Bilateral hip surgery in severe cerebral palsy
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Bilateral hip surgery in severe cerebral palsy
A PRELIMINARY REVIEW
K. L. Owers, J. Pyman, M. F. Gargan, P. J. Witherow,
N. M. A. Portinaro
From Southmead Hospital and Bristol Royal Hospital for Sick Children, Bristol,
England
hen cerebral palsy involves the entire body bony deformities may develop around the hip including
W pelvic asymmetry indicates that both hips are
‘at risk’. We carried out a six-year retrospective
coxa valga, malorientation of the femoral head and ace-
tabular dysplasia. These often progress and joint stiffness
2,3
clinical, radiological and functional study of 30 and/or dislocation may occur.
children (60 hips) with severe cerebral palsy involving The incidence of hip dysplasia and dislocation in patients
the entire body to evaluate whether bilateral with cerebral palsy varies according to the clinical pattern.
simultaneous combined soft-tissue and bony surgery of The frequency of incongruity of the hip correlates well with
4
the hip could affect the range of movement, achieve the severity of motor and intellectual impairment and the
5
hip symmetry as judged by the windsweep index, patients’ independent mobility. Patients with spastic diple-
improve the radiological indices of hip containment, gia who can walk rarely present with subluxation of the
6
relieve pain, and improve handling and function. hip. In patients with cerebral palsy producing quadriplegia
7
The early results at a median follow-up of three instability of the hip has been reported in up to 60%. This
years showed improvements in abduction and figure increases as fixed flexion-adduction contractures
6
adduction of the hips in flexion, fixed flexion develop. Subluxation or dislocation of the hip in total body
4
contracture, radiological containment of the hip using involvement is often bilateral with the clinical pattern of
both Reimer’s migration percentage and the contractures being asymmetrical. This can be associated
4
centre-edge angle of Wiberg, and in relief of pain. with pelvic obliquity and spinal deformity. A windsweep
Ease of patient handling improved and the satisfaction deformity may develop with the adducted, flexed and inter-
7
of the carer with the results was high. There was no nally rotated hip tending to sublux posteriorly and the
difference in outcome between the dystonic and contralateral abducted, extended and externally rotated hip
hypertonic groups. displaced anteriorly (Fig. 1). Subluxation of the abducted
J Bone Joint Surg [Br] 2001;83-B:1161-7. hip is difficult to identify on the anteroposterior (AP) pelvic
Received 13 April 2000; Accepted after revision 11 January 2001 radiographs, but can be detected clinically by a palpation of
the femoral head in the groin and a progressive abduction
contracture.
Cerebral palsy is a heterogeneous non-progressive central Our aim was to determine whether bilateral, simultane-
motor deficit which is seen in 1.5 to 2.5 children per 1000 ous, soft-tissue and bony surgery around the hip can
1
live births. The non-progressive central pattern of the improve the range of movement, achieve hip symmetry,
disease contrasts with the progressive deterioration of increase radiological congruity of the hip, relieve pain, ease
peripheral motor function which occurs during growth. the handling of the patient and improve the level of func-
This results in increasing muscle contractures. Secondary tion in patients with total body involvement. We also
compared the outcomes for dystonic and hypertonic
patients.
K. L. Owers, MRCS, Clinical Research Fellow Patients and Methods
Department of Orthopaedics and Trauma, Southmead Hospital, Westbury-
on-Trym, Bristol BS10 5NB, UK.
The medical records of 30 children (60 hips) with cerebral
J. Pyman, MCSP, Superintendent Physiotherapist
M. F. Gargan, FRCS, Consultant Orthopaedic Surgeon palsy with total body involvement were reviewed after
P. J. Witherow, FRCS, Consultant Orthopaedic Surgeon approval of the Ethical Committee had been obtained. All
N. M. A. Portinaro, MD, MSc, Consultant Orthopaedic Surgeon
The Bristol Royal Hospital for Sick Children, St Michael’s Hill, Bristol had undergone bilateral simultaneous combined soft-tissue
BS2 8BJ, UK. and bony surgery around the hip between 1991 and 1997.
Correspondence should be sent to Miss K. L. Owers. There were 12 boys and 18 girls with a mean age at the
©2001 British Editorial Society of Bone and Joint Surgery time of surgery of 7.7 years (3.1 to 12.2). In the 19 dystonic
0301-620X/01/811266 $2.00 patients, five were windswept towards the left and 14 to the
VOL. 83-B, NO. 8, NOVEMBER 2001 11611162 K. L. OWERS, J. PYMAN, M. F. GARGAN, P. J. WITHEROW, N. M. A. PORTINARO
died from an unrelated cause and was therefore excluded.
Since our study was carried out retrospectively information
was incomplete for the preoperative range of movement in
a further six patients. Statistical analysis was only carried
out on hips with complete preoperative and postoperative
data (Table I). Radiological records were complete for 25
patients and functional assessment for 29. Only 25 carers
returned completed questionnaires.
Clinical assessment. Our physiotherapist (JP) and the
paediatric orthopaedic consultant surgeons (PJW and MFG)
recorded the full range of passive movement and fixed
flexion contractures of each hip before operation under
general anaesthesia and at the latest follow-up clinic. The
total range of hip movement was assessed before and after
operation as the sum of the difference between hip flexion
and fixed flexion deformity, plus the internal rotation,
external rotation, abduction and adduction of the hip recor-
ded in flexion.
The windsweep deformity was assessed by measuring
hip abduction and adduction in flexion. The windsweep
8
index (WI) was calculated as the difference between the
sum of abduction of one hip and adduction of the con-
tralateral hip and vice versa (i.e. abduction R + adduction L
minus adduction R + abduction L) and was also recorded
before and after operation. A reduction in the WI at follow-
up was considered to be an improvement in the symmetry
of the movements of the hip. The length of follow-up was
Fig. 1 compared with the difference in WI.
Classical windsweep deformity of the lower limbs seen in severe Radiological assessment. The Reimers migration per-
cerebral palsy. centage (RM%) and centre-edge angle of Wiberg (CEA)
were measured for both hips on AP radiographs of the
9-11
pelvis. Hips were considered to be subluxed when the
right, while in the 11 hypertonic patients, five were wind- RM% was greater than 33 and dislocated when it was
9
swept towards the left and six to the right. 100.
Seven of the dystonic patients and three of the hyper- Functional assessment. The patients’ level of function was
tonic patients had undergone various soft-tissue procedures assessed and recorded by the physiotherapist at the time
at other hospitals before being assessed in our tertiary that the other examinations were carried out. This involved
referral centre. The effect of previous soft-tissue surgery on determining the ability to sit independently, to go into a
the outcome was determined. standing frame and to walk with an assistant or with a
All patients were called back to designated research frame. Pain in the hip was also noted. Since this group of
clinics at a median time of follow-up of three years (0.25 to patients had little verbal communication this could only be
6.5). Two were unavailable for clinical review, but func- subjectively assessed by the carers and physiotherapist and
tional data were obtained from the physiotherapist. One had was dependent upon apparent distress on movement of the
Table I. Preoperative and postoperative data for children with quadriplegic cerebral palsy who underwent bilateral hip surgery
Preoperative
Number Number Preoperative Postoperative postoperative
of patients of hips (mean ± SD) (mean ± SD) difference 95% CI p value
Abduction and adduction 21 97 ± 38 135 ± 30 38 ± 39 21 to 56 < 0.001
in flexion in degrees
Fixed flexion deformity 53 15 ± 14 8 ± 11 -7 ± 14 -11 to -3 < 0.001
in degrees
Flexion in degrees 54 113 ± 11 111 ± 15 -2 ± 17 -6 to 3 0.446
Total range of 43 251 ± 54 256 ± 42 5 ± 56 -12 to 22 0.536
movement in degrees
WI in degrees 21 50 ± 45 44 ± 44 -6 ± 55 -31 to 19 0.626
THE JOURNAL OF BONE AND JOINT SURGERYHIP SURGERY IN TOTAL-BODY-INVOLVED CEREBRAL PALSY 1163
carried out unilaterally in three patients and bilaterally in
seven. All patients were put into a full hip spica for six
weeks and given a spica transporter to take home, allow-
ing early reintegration of the child into the community
(Fig. 2). Pain and spasm were controlled postoperatively
by intravenous midazolam and epidural anaesthesia for
two to four days under close supervision from the pain
control team.
At six weeks, all patients were readmitted to hospital for
removal of the spica and controlled mobilisation in gaiters
with physiotherapy for five days. The aim was to reduce the
risk of fractures in osteoporotic bone and to retrain the
carers in the handling and positioning of their child who
had a new body shape after the corrective surgery. They
were then discharged home with a fitted sleeping shell and
continued to have daily community physiotherapy for one
month. The initial postoperative combined follow-up
assessment was carried out after three months.
Postoperative complications in both groups were
recorded.
Statistical analysis. The results were analysed using Stu-
dent’s t-test for paired data and McNemar’s and marginal
Fig. 2 homogenicity tests to assess improvement of radiological
and functional indices. The non-parametric Mann-Whitney
The spica transporter eases management of the patient
postoperatively and allows early reintegration of the U test was used to determine the significance of any
child to the community. difference in change between the dystonic and hypertonic
groups, and between those who had had previous surgery
and those who had not.
hip during daily activities. A confidential questionnaire was
given to all parents/carers to establish their opinion on the Results
ease of handling of the patient and function, including
transfers, seating and nursing before and after operation, Clinical outcome. The combined range of abduction and
and their overall satisfaction with the outcome. adduction in flexion for each hip increased significantly
Operative technique. The indications for surgery were from a mean of 97° to 135° (Table I). There was no
increased asymmetry, radiological subluxation or disloca- difference between the dystonic and hypertonic groups
tion and pain. The operation involved both bony and soft- (p = 0.62) or between those who had had previous surgery
tissue surgery. All patients underwent bilateral femoral and those who had not (p = 0.72). Fixed flexion deformity
varus derotation osteotomies placing the femur in the mid- improved from a mean of 15° preoperatively to 8° at
range of rotation in flexion and extension. All femora were follow-up and mean hip flexion and total range of move-
shortened. Each acetabular osteotomy was hinged on the ment were preserved (Table I).
triradiate cartilage to give the most congruous joint. All The WI was not significantly reduced from a mean of
hips were fully reduced at the end of the operation as 50° preoperatively to 44° at follow-up (Table I). Only nine
confirmed by an AP radiograph of the pelvis in the spica. of 21 (43%) patients with complete data were found to have
Soft-tissue surgery was also carried out selectively, depend- more symmetrical hips using the WI, despite increased
ing on the preoperative assessment and the intraoperative abduction and adduction in flexion in 17 (81%). The medi-
examination of hip movements under anaesthesia after the an time of follow-up for these patients was 2.8 years. The
corrective bony procedure had been carried out. nine patients with improved symmetry had a median fol-
Of the 19 dystonic patients, eight had an acetabuloplasty low-up of 2.3 years compared with 3.2 years for those with
on the adducted side only and three had bilateral pro- deteriorating symmetry (p = 0.25). There was no age-relat-
cedures. Psoas and/or adductor releases were carried out ed difference between these two groups.
unilaterally in 11 patients and bilaterally in six. Con- There was no statistical difference between the dystonic
tralateral releases of gluteus maximus and tensor fascia lata and hypertonic groups (p = 0.54) or between those who had
were carried out in four patients. had previous surgery and those who had not (p = 0.11) in
Of the 11 hypertonic patients, six underwent acet- the evaluation of change in mean WI (Table II).
abuloplasty on the adducted side only and three had Radiological outcome. On the 25 complete sets of radio-
bilateral operations. Psoas and/or adductor releases were graphs a significant reduction in the incidence of subluxa-
VOL. 83-B, NO. 8, NOVEMBER 20011164 K. L. OWERS, J. PYMAN, M. F. GARGAN, P. J. WITHEROW, N. M. A. PORTINARO
Table II. Mean preoperative and postoperative WI values (degrees) for 21 patients divided
into their different patient groups and as to whether they had undergone previous surgery
Number of
patients Preoperative WI Postoperative WI
Group
Hypertonic 9 36 19
Dystonic 12 60 59
Previous surgery
Yes 10 76 50
No 11 37 40
tion and dislocation of the hip was detected. Preoperatively, Discussion
36 of 50 hips (72%) had an RM% >33 of which nine of the
25 patients (36%) had bilateral subluxation or dislocation. In patients with cerebral palsy involving the total body
This compares with eight of 50 hips (16%) found to be pelvic asymmetry is indicative of both hips being ‘at risk’.
subluxed postoperatively. Of the 36 with an RM% >33 Muscle imbalance and bony dysplasia are progressive and
4
preoperatively, six had an RM% of 100 and were, by bilateral, but often asymmetrical. A windswept attitude of
definition, dislocated, compared with none at follow-up. Of the lower limbs often develops, associated with pelvic and
4,7
the 50 hips which suffered recurrent subluxation, eight did spinal deformity, possibly due to asymmetrical central
so over a median follow-up of 3.0 years compared with 2.7 involvement of the motor cortex. When deformity is estab-
years for the rest of the group (p = 0.70) although the range lished, mobility, seating and nursing care are always affec-
was the same. Six of these eight hips were in the dystonic ted. In the long term, instability of the hip causes premature
patient group, with a mean age of 7.9 years compared with and progressive degeneration of the joint with dislocation
7,12
two in eight hypertonic patients who resubluxed at a mean and severe pain in at least 50% of patients.
age of 4.4 years. The aim of surgery to the hip in these patients is to
The overall mean RM% improved from 50 preoper- reduce subluxation/dislocation and to relieve pain, improve
atively to 18 at follow-up (p < 0.001). The CEA also symmetry and aid management. This is achieved by cor-
improved significantly from -5° to 18° (p < 0.001). recting the bony deformity and releasing soft-tissue con-
There was no significant difference between the hyper- tractures. A variety of procedures has been suggested to
tonic and dystonic groups for change in RM% (p = 0.85) or improve the containment and biomechanics of subluxing
CEA (p = 0.89). There was also no difference in RM% hips in these patients. Soft-tissue operations alone have had
13
(p = 0.83) or CEA (p = 0.54) between those who had had variable rates of success and uncertain long-term results.
previous soft-tissue surgery and those who had not. Early soft-tissue surgery may prevent further subluxation
Functional outcome. The patients’ level of function was and dislocation, improve function and relieve pain when
14,15
assessed and recorded by the physiotherapist. Of the 29 compared to conservative treatment. When soft-tissue
patients, the number capable of independent seating surgery was combined with bony operations better long-
13
improved from six (21%) preoperatively to eight (28%) term results are achieved.
postoperatively. The number with the ability to go into a The traditional concept of concentrating on the obviously
standing frame improved from 26 (90%) to 28 (97%). subluxed or dislocated adducted hip results in incomplete
Preoperatively, 20 patients (69%) could not walk but six treatment. Our original rationale for bilateral simultaneous
(21%) could do so with the help of an assistant and three hip surgery in this group of patients was that both hips are
(10%) with a frame. Postoperatively, the level of mobility ‘at risk’. We addressed both hips simultaneously in an
improved in two of the non-walkers and three of six aided attempt to minimise the risk of subsequent progressive
walkers (50%). subluxation or dislocation of both the abducted hip, which
Of the 13 patients who were thought to have pain in the was found to be a problem in previous unilateral surgery, as
hip before operation, 11 (85%) were free from pain at well as the adducted hip. This approach aims to correct
follow-up (p = 0.001). pelvic asymmetry and balance, creating a more anatomical
From the completed questionnaires, 23 of 25 carers basis for retraining and physiotherapy (Fig. 3).
(92%) reported ‘ easier handling’ of their children. Par- All patients had bilateral femoral varus derotation and
ental/carer satisfaction with their child’s management was shortening osteotomies, placing the femur in the mid range
high at 22 of 25 (88%). of rotation. Femoral shortening effectively lengthens the
3,16-19
Complications. Complications included four supracondy- muscles which alters their action on the hip. This is
lar fractures of the femur in three patients and one case necessary in addition to the soft-tissue surgery, particularly
each of trochanteric bursitis, a sinus over a plate and a in the younger child with significant growth potential.
plaster sore. Unilateral or bilateral hinged acetabuloplasties were carried
THE JOURNAL OF BONE AND JOINT SURGERYHIP SURGERY IN TOTAL-BODY-INVOLVED CEREBRAL PALSY 1165
Fig. 3a Fig. 3b
Photographs showing a) the preoperative clinical picture of windsweep deformity and b) postoperative improvement in hip
symmetry and fixed flexion deformity.
23
out in 36 hips to correct acetabular dysplasia and to indication of acetabular dysplasia because it does not
20 21
improve cover of the femoral head. correlate well with the migration percentage. Radio-
The extent of soft-tissue surgery was tailored to each logical assessment of containment of the abducted hip can
individual patient and depended on intraoperative examina- be difficult and should be supplemented by clinical
tion under anaesthesia to assess fixed flexion and/or abduc- review.
tion/adduction contractures after completion of the bony Functional assessment has not been standardised for
surgery. There was no significant difference in the clinical patients with severe cerebral palsy, but reliable methods
or radiological outcome between those patients who had were selected for this study. Pain can be difficult for an
had previous surgery and those who were being operated unfamiliar observer to determine in patients with little or no
on for the first time. A balanced soft-tissue release allowed verbal communication, but we feel that the close relation-
us to prevent unnecessary disruption of the soft tissues and ship which develops between the patient, the carer and
to obtain a better range of movement while preserving physiotherapist means that this can be assessed, although
stability of the hip. there are no objective criteria available.
The need for acetabuloplasty and the extent of soft-tissue Our results have shown a significant improvement in the
surgery carried out varied between the patients, but there range of abduction and adduction of the hips in flexion,
was a similar distribution of the type of procedure between which is important for seating and nursing care. There was
the dystonic and hypertonic groups, making statistical anal- a significant reduction of fixed flexion deformity without
ysis feasible. loss of the range of flexion. All except one hip had a
Patients with this severe cerebral palsy are difficult to postoperative range of flexion greater than 90°. There was
evaluate. Clinical assessment is complicated by the varia- no significant change in the total range of hip movement.
bility of tone affecting the range of movement at different We feel that the effect of surgery on abduction and adduc-
times. In order to minimise the interobserver and intra- tion, flexion, fixed flexion deformity and hip asymmetry is
observer errors all patients were assessed jointly by the more relevant to the clinical picture than is a “total range of
13
surgeon and the physiotherapist. hip movement” as suggested by some authors.
Radiological evaluation of containment of the hip can be Windsweep has previously been used as a descriptive
assessed by using methods such as the RM%, the CEA or term without any means of definitive clinical evaluation. We
8
the acetabular index, but there may be significant intra- described the WI to aid in the clinical assessment of
observer and interobserver errors because of the established symmetry in such patients, and it was found to be a simple
21,22
deformity of the pelvis and spine. Standardisation of and useful method of evaluating a windsweep deformity. It
the position of the pelvis during radiography is virtually is quick and easy to calculate from the standard chart for
impossible in this group of patients. In our study, we used range of movement and we feel that it adds to the overall
both RM% and CEA to decrease the errors in measurement picture in the assessment of this complicated group of
of hip congruity. We did not use the acetabular index as an patients. Hip symmetry using the WI, however, improved
VOL. 83-B, NO. 8, NOVEMBER 20011166 K. L. OWERS, J. PYMAN, M. F. GARGAN, P. J. WITHEROW, N. M. A. PORTINARO
Fig. 4a Fig. 4b
Radiographs of the pelvis showing the preoperative windsweep deformity
with subluxation of the right hip with acetabular and femoral head
dysplasia (a), after bilateral femoral derotational osteotomies and a right
Pemberton acetabuloplasty, (b) and after removal of metalwork showing
improved radiological joint congruity and symmetry (c).
Fig. 4c
postoperatively in only nine of 21 patients (43%) despite a The rate of supracondylar fracture was low compared with
24
marked improvement in the range of movement in abduction other series.
and adduction. The patients with improved symmetry had There were no significant differences between the pre-
overall a shorter median duration of follow-up which sug- operative and postoperative changes for the dystonic and
gests that there may be a deterioration in symmetry with hypertonic groups in all parameters assessed. Dystonic
time. A longer follow-up will be necessary to confirm this. patients have often received minimal treatment in previous
There was a statistically significant improvement in both series because of the unpredictability of their response to
radiological indices of hip containment after surgery (Fig. surgery. This has not been shown to be the case so far in
4). Resubluxation was more common in the dystonic group our patients.
and occurred in younger patients in the hypertonic group. These early results, at a median three-year follow-up,
Surgery was apparently effective in relieving pain in the suggest that bilateral simultaneous combined soft tissue and
hip in 11 of 13 patients (85%), although objective assess- bony surgery should be considered in this challenging
ment of pain in this group of patients is not possible. group of children with cerebral palsy.
Our management appears to have some beneficial effect The authors would like to thank, for their assistance, S. L. Whitehouse and
on the mobility of these patients, although not to a sig- K. Waters of the North Bristol NHS Trust.
No benefits in any form have been received or will be received from a
nificant extent. Ease of patient handling improved and commercial party related directly or indirectly to the subject of this
satisfaction of the carers with the outcome of surgery was article.
high. In a situation where there has been considerable input References
from all members of the multidisciplinary team, this may 1. Campbell AGM, McIntosh N, eds. Forfar and Arneil’s textbook of
be a predictable sentiment whatever the clinical outcome. paediatrics. Edinburgh: Churchill Livingstone, 1998:738.
THE JOURNAL OF BONE AND JOINT SURGERYHIP SURGERY IN TOTAL-BODY-INVOLVED CEREBRAL PALSY 1167
2. Samilson RL, Tsou P, Aamoth G, Green WM. Dislocation and 14. Sharrard WJW, Allen JMH, Heaney SH, Prendiville GRG. Surgi-
subluxation of the hip in cerebral palsy: pathogenesis, natural history cal prophylaxis of subluxation and dislocation of the hip in cerebral
and management. J Bone Joint Surg [Am] 1972;54-A:863-73. palsy. J Bone Joint Surg [Br] 1975;57-B:160-6.
3. Tylkowski CM, Rosenthal RK, Simon SR. Proximal femoral osteot- 15. Miller F, Cardoso Dias R, Dabney KW, Lipton GE, Triana M.
omy in cerebral palsy. Clin Orthop 1980;151:183-92. Soft-tissue release for spastic hip subluxation in cerebral palsy. J
4. Pritchett JW. The untreated unstable hip in severe cerebral palsy. Clin Pediatr Orthop 1997;17:571-84.
Orthop 1983;173:169-72.
16. Hoffer MM, Stein GA, Koffman M, Prietto M. Femoral varus-
5. Howard CB, McKibbin B, Williams LA, Mackie I. Factors affecting derotation osteotomy in spastic cerebral palsy. J Bone Joint Surg [Am]
the incidence of hip dislocation in cerebral palsy. J Bone Joint Surg 1985;67-A:1229-35.
[Br] 1985;67-B:530-2.
17. Bos CFA, Rozing PM, Verbout AJ. Surgery for hip dislocation in
6. Eilert RE. Editorial: Hip subluxation in cerebral palsy: what should be cerebral palsy. Acta Orthop Scand 1987;58:638-40.
done for the spastic child with hip subluxation? J Pediatr Orthop
1997;17;5:561-2. 18. Brunner R, Baumann JU. Clinical benefit of reconstruction of
7. Lonstein JE, Beck K. Hip dislocation and subluxation in cerebral dislocated or subluxated hip joints in patients with spastic cerebral
palsy. J Pediatr Orthop 1986;6:521-6. palsy. J Pediatr Orthop 1994;14:290-4.
8. Portinaro NMA, Owers KL, Pyman J, Gargan MF. The windsweep 19. Brunner R, Baumann JU. Long-term effects of intertrochanteric
index: a method of evaluating pelvic asymmetry and monitoring the varus-derotation osteotomy on femur and acetabulum in spastic cere-
efficacy of treatment in cerebral palsy. Hip Int 2000;10;3:170-3. bral palsy: an 11- to 18-year follow-up study. J Pediatr Orthop
9. Reimers J. The stability of the hip in children: a radiological study of 1997;17:585-91.
the results of muscle surgery in cerebral palsy. Acta Orthop Scand 20. Mubarak SJ, Valencia MD, Wenger DR. One-stage correction of the
1980:Suppl 184. spastic dislocated hip: use of pericapsular acetabuloplasty to improve
10. Cornell MS, Hatrick NC, Boyd R, Bird G, Spencer JD. The hip in coverage. J Bone Joint Surg [Am] 1992;74-A:1347-57.
children with cerebral palsy: predicting the outcome of soft tissue 21. Brunner R, Robb JE. Inaccuracy of the migration percentage and
surgery. Clin Orthop 1997;340:165-71. centre-edge angle in predicting femoral head displacement in cerebral
11. Wiberg G. Studies on dysplastic acetabula and congenital subluxation palsy. J Pediatr Orthop B 1996;5:239-41.
of the hip joint: with special reference to the complication of osteo-
arthritis. Acta Chir Scand 1939;83:Suppl. 58. 22. Portinaro NM, Murray DW, Bhullar TP, Benson MK. Errors in
measurement of acetabular index. J Pediatr Orthop 1995;15:780-4.
12. Cooperman DR, Bartucci E, Dietrick E, Millar EA. Hip dislocation
in spastic cerebral palsy: long term consequences. J Pediatr Orthop 23. Sharp IK. Acetabular dysplasia: the acetabular angle. J Bone Joint
1987;7:268-76. Surg [Br] 1961;43-B:2:268-72.
13. Barrie JL, Galasko CSB. Surgery for unstable hips in cerebral palsy. 24. Stasikelis PJ, Lee DD, Sullivan CM. Complications of osteotomies
J Pediatr Orthop B 1996;5:225-31. in severe cerebral palsy. J Pediatr Orthop 1999;19:207-10.
VOL. 83-B, NO. 8, NOVEMBER 2001You can also read
