994 and 2002, and born between 1990 and 1991, developed a
dislocation of the hip before the age of six years.
The decreased incidence of dislocation after the introduction of the prevention
programme was significant (p < 0.001). Dislocation of the hip in cerebral palsy remains a
serious problem, and prevention is important. Our screening programme and early
intervention when lateral displacement of the femoral head was detected appear to be
successful.
Children with cerebral palsy have an increased Patients and Methods
risk of lateral displacement of the femoral In 1994, a cerebral palsy register and health- head, leading in some cases to dislocation. The care programme for children with cerebral reported incidence of lateral displacement or palsy was started in southern Sweden involving dislocation is related to the severity of the con- all children born with cerebral palsy in the area dition and varies between 7% in ambulatory since 1990. The condition was defined as “an children and 60% in those with total-body involvement.1,2 However, no population-based progressive, but often changing, motor impair- NeuropaediatricianDepartment of Paediatrics ment syndromes secondary to lesions or anom- alies of the brain arising in the early stages of results in significant morbidity in terms of development”.11 Only children who were alive pain,3-5 contractures, problems with sitting, at two years of age and had their cerebral Orthopaedic SurgeonDepartment of Orthopaedics standing or walking,6 fractures,7,8 skin ulcera- lesion before that time were included. The sub- tion and difficulty with perineal care,6,8 pelvic type of cerebral palsy was determined after the fourth birthday according to the method of Correspondence should be sent to Dr G. Hägglund.
The first suggestion that dislocation in cere- function was classified according to the classi- 2005 British Editorial Society of Bone and nearly 50 years ago.10 We initiated a popula- fication system (GMFCS) of Palisano et al13 tion-based prevention programme for chil- which is an age-related five-level system in which level I is the least and level V the most preventing dislocation of the hip by early J Bone Joint Surg [Br] 2005;87-B:95-101.
detection and intervention. We now describe The study area has a population of 1.3 mil- and analyse the results of the first ten years of lion. The total population of children born in Accepted after revision 25 May 2004 1990 and later were systematically reviewed in G. HÄGGLUND, S. ANDERSSON, H. DÜPPE, H. LAUGE-PEDERSEN, E. NORDMARK, L. WESTBOM Table I. The gross motor function classification system (GMFCS).13 The levels represent the highest level of mobility that
a child is expected to achieve between six and 12 years of age
Walks without restrictions; limitations in more advanced gross motor skills Walks without assistive devices; limitations walking outdoors and in the community Walks with assistive devices; limitations walking outdoors and in the community Self-mobility with limitations; children are transported or use power mobility outdoors and in the community Self-mobility is severely limited even with the use of assistive technology Table II. Subdiagnosis and GMFCS* level in 103 children with cere-
In the present study children born between 1990 and bral palsy born between 1990 and 1991 (control group) and 258 chil- 1997 and living in the area between 1994 and 2002 were dren born between 1992 and 1997 (study group) analysed. Data for children born in the area were used Born 1990 to 1991
Born 1992 to 1997
until the children had moved out of the area or died. Chil- dren who had moved into the area were included in the fol- low-up programme, but their results were analysed The control group. There were 103 children born between
1990 and 1991; 87 lived in the area when the programme started and 16 had moved in at a median age of nine years (5 to 11; inter-quartile range 8 to 9). By January 1, 2003, three children had died at six, nine and 11 years of age, respectively. One child had moved out of the area at the age The study group. There were 258 children born between
1992 and 1997; 210 were born in the area and 48 had * GMFCS, gross motor function classification system moved in at a median age of three years (0 to 9; inter-quar-tile range 2 to 5). Twenty-two children did not participatein the prevention programme. Four families declined to 1998 and 2002 in order to identify all children with prob- participate, and 18 children were identified during the able cerebral palsy.14 In those aged four years and older the review undertaken in 2002 and then included in the pro- diagnosis and subtypes of cerebral palsy were established.
gramme. However, information regarding their hips at the The prevalence of children with cerebral palsy aged from census date (January 1, 2003) was collected from their clin- four to seven years and living in the area on January 1, ical records. Of the remaining 236 children three had died 1998 was 2.4 per 1000 children.14 Our study of dislocation at two, seven, and eight years of age, respectively, all from of the hip also included children with cerebral palsy who causes not related to the prevention programme. Four chil- moved out of and into the area or died during the study dren had moved out of the area at five, six, eight and nine In addition to an active search for children with cerebral The control and study groups were comparable with palsy in order to offer them participation in the programme regard to the proportion of subtype of cerebral palsy and as early as possible, the health-care programme included a GMFCS level (Table II). In the classification system,13 spas- continuing standardised follow-up of the diagnosis, gross tic diplegia includes all children in whom the lower limbs motor function, clinical findings and treatment. The local are more affected than the upper ones. This is in contrast to physiotherapist and occupational therapist completed a spastic tetraplegia which is defined as massive total motor record twice a year until the age of six years and then once disability with all four limbs severely involved, the upper a year. The results were computerised and the local team limbs at least as severely as the lower.
received a report showing the development of the child over The migration percentage15 (MP) and acetabular index16 (AI) were measured on the radiographs (Fig. 1). All radio- A standardised radiological follow-up of the hips was graphs were assessed by one of the authors (GH). Both carried out. They were examined on an anteroposterior measurements have been shown to be reliable.15,17 How- radiograph at diagnosis, then at least once a year in chil- ever, recently some doubt on the inter- and intra-measurer dren with the diplegic, tetraplegic or dystonic type cerebral reliability has been raised, suggesting that repeated meas- palsy until the age of eight years, and then on an individual urements should be made by one individual.18 Hips with an basis. Children with spastic hemiplegia or pure ataxia were MP < 33% and an AI < 30˚ were regarded as normal. In only examined radiologically at four years of age. The hip children with lateral displacement or acetabular dysplasia, programme began in 1994 and included children born in the findings were compared with earlier measurements and data from the physiotherapist’s follow-up. The decision PREVENTION OF DISLOCATION OF THE HIP IN CHILDREN WITH CEREBRAL PALSY osteotomy. Five children in the control group had beentreated by SDR, and three with ITB.
None of the 206 children monitored by the prevention programme from the time of diagnosis has developed dislo-cation of the hip (p < 0.001). None of the 48 who movedinto the area between 1995 and 2002 developed any furtherdislocation.
One boy had a dislocated hip when he moved into the area. One girl whose family were unwilling to participate inthe prevention programme has developed bilateral disloca-tion. One boy, born in 1993, with spastic tetraplegia had alateral displacement of the hips (MP 40 left, 74 right) at theage of three years. Operation by adductor-iliopsoas tenot-omy was recommended, but the child’s respiratory condi- tion was such that he was only expected to survive for a Diagram showing measurement of the acetabular index (AI) and migra- short period. Surgery was therefore not undertaken. The tion percentage (MP; MP = A/B x 100).
right hip dislocated within two years and the child died ateight years of age.
In the study group 50 children (78 hips) showed lateral- with regard to preventative treatment was made together isation of the hip with an MP exceeding 33% (Table III). At with the child’s family and the local treatment team.
the date of census, 54 of these hips had been corrected to In the younger and in older children with minor displace- normal (MP < 33%). Of these, 11 were operated on by ment, orthopaedic surgery usually involved bilateral tenot- adductor-iliopsoas tenotomy, and 16 by proximal varus omies of the adductors and iliopsoas. With marked femoral osteotomy. Five children (six hips with lateralisa- displacement, or if the lateral displacement was not reduced tion) have been treated by SDR, and two children (three within one year after adductor-iliopsoas tenotomy, a varus hips) with ITB. In 18 hips the MP corrected to normal with osteotomy of the proximal femur was performed. In chil- no operative treatment. These hips had a mean MP of 35% dren with marked dysplasia, acetabular reconstruction was sometimes necessary. Post-operatively, the children were At the date of census 24 hips still had an MP equal to or treated by positioning the hips in extension and abduction exceeding 33% (Table III). Three children (five lateralised during lying and sitting. Standing frames and abduction hips) had died. In nine hips the MP was decreasing, and in splints were often used. Treatment with a hip spica plaster two it remained at 33%. In seven hips preventative surgery was planned. One child was operated on just before the In children for whom reduction of spasticity was planned using selective dorsal rhizotomy (SDR) or continuous infu- The AI was increased (> 30˚) in 18 children (23 hips) in sion of intrathecal baclofen (ITB), it was often possible to the study group. The mean AI in these hips was 34˚ (30 to wait and see if the decreased spasticity reduced lateral hip 40). All hips with an increased AI also had an MP greater displacement. The indications for SDR and ITB were when than 33%. At the census date the AI was normal in all problems were directly related to increased muscle tone, except seven hips. In one case, with an AI of 40˚, a pelvic and the decision to employ such treatment was made irre- osteotomy was performed after the date of census. It was spective of the degree of lateralisation of the hip.
decreasing or approaching 30˚ in the remaining six hips.
Statistical analysis. Fisher’s test was used to evaluate the
In total, 25 children, five in the control group and 20 in efficiency of the prevention programme.
the study group, have been operated on by SDR at a medianof four years of age (3 to 6). Of these, 17 had normal hip radiographs before and after operation. In the remaining At the date of census, eight children in the control group eight children, eight hips showed lateral migration before had developed unilateral dislocation of the hip, always SDR. In five hips the lateral migration corrected without between the age of three and six years. All children with further treatment post-operatively. In a further two hips the dislocated hips had severe pain, at least periodically. All had MP decreased, and in one hip it remained at 33%. Two hips pelvic obliquity and scoliosis. In three patients, resection of were normal before SDR, but have developed lateralisation the femoral head and subtrochanteric valgus osteotomy after the procedure. In one boy (case 24, Table III) the MP were performed to reduce pain. Three of the children with increased to 42% at eight years of age, and a varus femoral dislocation had died. A further nine of the 103 children in osteotomy was performed. In the other case the MP the control group had developed lateral displacement. All increased to 34%. This hip will be investigated by further had been operated upon to prevent dislocation, six by radiography before a decision is made regarding preventa- adductor-iliopsoas tenotomy and three by varus femoral G. HÄGGLUND, S. ANDERSSON, H. DÜPPE, H. LAUGE-PEDERSEN, E. NORDMARK, L. WESTBOM Table III. Data on the 50 children (78 hips) with lateral migration in the study group
Age at latest
GMFCS†
Highest operation
follow-up
MP at latest
Diagnosis*
MP (yrs) MP
Treatment‡
follow-up
Comments
Planned for new femoral + pelvic osteotomy * according to Hagberg et al12; H, spastic hemiplegia; D, spastic diplegia; T, spastic triplegia; Dy, dystonic type; A, athetonic type† GMFCS, gross motor function classification system‡ 0, no treatment; ad, adductor-psoas tenotomy; ost, varus osteotomy of the proximal femur; SDR, selective dorsal rhizotomy; ITB, intrathecalbaclofen pump PREVENTION OF DISLOCATION OF THE HIP IN CHILDREN WITH CEREBRAL PALSY Table III (cont.) Data on the 50 children (78 hips) with lateral migration in the study group
Age at latest
GMFCS†
Highest operation
follow-up
MP at latest
Diagnosis*
MP (yrs) MP
Treatment‡
follow-up
Comments
* according to Hagberg et al12; H, spastic hemiplegia; D, spastic diplegia; T, spastic triplegia; Dy, dystonic type; A, athetonic type† GMFCS, gross motor function classification system‡ 0, no treatment; ad, adductor-psoas tenotomy; ost, varus osteotomy of the proximal femur; SDR, selective dorsal rhizotomy; ITB, intrathecalbaclofen pump Seven children have been treated with ITB, three in the in southern Sweden, who have a dislocated hip. These are control group and four in the study group. The hips in two the boy who moved into the area with an established dislo- children in the study group showed lateral displacement cation and the girl who did not participate in the pro- before treatment with ITB. At the date of census three of these hips were normal and one showed an MP of 33%.
A dislocated hip in a child with cerebral palsy is a serious problem. The reported frequency of children with pain var- Discussion
ies, probably due to the difficulty in assessing pain in these Of the 103 children in the control group, eight have devel- children. Cooperman et al3 and Bagg et al5 who used stand- oped dislocation of the hip. This cannot be looked upon as ardised criteria for assessment of pain have reported the the natural history of untreated hips. Some of the nine chil- highest incidence of 50% to 90%. Pain often results in dren operated upon for lateral displacement would proba- increased muscle tone with increased energy requirements.
bly have had dislocation if they had not had the operation.
A child with dislocated hips often deteriorates with under- Some of the children with lateral displacement and treated nourishment, increased contractures, postural difficulties, by SDR or with ITB could also have developed dislocation skin ulceration and problems with perineal care. Perhaps it of the hip. In the study group, 50 (21%) of 236 children is no mere coincidence that four of the five children who showed lateral displacement of the hip. The displacement died at more than four years of age during the study period returned to normal without operative treatment in 12. The remaining 38 (16%) would probably have had dislocation We chose an MP of 33% as the indication for interven- without preventative surgery. This suggests that the natural tion according to the definition of subluxation of Reim- risk for hip dislocation in a total population of children ers.15 Other reports have used an MP of 30% to 40% as the with cerebral palsy is between 15% and 20%.
indication for surgery.20,21 Several hips with an MP The children in the study group were aged between five between 33% and 40% returned to normal levels without and 11 years at the date of census. They are still at potential operative treatment. These children had not received any risk of dislocation, but they are older than the age at dislo- additional non-operative treatment as a result of the radio- cation in the control group, and it is known that most dis- logical findings. No hip with an MP exceeding 42% locations of the hip occur before seven years of age.19 The returned to normal without operative treatment. We lower incidence at this age in the study group as compared recommend radiological follow-up at intervals of six with the control group is statistically significant (p < months before a decision about surgery is made in hips with 0.001). At present, there are only two children with cere- an MP of between 33% and 40%, especially if the child is bral palsy in a total of about 400 up to the age of 12 years walking and has a good range of movement in the hip.
G. HÄGGLUND, S. ANDERSSON, H. DÜPPE, H. LAUGE-PEDERSEN, E. NORDMARK, L. WESTBOM Radiographs showing the hips of a boy (case 32) with dystonic-type cerebral palsy, GMFCS 5. a) At five years of age witha migration percentage (MP) of 79% on the left and 33% on the right. He underwent bilateral adductor-psoas tenotomyand b) at ten years of age with an MP of 28% on the left and 58% on the right.
Radiographs showing the hips of a boy (case 13) with dystonic-type cerebral palsy, GMFCS 5. a) At three years of age with amigration percentage (MP) of 44% on the left and 50% on the right. He underwent bilateral adductor-psoas tenotomy, b) atfive years of age with an MP of 47% on the left and 73% on the right, c) after operation with proximal femoral varus osteotomyand d) at six years of age with an MP of 46 % on the left and 40% on the right.
PREVENTION OF DISLOCATION OF THE HIP IN CHILDREN WITH CEREBRAL PALSY The AI was increased in 23 hips, all of which also the hips, arranging radiological screening and the analysis showed an increased MP. Our results do not support the findings of Cooke, Cole and Carey22 that lateral displace- In the future we hope to be able to discover which chil- ment is always preceded by an increased AI. Our findings dren in the population are at risk of dislocation of the hip in indicate that the MP could be used as the only measurement relation to subtype of cerebral palsy, function and other in a screening programme for dislocation of the hip. It also information from the collected data. We also hope to suggests that lateral displacement precedes acetabular dys- improve the timing of radiological examination and the choice of intervention for lateral displacement.
Children with spastic hemiplegia and pure ataxia are The study was supported by the Medical faculty, Lund University and Stiftelson included in the radiological follow-up, but normally radio- för bistånd åt vanföra i Skåne.
graphs were only taken at diagnosis and at four years of No benefits in any form have been received or will be received from a com- mercial party related directly or indirectly to the subject of this article.
age. One reason for their inclusion is that some childrenwith spastic diplegia in their early years may present withunilateral symptoms. Some children with ataxic diplegia References
could present with mainly ataxic symptoms, and be initially 1. Howard CB, McKibbin B, Williams LA, Mackie I. Factors affecting the incidence
of hip dislocation in cerebral palsy. J Bone Joint Surg [Br] 1985;67-B:530-2.
2. Lonstein JE, Beck K. Hip dislocation and subluxation in cerebral palsy. J Pediatr
The combination of adduction and flexion spasticity or contracture appears to cause the lateral displacement.23 3. Cooperman DR, Bartucci E, Dietrick E, Millar EA. Hip dislocation in cerebral
Soft-tissue surgery should address both of these deformi- palsy: long-term consequences. J Pediatr Orthop 1987;7:268-76.
ties. We always combine adductor and iliopsoas tenotomy, 4. Moreau M, Drummond DS, Rogala EJ, Aschworth A, Porter T. Natural history of
dislocated hip in spastic cerebral palsy. Dev Med Child Neurol 1979;21:749-53.
and we always perform the surgery bilaterally in children 5. Bagg MR, Farber J, Miller F. Long-term follow-up of hip subluxation in cerebral
with bilateral spasticity or dystonia. If a child who cannot palsy patients. J Pediatr Orthop 1993;13:32-6.
walk has a knee contracture exceeding 20˚, the knee is 6. Samilson RKL, Carson JJ, James P, Raney FL. Results and complications of
adductor tenotomy and obturator neurectomy in cerebral palsy. Clin Orthop treated post-operatively by serial casting. The combination of iliopsoas tenotomy and treatment of the knee contrac- 7. McIvor WC, Samilson RL. Fractures in patients with cerebral palsy. J Bone Joint
ture is probably the reason why no child has developed 8. Pritchett JW. The untreated unstable hip in severe cerebral palsy. Clin Orthop
hyperabduction post-operatively, a reported complica- 9. Letts M, Shapiro L, Mulder K, Klassen O. The windblown hip syndrome in total
Varus osteotomy of the proximal femur has been carried body cerebral palsy. J Pediatr Orthop 1984;4:55-62.
out in 15 children (21 hips) in the study group. During the 10. Tachdjian MO, Minear WL. Hip dislocation in cerebral palsy. J Bone Joint Surg
first years of the programme some children were diagnosed 11. Mutch L, Alberman E, Hagberg B, Kodasma K, Perat MV. Cerebral palsy epide-
and referred late when it was considered to be too late for an miology: where are we now and where are we going? Dev Med Child Neurol adductor-psoas tenotomy. However, in doubtful cases, we 12. Hagberg B, Hagberg G, Olow I. The changing panorama of cerebral palsy in Swe-
often prefer to do an adductor-psoas tenotomy first. If no den 1954-1970. Acta Paediatr Scand 1975;64:187-92.
decrease in lateral displacement is seen within one year, we 13. Palisano R, Rosenbaum P, Walter S, et al. Development and reliability of a sys-
perform a proximal varus femoral osteotomy (Figs 2 and 3).
tem to classify gross motor function in children with cerebral palsy. Dev Med ChildNeurol 1997;39:214-23.
The new techniques of reducing spasticity by SDR, ITB 14. Nordmark E, Hägglund G, Lagergren J. Cerebral palsy in southern Sweden: I prev-
and botulinum toxin have probably prevented lateral dis- alence and clinical features. Acta Paediatr 2001;90:1271-6.
placement in some cases. In one boy a lateral displacement 15. Reimers J. The stability of the hip in children: a radiological study of the results of
of 44% corrected to normal within one year after treatment muscle surgery in cerebral palsy. Acta Orthop Scand 1980;184(Suppl):1-97.
16. Hilgenreiner H. Zur frühdiagnose und frühbehandlung der angeborenen hüftgel-
with ITB. It has been suggested that SDR could increase the lenkeverrenkung. Med Klin 1925;37-38:1385-429.
risk of dislocation by increasing the muscle imbalance at 17. Parrott JK, Boyd RN, Dip OG, et al. Hip displacement in spastic cerebral palsy: reli-
the hip.25 All children treated by SDR were operated upon ability of radiological measures. J Paed Orthop 2002;22:660-7.
up to L2 and no signs of increased muscle imbalance were 18. Faraj S, Atherton WG, Stott NS. Inter- and intra-measurer error in the measure-
ment of Reimer’s hip migration percentage. J Bone Joint Surg [Br] 2004;86-B:434-7.
19. Rang M, Silver R, de la Garza J. Cerebral palsy. In: Lovell WW, Winter RB, eds.
The main challenge for this programme was the early Pediatric orthopaedics. Second ed. Philadelphia: JB Lippincott Co, 1986:345-96.
identification of all children with cerebral palsy in the pop- 20. Onimus M, Allamel G, Manzone P, Laurain JM. Prevention of hip dislocation in
cerebral palsy by early psoas and adductors tenotomies. J Pediatric Surg ulation.14 The health-care programme has been developed in collaboration with the local child rehabilitation team of 21. Dobson F, Boyd RN, Parrott J, Nattrass GR, Graham HL. Hip surveillance in chil-
physiotherapists, occupational therapists, paediatricians dren with cerebral palsy. J Bone Joint Surg [Br] 2002;84-B:720-6.
and orthopaedic surgeons. All interventions have the sup- 22. Cooke PH, Cole WG, Carey RPL. Dislocation of the hip in cerebral palsy. J Bone
Joint Surg [Br] 1989;71-B:441-6.
port of both the local team and the paediatric orthopaedic 23. Kalen V, Bleck EE. Prevention of spastic paralytic dislocation of the hip. Dev Med
surgeon. This collaboration has been essential both to start and run the programme. However, it is our experience that 24. Silver RL, Rang M, Chan J, de la Garza J. Adductor release in nonambulant chil-
dren with cerebral palsy. J Pediatr Orthop 1985;5:672-7.
once the child has been included in the follow-up pro- 25. Greene WB, Dietz FR, Goldberg MJ, et al. Rapid progression of hip subluxation in
gramme, an orthopaedic surgeon should be responsible for cerebral palsy after selective posterior rhizotomy. J Pediatr Orthop 1991;11:494-7.

Source: http://www.bjj.boneandjoint.org.uk/content/87-B/1/95.full.pdf