SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Aim  The purpose of this study was to evaluate a population-based radiographic hip surveillance programme for children with cerebral palsy (CP) and to assess the natural history of hip displacement.

Method  The study comprised 335 children (188 males, 147 females), born during 2002 to 2006 in the 10 south-eastern counties in Norway. Their mean age at the first radiograph was 3 years (range 6mo–7y 11mo) and the mean age at the most recent follow-up was 5 years 5 months. Distribution according to CP type was spastic hemiplegia in 38%, diplegia in 27%, quadriplegia in 21%, dyskinesia in 10%, and ataxia in 3%; Gross Motor Function Classification System (GMFCS) levels I to V were, 44%, 14%, 8%, 11%, and 23% respectively. Migration percentage (MP), acetabular index, and pelvic obliquity were measured on the radiographs.

Results  Hip displacement (MP>33%) occurred in 26% of all children (subluxation in 22% and dislocation in 4%) and in 63% of those in GMFCS levels IV or V. Dislocation occurred in 14 children at a mean age of 4 years 5 months (range 1y 10mo–9y 7mo). The mean migration percentage was 20.4% at the initial radiographs and 34.0% at the last follow-up. Mean progression in migration percentage increased markedly with decreasing functional level, from 0.2% per year at GMFCS level I to 9.5% at level V.

Interpretation  There is a pronounced trend towards hip displacement in nonambulant children. Close surveillance from age 1 to 2 years is needed to find the appropriate time for preventive surgery. Since 12% of the nonambulant children developed dislocation, our routines for hip surveillance need improvement.


What this paper adds

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  •  Hip displacement occurred in 63% of children in GMFCS levels IV or V.
  •  There was a pronounced increase in lateral hip migration (9.5% per year) in GMFCS level V.
  •  Radiographic hip surveillance from age 1 to 2 years is needed for non-ambulatory children.

There is increased risk of hip dysplasia and dislocation in cerebral palsy (CP) and the risk is highest in children with the most severe forms of CP and in patients without gait function.1–3 Dislocation may lead to pain and severe problems with ambulation, sitting balance, perineal nursing care, and decubitus ulceration.4,5 Therefore, special screening programmes aimed at early diagnosis and treatment of hip displacement to prevent complete dislocation have been developed.2,6 These programmes have provided important information regarding risk factors for hip discplacement and the effect of early diagnosis. There is, however, still a need for further research in this field, since the rate of deterioration of the hips has not been established. This is important because the effectiveness of preventive treatment cannot be properly assessed unless the natural history of hip displacement is fully understood.

Based on the experience from the Swedish CP hip surveillance programme,6 a systematic population-based registration of children with CP was initiated in southeast Norway in 2006. The aims of the present prospective study on hip development were to (1) evaluate the quality of this screening programme; (2) assess the natural history of hip displacement during the first years of life according to functional level and type of CP, and (3) define risk factors for deterioration.

Method

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Participants

In January 2006 the CP follow-up programme (CPOP) for southeast Norway was started, encompassing about 50% of the population in the country. All children with CP born after 1 January, 2002 and living in one of the 10 south-eastern counties were invited to participate in a follow-up programme of motor function and were included in the programme after informed consent from their parents. The follow-up programme had approval from the Norwegian Data Inspectorate. The programme is led and coordinated by two physiotherapists and an occupational therapist at our hospital, where the database with patient information is stored. The diagnosis and type of CP were determined by neuropaediatricians working at the child habilitation centres (one in each county). The CPOP programme includes a systematic clinical and radiographic follow-up.

At the end of 2011, the CPOP register contained 389 children born during the 5-year period 2002 to 2006. This prospective radiographic study on hip development comprised 335 children, 147 females and 188 males. The reason for lacking radiographs of the remaining children (14%) was that the routines were changed from 2009, after which children with near normal gait function did not need to have radiographs taken of their hips. Radiographs were available for all non-ambulatory children (Gross Motor Function Classification System [GMFCS] levels IV and V). Of the 335 children, 285 (85%) had spastic CP, of whom 69 had quadriplegia, 89 had diplegia, and 127 had hemiplegia. The numbers with dyskinesia and ataxia were 34 and 11 respectively, and in five children the CP type had not been definitively determined.

The functional level of the children was determined by physiotherapists in the habilitation centres, who examined the children twice a year up to the age of 6 years and then once a year. Gait function was assessed using the five levels of the GFMCS7,8 with decreasing functional level as the GMFCS class increases. The distribution of patients from level I to V was 44%, 14%, 8%, 11%, and 23%. There was a strong correlation (p<0.001) between the type of CP and gait function, as none of the children with quadriplegia could walk, whereas all those with hemiplegia could walk (two with support). Of those with diplegia, 52 children walked without support, 20 had to use support, and 17 had no gait function. Only one child with dyskinesia could walk without support, whereas six walked with support and 27 could not walk. Ten of the 11 children with ataxia walked without support and one child was in GMFCS level IV.

Radiographic examination and follow-up routines

Radiographic examination was carried out at the local county hospital according to a specified procedure, to ensure that the same technique was used. An anteroposterior radiograph of the pelvis and hip joints was obtained with the child in the supine position. Care was taken to position the child correctly with the legs parallel and to avoid rotation of the pelvis and legs. All radiographs were sent (by CD or electronically) to the physiotherapist who coordinated the screening programme. The radiographs were transferred to our Picture Archiving and Communication System (PACS; Sectra, Linköping, Sweden) and measured by the author, who has many years of experience in evaluating and measuring radiographs of children’s hips. The radiographs were twofold enlarged in order to obtain better visualisation of the landmarks and measurements were performed digitally with the standard equipment in PACS.

The following parameters were measured: migration percentage9 acetabular index,10 and pelvic obliquity (Fig. 1). Migration percentage is the percentage of the femoral head lateral to the acetabulum (lateral to Perkins’ line), measured parallel to Hilgenreiner’s line. When the lateral margin of the femoral head was medial to Perkins’ line and the migration percentage is in fact a negative value, it was given the value 0%. When the whole femoral head was lateral to Perkins’ line, the migration percentage was registered as 100%. Depending on their migration percentage, the hips were classified as normal (MP under 33%), subluxation (MP 33–89%), and dislocation (MP 90% or higher). Based on the initial and the last radiograph, the progression in migration percentage per year was calculated.

image

Figure 1.  (a) Anteroposterior radiograph of a 4 year 2 months old female with spastic quadriplegia, showing the radiographic measurements. Migration percentage is the lateral displacement of the femoral head (a/b×100). Migration percentage was 33% in the right hip and 29% in the left. The other radiographic parameters are acetabular index (AI) and (pelvic obliquity (PO). (b) Radiograph of the same female as in (a) after a follow-up period of 16 months. There is subluxation of the left hip (MP 39%) whereas the right is within the normal range (29%).

Download figure to PowerPoint

The acetabular index is the slope of the acetabular roof, which is the angle between the acetabular roof and Hilgenreiner’s line. Pelvic obliquity is the angle between the horizontal line and the line between the lowest points of the pelvic bones on the right and left side (Fig. 1). Angles less than 3° were not registered as pelvic obliquity.

According to the study protocol a pelvic radiograph should be taken shortly after diagnosis, preferably at the age of 1 year, in children with pronounced spasticity. For all other children, a radiograph at the age of 2 years should be obtained. Follow-up radiographs should be taken once a year until the age of 8 years in children with spastic bilateral CP (diplegia and quadriplegia) and dyskinetic CP, whereas a follow-up radiograph at age 4 years should be taken in children with spastic hemiplegia and ataxia and independent walking ability (GMFCS level I–II). From January 2009, the follow-up routines were changed in accordance with the revised guidelines for the Swedish CP programme.11 Children in GMFCS levels III to V should now have annual radiographs as before, whereas those in level II should have radiography at age 2 years and 6 years. Children in level I did not need radiographic examination, provided that the clinical examination did not arouse suspicion of a hip displacement.

The children were followed up until operative treatment for hip displacement or until the most recent radiograph in those who had not undergone hip surgery. Three children had died and three had been operated on before a follow-up radiograph had been taken. In the remaining, follow-up radiographs were available in 223 children, all of those in GMFCS level V, 92% of children in level IV, 93% of those in level III, 76% of those in level II, and 38% of children in level I.

Statistical analysis

To assess the statistical differences between groups, Student’s t-tests and χ2 test were used when two groups were compared and analysis of variance (ANOVA) was used when more than two groups were compared. Correlation between parameters was evaluated by Pearson’s correlation coefficient (r). Differences were considered significant when the p-value was <0.05.

Results

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Prevalence of hip displacement

The mean age of all children at the time of the primary radiographs was 3 years (range 6mo–7y 11mo). Of the 115 non-ambulant children in GMFCS level IV or V, 25% had their first radiograph under the age of 1 year 6 months, 64% under 3 years, and 78% under 4 years. There was a reduction in mean age from 4 years 2 months for non-ambulant children born in 2002 to 1 year 5 months for those born in 2006.

Eighty-nine of the 335 children (26%) had hip displacement (subluxation in 22% and dislocation in 4%) at the initial radiographic examination or at follow-up. There was no difference in prevalence between females and males (p=0.99). The mean patient age at the first radiograph showing subluxation was 3 years 7 months (range 8mo–7y 6mo). Complete dislocation occurred in 14 children (12% of the non-ambulators) at a mean age of 4 years 5 months (range 1y 10mo–9y 7mo).

The development of the hips was strongly associated with functional level. The percentage with normal hips decreased from 99% to 28% with increasing GMFCS level from I to V (Table I). Pronounced subluxation with a migration percentage in the range 50% to 89% was not seen in children with independent gait function and occurred with increasing frequency in children in GFMCS level III (3%) to level IV (13%) and level V (32%). Complete dislocation was seen only in GFMCS level IV (one patient) and level V (13 patients).

Table I.   Radiographic diagnosis according to functional level and CP type, given as percentage of patients
ClassificationRadiographic diagnosis n
NormalSubluxationDislocation
MP 33–39%MP 40–89%
  1. n, number of children; GMFCS, Gross Motor Function Classification System; MP, migration percentage.

GMFCS level
 I99100146
 II9244046
 III612118028
 IV55537338
 V2812431777
CP type
 Hemiplegia97300127
 Diplegia78616089
 Quadriplegia1912521769
 Dyskinesia76912334
 Ataxia10000011

Hip displacement differed markedly in relation to type of CP (Table I). In children with hemiplegia, all had normal hips except 3% who had slight subluxation. All children with ataxia had normal hips. The prognosis was worse in dyskinesia with 24% of the children having abnormal hips. Of the children with spastic diplegia, 22% had subluxation. Children with quadriplegia had the highest proportion of abnormal hips with 64% subluxation and 17% dislocation.

Of the 115 children in GMFCS level IV or V, the proportion who developed abnormal hips was related to type of CP. Whereas 81% (56 of 69 children) with quadriplegia developed hip displacement, this occurred in 47% of those with diplegia and in 26% of those with dyskinesia. This shows that it is the bilateral spastic CP types that are particularly at risk of hip displacement.

Progression of radiographic measurements

Taking the side with the largest displacement (the ‘worst’ hip), the mean migration percentage at the initial radiograph was 20.4% (range 0–100%). Migration percentage was strongly related to CP type and GMFCS level (Table II). There was a significant correlation in migration percentage between the right and left hip (r=0.67; p<0.001).The mean length of follow-up was 2 years 9 months (range 6mo–7y 3mo) and the mean age of the patients at the most recent follow-up radiograph was 5 years 5 months (range 2y–9y 7mo). At the last follow-up, the mean migration percentage of the worst side had increased to 34% (range 0–100%), again with a strong association with CP type and functional level.

Table II.   Migration percentage (%) initially and at follow-up, and migration percentage progression per year (mean and SD) in relation to GMFCS level and CP type (only the side with the largest migration percentage is included)
ClassificationNumber of patientsF-U, yMigration percentage
InitialFollow-upInitialFollow-upProgression per year
  1. GMFCS, Gross Motor Function Classification System; F-U, follow-up; CP, cerebral palsy.

GMFCS level
 I146551.914.4 (8.9)17.3 (8.7)0.2 (3.7)
 II46352.917.4 (11.9)22.7 (9.5)1.2 (3.2)
 III28262.926.5 (10.7)30.8 (9.6)1.3 (3.1)
 IV38353.026.2 (20.2)36.2 (20.0)3.9 (4.8)
 V77723.128.6 (24.3)52.3 (29.4)9.5 (9.4)
CP type
 Hemiplegia127512.114.7 (9.3)17.2 (9.7)0.2 (3.7)
 Diplegia89672.820.4 (12.6)28.2 (14.0)1.8 (4.2)
 Quadriplegia69673.034.1 (25.2)56.7 (26.4)9.2 (8.4)
 Dyskinesia34313.114.9 (12.2)27.7 (19.4)5.0 (9.0)
 Ataxia1162.214.3 (10.6)23.7 (2.3)2.2 (2.9)

During the follow-up period the change in migration percentage of the side with the largest progression was called maximal progression. The mean maximal progression was of 4% (range −9% to 49%) per year. The progression was more than 20% per year in eight children (all non-ambulators). Migration percentage progression increased with decreasing functional level (Table II), from 0.2% per year for those in GMFCS level I to 9.5% per year for children in GMFCS level V. The differences between children with no gait function (GMFCS levels IV and V) and the groups with gait function were significant (p<0.001).

There was a clear association between deterioration of hip displacement and type of CP, as the maximal migration percentage progression per year was 9.2% in quadriplegia, 1.8% in diplegia, 0.2% in hemiplegia, and 5.0% in dyskinesia. The differences between quadriplegia and diplegia and hemiplegia were highly significant (p<0.001).

The progression in migration percentage decreased with age (r=−0.22; p=0.001). Children under 2 years 6 months of age at the first radiograph had larger maximal migration percentage progression per year (5%) than older children (2.9%; p=0.033). In children without gait function, progression in migration percentage was larger in those under 4 years at the first radiograph (8.6%) than in older children (3.8%; p=0.023).

There was no significant correlation between initial migration percentage and maximal migration percentage progression per year (r=0.15; p=0.14). Of the 13 children who had initial subluxation with migration percentage of 50% or more, one underwent surgical correction and one died before follow-up. Of the remaining 11 children, three had a change in migration percentage of less than 10% (from 50% to 43%, from 50% to 47%, and from 66% to 74% respectively). Eight children had deterioration (more than 10% increase in migration percentage) with complete dislocation in four children and increasing subluxation in four. Of the 14 children with initial migration percentage between 40% and 50%, one deteriorated to dislocation, five had increasing subluxation, six were mainly unchanged (<10% change in MP), one improved (from MP 41% to 24%) and one underwent surgery before follow-up.

Twenty-one children had slight subluxation initially (MP 33% to 39%). At the last follow-up, deterioration had occurred in nine children, dislocation in two, and increased subluxation (MP over 40%) in seven. Five children still had slight subluxation at follow-up. One child was operated upon before follow-up and one had no follow-up. The hips of five children (three in GMFCS level II and two in level III) improved from slight subluxation to normal. Five children, who had normal hips at the initial examination, developed slight subluxation during follow-up, but had hips within the normal range at the last examination. Thus, the hips of 10 children improved from slight subluxation to normal during the follow-up period.

At the initial examination, 32 children had unilateral abnormal hips and 20 had bilateral displacement. The maximal migration percentage progression per year was 3.4% in children with initially normal hips, 6.0% in patients with unilateral displacement, and 6.6% in those with bilateral affection. The differences between children with bilateral normal hips and the two groups with abnormal hips were not statistically significant. Neither was there any significant correlation between side difference in migration percentage at the initial examination and yearly progression in migration percentage (r=0.09; p=0.20).

The mean acetabular index (the side with the highest angle) was 20.5° at the primary radiographs and 21.3° at the last follow-up (p=0.37). Initial acetabular index was higher in patients who developed subluxation (24.2°) and dislocation (29.3°) than in hips that remained normal (18.9°; p=0.001). There were significant correlations between acetabular index and migration percentage at the initial examination (r=0.71; p<0.001) and at follow-up (r=0.80; p<0001). In children with migration percentage at the primary radiograph of 40% or higher, 55% of these had an acetabular index of more than 27° and 31% had 30° or higher, showing that acetabular index was not reliable as an indicator of hip displacement at this early age. At follow-up, 47 children had an acetabular index of 27° or higher and hip displacement was seen in all except two. Of 24 children with acetabular index of 30° or higher, all had displacement.

The mean pelvic obliquity was 4.9° primarily and 5.3° at follow-up (p=0.19). Initial pelvic obliquity was higher in children who developed dislocation (7.5°) than in those who developed subluxation (4.7°; p=0.035) and those with continuous normal hips (4.7°; p=0.033), whereas there was no difference between the latter two groups (p=0.98). There was a significant correlation between initial pelvic obliquity and initial migration percentage of the worst hip (r=0.23; p=0.005) and migration percentage at follow-up (r=0.33; p=0.001).

Children with dislocation

All the 14 children who developed complete hip dislocation were non-ambulators (GMFCS level IV in one child and level V in 13). CP type was spastic quadriplegia in 12, dyskinesia in one and unspecified CP in one. Four children had dislocation already at the initial radiograph at an age varying from 1 year 10 months to 3 years. The remaining children had subluxation between 1 year 1 month and 4 years 8 months and dislocation at a median age of 4 years 2 months (nine were under 6y and one was 9y). The mean primary migration percentage of these 10 children was 45.5% (range 12–88%) and the progression in migration percentage was 23.7% (range 8.9–49.1%) per year. The most frequent reasons for developing dislocation were an excessively long time on the waiting list for surgery (five children) and unexpectedly rapid deterioration with migration percentage increase greater than 30% per year (two children).

Discussion

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Natural history of hip displacement

Population-based patient data is needed for an adequate study of the natural history of hip displacement in CP. Because the children were included in a CP surveillance programme including 10 counties, the children in this study represent a true cross-section of the CP population. The distribution of functional level was in accordance with that of previous population-based studies.2,11,12 Thirty-four per cent of the present children could not walk (GMFCS levels IV and V), which is similar to the 34% and 31% reported, respectively, by Soo et al.2 and Connelly et al.,12 whereas Hägglund et al.11 had a lower proportion of non-ambulators (27%).

Prevalence of hip displacement depends on functional level and type of CP. According to recent population-based studies,2,6,12 hip displacement is directly related to GMFCS levels. This was confirmed by the results of the present study (Table III). Compared with Hägglund et al.11 who also used a migration percentage of 33% as the threshold value, the present study had somewhat lower prevalences of hip displacement in GMFCS levels I to IV but slightly higher prevalence in GMFCS level V. This was somewhat unexpected because the present patients were followed up to an age of less than 9 years whereas the patients of Hägglund et al.11 were followed until age 9 to 16 years. The reason is probably that hip displacement starts at an early age in non-ambulators and subluxation has usually already occurred before the age of 5 years. Table III shows that the frequency of hip displacement is less than 5% in children with nearly normal walking ability and increases to more than 60% in children with no walking capacity. In children in GMFCS levels IV or V, hip displacement was more frequent in spastic bilateral CP than in dyskinesia, indicating that spasticity is an important aetiological factor. In children with hemiplegia and ataxia, who both have good gait function, hip displacement was very rare, confirming previous studies.2,11

Table III.   Percentage of patients with hip displacement (subluxation or dislocation) according to functional level; a comparison between three previous studies and the present study
AuthorsMP limit (%)Gross Motor Function Classification System level
IIIIIIIVV
  1. MP limit, cut-off in migration percentage between normal hips and hips with displacement.

Soo et al.230015416990
Connelly et al.1230317465976
Hägglund et al.1133513506268
Present study 20123318394572

Hip displacement is more frequent in quadriplegia than in diplegia.1,3,13,14 This was supported by the present results where the progression in migration percentage was more than four times as large in quadriplegia. Reimers9 found a mean yearly migration percentage increase of 10% in spastic CP, but did not specify which subtypes the patients belonged to. A migration percentage increase of 7% to 9% per year in quadriplegic bedridden patients was reported by Vidal et al.13, which is in keeping with the present 9.2% progression in spastic quadriplegia. In the diplegic group the mean migration percentage progression per year was 1.8%, which is less than the 4% per year in children with a potential for independent walking reported by Vidal et al.13

One limitation of this study is that the follow-up was not long enough for the true prevalence of hip displacement to be seen. However, subluxation occurs before the age of 5 years in most children and should therefore be detected by the present follow-up. Another limitation is that the results of surgical treatment were not included. The reason is that a longer postoperative follow-up is needed and these results will be published later. Since early surgical treatment is not a guarantee for a successful final outcome at skeletal maturity,15 the effect of the present screening programme cannot yet be fully evaluated.

The mean age at dislocation is 6 to 7 years in clinical series.3–5 Actually, the age at dislocation is usually lower, as shown by systematic hip surveillance. Of 15 children with dislocation during a hip screening program,12 three had dislocation under 3 years of age. This is in accordance with the present study, where the mean age at dislocation was 4 years 5 months and four children were under 3 years.

Radiographic parameters and cut-off values

Previous studies have disagreed concerning the most adequate radiographic parameter in hip screening. Whereas Cooke et al.14 found that the acetabular index was the most powerful single radiographic predictor for hip displacement, Vidal et al.13 maintained that acetabular index should not be considered as a prognostic indicator below the age of 5 years. Hägglund et al.6 maintained that hip displacement preceded acetabular dysplasia and this was confirmed by the present results. Migration percentage measures the most important single qualification of a hip joint, which is an adequate coverage of the femoral head by the acetabular roof. In the present study, the initial migration percentage was of highly significant prognostic value for the development of hip displacement. Because this is an easy measurement and is little influenced by the rotational position of the femur9 and since it has sufficient intra- and interobserver reproducibility,16 most authors agree that migration percentage is appropriate as the main radiographic parameter in screening for hip displacement.2,6,9,13,17

Acetabular dysplasia is assessed by the acetabular index. Vidal et al.13 reported that acetabular deformity in CP does not develop until after the age of 30 months. In the present study, acetabular index was a significant prognostic factor for hip displacement in children less than 30 months as well as in older children. Acetabular index of more than 27°, and especially over 30°, was clearly associated with hip displacement. Scrutton et al.18 found a correlation between acetabular index and migration percentage which became stronger as the child approached 48 months. However, a disadvantage with acetabular index is that, in children with fixed flexion deformity causing anterior pelvic tilt, the landmarks used for measurement of the acetabular roof can be difficult or impossible to determine. The pelvis can be corrected to an adequate position by flexing the hips, by placing pillows under the legs.16 This is, however, difficult to implement as standard routine in all hospitals included in a multicentre screening programme. Acetabular index should be considered as a supplement to migration percentage and is useful when deciding the correct strategy for surgical treatment. Pelvic obliquity is unnecessary in routine screening of hips, but is a relevant measurement when associated with unilateral hip displacement and scoliosis with convexity to the opposite side.

Regarding grading of hip displacement, Reimers,9 who was the first to use migration percentage in the evaluation of hips in children with CP, recommended cut-off values of 33% and 90%, respectively, for subluxation and dislocation, and these limits have been used in most studies.3,6,18 Cut-off values of 30% for subluxation and 100% for dislocation have been used in some recent studies.2,12,19 Robin et al.19 increased the categories of abnormal hips from two to four. So far, no benefits of using these new cut-offs and gradings have been documented. Comparison between different studies is easier and more reliable when the same grading system is used and the grading of Reimers9 is still adequate and easy to use.

It would be an advantage if there was a clear cut-off value of migration percentage, so that percentages above this could serve as indication for preventive surgery to avoid dislocation, but no consensus on this point exists. Several patients with slight subluxation and gait function (GMFCS levels II and III) had improvement during follow-up and ended up with hips within the normal range. This has also been reported by others.6,12 It seems reasonable to consider hips with migration percentage in the range 33% to 39% as ‘hips at risk’6 and postpone the decision about surgical treatment until the natural history has been better established by longer follow-up. Miller and Bagg20 found that all hips with a migration percentage in the range 60% to 90% went on to dislocation and the present results showed that eight of 11 children with a migration percentage above 50% deteriorated to dislocation or severe subluxation. As an indication for preventive surgery we use a migration percentage value of about 40%, in accordance with others. 11,12,17

Routines and quality of hip joint screening

The pronounced differences in hip displacement between different functional levels indicated that the follow-up routines should vary. In non-ambulators (GMFCS levels IV and V), close follow-up was necessary, especially in those younger than 4 years who had the greatest migration percentage progression. Since subluxation occurred in many patients under age 3 years and even dislocation was seen in some, it is reasonable to start with an anteroposterior radiograph between 1 and 2 years or at the time of CP diagnosis, as proposed by Hägglund et al.,6 rather than taking the first radiograph at 30 months, as recommended by others.2,18 In non-ambulators repeat radiographs once a year has been a recommended procedure, although radiographs every 6 months was proposed by Soo et al.2 for spastic non-ambulators. The frequency of radiographic follow-up is also dependent on the rate of progression. If the progression in migration percentage is more than 15% to 20% per year, follow-up after 6 months seems reasonable.

In children in GMFCS levels II and III, such frequent follow-up is not necessary. Based on the slight annual increase in migration percentage, a radiograph at age 4 to 5 years seems sufficient. If migration percentage is within the normal range and hip abduction is satisfactory (more than 30°), further radiography is hardly necessary. In children with nearly normal gait function (GMFCS level I) a radiograph is needed only if there is clinical suspicion of hip abnormality.

The main goal of systematic hip screening is to prevent complete hip dislocation. The Swedish surveillance programme seems to be successful (although the final outcome of surgical treatment was not reported), since only one child, whose parents refused surgical treatment, developed dislocation.6 The present results were not that good, since dislocation developed in 4% of the children, which is in keeping with 6% to 7% dislocation in other screening programmes with longer follow-up.2,12

In our surveillance programme, patient age at the initial radiograph decreased during the study period, which indicates that the aim of early radiographic examination was better achieved after the routines had been practised for some time; this is promising for our continued screening. We had problems, however, regarding follow-up routines and needed to make requests to the habilitation teams about radiographs that had not been taken according to the guidelines or had not been sent to our central coordinator. It would be best if the radiographs were electronically transferred from the local hospitals to the orthopaedic surgeon who performs the measurements, but we have not reached that goal yet. Another weakness was the excessively long waiting period for surgery, which usually was more than 6 months and sometimes over 12 months and contributed to complete dislocation in some patients. Thus, the challenges for our screening programme are improvement of the routines for radiographic follow-up and increased resources for surgical treatment.

Acknowledgements

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The author thanks physiotherapist Gerd Myklebust, who coordinated the CPOP programme of the lower extremities and helped to collect the radiographs. He is also grateful to the physiotherapists of the child habilitation teams who coordinated the radiographic screening.

References

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References