Abstract
Optimum management of hip displacement in children with cerebral palsy (CP) is facilitated by an approach that focuses on anticipatory and preventive measures. Hip surveillance programs for children with CP were developed at the beginning of the new millennium, with the purpose of identifying hip displacement sufficiently early to permit a choice of effective management options. In the early years, hip surveillance was guided by epidemiological analysis of population-based studies of prevalence. In Australia, a National Hip Surveillance in CP Working Group was first convened in 2005. This resulted in a 2008 Consensus Statement of recommendations published and endorsed by Australasian Academy of Cerebral Palsy and Developmental Medicine (AusACPDM). The group undertook that the recommendations should be reviewed every 5 years to ensure currency and congruency with the emerging evidence base. As new evidence became available, hip surveillance guidelines developed, with the most recent 2020 Australian Hip Surveillance Guidelines endorsed by the AusACPDM. Implementing comprehensive hip surveillance programs has now been shown to improve the natural history of hip dislocations and improve quality of life. Standardised hip surveillance programs can also facilitate planning for multicentre research through harmonisation of data collection. This, in turn, can help with the identification of robust new evidence that is based on large cohort or population studies. Here a review of evidence informing the updated 2020 Hip Surveillance Guidelines is presented.
What this paper adds
Hip dislocation is a major contributor of pain and reduced quality of life in children with cerebral palsy (CP), particularly those with the most severe impairments.
Since the commencement of hip surveillance, the incidence of hip dislocations in children at risk has decreased.
Early identification of hips at risk of progressive displacement through hip surveillance could have an important impact on health-related quality of life for the children and burden of care for their families.
Background
Progressive musculoskeletal deformity remains a significant cause of disability and pain in people living with CP [1, 2]. One of the most prevalent and difficult to manage musculoskeletal complications of CP is hip dislocation [3]. Painful hip dislocations result in considerable reduction in quality of life [4]. Prevention of hip dislocation is a musculoskeletal priority for children with CP.
By 2005, surveillance was starting in a range of formats in Australia. In 2008, a nominal expert group of clinicians and researchers in hip displacement in CP from across Australia (called the Australian National Hip Surveillance Working Group) formulated the ‘Consensus Statement on Hip Surveillance for Children with Cerebral Palsy: Australian Standards of Care’ with a commitment to review and revise every 5 years [5]. The initial version was called a Consensus Statement due to the lack of published evidence on which to base recommendations. In 2014 the group conducted a systematic review that identified additional evidence that hip surveillance programs could prevent hip dislocations in children with CP [6]. This new knowledge enabled the document to be defined as a guideline [7].
In 2019, an updated review of the 2014 Guideline commenced, resulting in the ‘Australian Hip Surveillance Guidelines for Children with Cerebral Palsy 2020’ [8]. This continued the effort to design an evidence-based, comprehensive, succinct compilation of practical guidance for the prevention of hip dislocation in children with CP. Optimal dissemination and practice uptake of information was promoted using concise language and familiar formatting. A consensus development method was used to update the 2020 Guideline. This consisted of a review of relevant publications relating to the recommendations contained in the hip surveillance guidelines and recommended changes to the 2014 Guidelines by an expanded Australian National Hip Surveillance Working Group.
What is the purpose of hip surveillance guidelines?
Hip surveillance guidelines provide evidence-based recommendations for the routine surveillance of hip displacement for all children with CP with an overarching aim of preventing hip dislocations, reducing pain and maintaining quality of life. They are based on the key principle of minimising radiation exposure for children in lower risk categories but enabling early identification of children at higher risk.
Hip surveillance guidelines also ensure that the education and knowledge provided to families regarding the risk for their child’s hips is evidence-informed and congruent with national and international practices. This allows informed choice and assists in engagement with services.
Hip surveillance guidelines also provide the means to standardise monitoring programs across Australia to enable harmonisation of data collection that can be used for future multicentre research.
Hip surveillance programs are effective
Provision of an effective hip surveillance program requires multiple factors to ensure its success. This includes good quality radiographic services and clinical databases to track and monitor children over time to ensure timely imaging and assessment as well as both referral and access to a supportive, expert orthopaedic service.
Population-based studies have shown that hip surveillance programs are effective at reducing hip dislocations. Prevalence of hip displacement fell from 10.0% to 4.5% and dislocation fell from 2.5% to 1.3% in the 5 years since surveillance began in one study [9], and dislocation fell from 9% to 0% in a 10 year review of surveillance in another study [10].
Children who have accessed hip surveillance and responsive, proactive paediatric orthopaedic services have improved hip health into young adulthood compared with children who did not [11, 12].
Main findings of the updated review of the guidelines
The summary of changes from the 2014 to the 2020 Guideline is provided in Supplementary material.
The Guideline should be based on Gross Motor Function Classification to stratify risk
There are now a number of studies confirming the linear relationship of Gross Motor Function Classification System (GMFCS) levels and risk of hip displacement, with broadly similar prevalence rates [6, 13–18]. Newly published data confirmed children classified at GMFCS IV and V were at highest risk of developing hip displacement [6, 19–21]. In the 2014 review of the guidelines, Australia-wide data was used to inform recommendations for hip surveillance for children at GMFCS level I, due to the paucity of published data [22]. There is now additional evidence that children at GMFCS level I have a negligible risk of dislocation [6, 23]. A paucity of data relating to hip risk in children with hemiplegia Winters, Gage, and Hicks (WGH) gait classification [24] continues, despite ongoing concern of increased risk of symptomatic dysplasia (rather than dislocation) in adolescents in this group [3, 6].
Migration percentage (MP) is reliable and clinically useful to measure hip displacement for surveillance programs
Since the first consensus statement was published, Reimer’s migration percentage (MP) [25] remains the radiological measure of choice for detecting and predicting risk of hip displacement because of its good interrater reliability [21, 26–28], and clinical utility [29]. Additional evidence for utility is the use of MP in population-based studies of prevalence [9, 31] and hip displacement prediction studies [13, 32].
A new method for measurement of MP that takes pelvic obliquity (PO) into consideration has been proposed (pelvic adjusted, migration percentage: PAMP) [28, 32]. Presence of PO was found to influence measurement of the MP once PO reaches 10 degrees or more [27, 33]. The effect of PO on frequency of routine hip surveillance may be small because the majority of younger children with CP have a level pelvis or PO less than 5 degree [27, 33]. The PAMP may be a better measure than traditional MP when PO is greater than 10 degrees, and this is an area requiring further study. Once PAMP has been used, it should be used consistently for that child. These changes have been included in the revised surveillance recommendations [27].
Data from three previously summarised studies reported that femoral neck anteversion, neck shaft angle, and MP all demonstrated a linear relationship with GMFCS, but ambulatory function itself had a clearer relationship with progressive MP [6] This updated review identified an additional five studies reporting the relationship between femoral neck anteversion, neck shaft angle and MP [19, 35]. Results were mixed across the five studies [19, 35], and, by group consensus, MP remains the radiological measure of choice.
Hermanson et al. have developed an equation for calculating a risk score for hip dislocation taking into consideration the variables of GMFCS, MP > 40 and head shaft angle [32]. These three variables were found to increase risk of developing hip dislocation within five years of the first x-ray, whereas older age decreased risk. Whilst the newly developed risk score had a good ability to differentiate between individuals at high or low risk for developing hip displacement, it requires further testing of clinical utility and reliability before recommendations for routine use in hip surveillance can be made. A shortcoming is that the risk score is only applicable when MP is greater than 40%.
Other radiological measures reported in the literature relate to acetabular morphology and hip displacement [13, 35–37]. Associations between acetabular dysplasia and GMFCS were evaluated only in GMFCS levels III-V. Acetabular dysplasia was evident only in the presence of hip displacement, and it was uncommon for acetabular dysplasia to occur independently of hip displacement [36, 37].
Frequency of radiological review can be determined by GMFCS level
New information supports a reduction in number and frequency of radiographs [8]. Two population studies support that annual surveillance using MP is effective at identifying risk and reducing incidence of hip dislocation for GMFCS levels III, IV and V in absence of PO and/or scoliosis, and children classified at GMFCS I do not develop hip dislocation and may not require radiological review [9, 10]. There is additional evidence from a large cross-sectional cohort that children at GMFCS I did not require any intervention and, in the absence of other risk factors, children at GMFCS level II could be safely monitored with a reduced frequency of radiographs [23].
Hip surveillance should commence as soon as CP is identified
Age was previously reported to be a poor determinant of hip displacement as hip dislocations have been reported in children as young as 2 years of age in population-based studies [6, 31]. More recent population studies have also reported cases of dislocation and hip displacement at first radiological review as young as two years of age [9, 32], with some as young as 8 months of age [32, 34]. This supports the recommendation that hip surveillance should commence as soon as CP is identified.
Hip surveillance should continue after skeletal maturity
Hip displacement after skeletal maturity has been reported in the presence of other risk factors such as PO and scoliosis [6, 38]. One large cohort study of adults with CP identified correlations between hip dislocations and postural asymmetry [39].
Recent findings of hip morphology and the presence of pain in adolescents and young adults who did or did not have hip surveillance identified hip dislocations in both groups [11]. Whilst the prevalence of hip dislocation was notably less in the adolescents under hip surveillance, the study highlighted that hip dislocation can progress after skeletal maturity. The new guidelines suggest adolescents at risk (hip displacement and/or PO and scoliosis) should continue hip surveillance to leave options open for reconstructive surgery and avoid the need for salvage surgery [11].
Pain is a poor indicator of risk of hip displacement
Pain is associated with hip displacement but is a poor indicator of hip displacement in younger children [3]. Whilst one cross-sectional study identified pain at low MP [40], two others identified that pain was often not present until hip displacement was above 40%, especially in older children and adolescents [40, 41]. Pain severity and frequency were higher in individuals who had no hip surveillance, with pain associated with poor hip morphology [11], higher MP, higher GMFCS level and reduced hip range of motion [40–42]. Decreased health related quality of life has been found with increased MP [4, 42]. These studies add further support for the need for hip surveillance programs to identify early progression of MP and facilitation of early referral for orthopaedic review at MP > 30% to ensure timely intervention before hip displacement increases and becomes painful.
The role of musculoskeletal measures in hip surveillance
The working group’s previous systematic review found hip range of movement to be a poor indicator of hip displacement, with only two studies addressing this factor [16, 43]. No new evidence relating hip range of movement to risk of hip displacement was found; however, reduced hip range of movement in the presence of hip displacement was associated with increased pain [40]. No evidence was found that spasticity is an indicator of hip displacement or contributes to risk of hip displacement [3].
Whilst there is no strong evidence for a relationship between musculoskeletal measures and hip MP, it is recommended that they are assessed as part of hip surveillance to ensure these factors are included in clinical decision-making and until stronger evidence emerges that they are not related. The Modified Ashworth Scale has been replaced with the more reliable Australian Spasticity Assessment Scale [44].
Hip surveillance should continue even after interventions for the hip
There is evidence that hip re-displacement and re-dislocation may occur after surgical interventions [45] and therefore, hip surveillance should continue post-operatively. Re-subluxation of the hip has been identified in a cohort of children at GMFCS IV and V within 2 to 15 years follow up [23], suggesting that routine hip surveillance should re-commence at least two years after the surgical intervention.
Discussion
This review firstly confirmed and added weight to the body of evidence that has grown over the last 20 years that hip surveillance programs are effective at identifying progressive problems at the hip. This allows for early triage to orthopaedic services. It also confirmed GMFCS as the best classification to stratify risk and MP as the best outcome measure to quantify the amount of displacement present.
One of the key objectives of the guideline is the standardisation of monitoring, ensuring a balance of risk and benefit. The aim is to minimise exposure to radiation from radiological review whilst balancing the risk of missing a progression of hip displacement or dislocation. No high-level evidence was found to replace radiological review for detecting progression of hip displacement. Both pain and range of hip motion are poor predictors. There was, however, new evidence that informed reduced frequency of radiographic review for lower risk GMFCS levels.
Most of the work published since the 2015 systematic review has focussed on GMFCS III-V, confirming that children functioning at these GMFCS levels have increased risk of hip displacement. There is new evidence to support that frequency of hip x-ray for children at GMFCS III-V could be reduced to annual review without increasing risk of undetected dislocation when there are no other musculoskeletal risk factors present, noting that effectiveness relies on the availability of a responsive, proactive orthopaedic surgical service. There is sufficient data from the population studies to support that, in the absence of other risk factors (namely gait classified as WGH IV), children in GMFCS level I are at low risk of dislocation and may not require routine radiological review.
MP remains the radiological measure of choice for hip surveillance due to its good reliability and clinical utility. Head shaft angle may be of use as a complementary measure of hip displacement when it is difficult to get good positioning for pelvic radiograph as it was found not to be dependent on femoral position on radiograph [34]. However, in these cases it is likely risk would have been identified by GMFCS and MP.
There are opportunities for future research to inform hip surveillance guidelines. More studies are required to test hip dislocation risk related to postural asymmetry, hip muscle contracture as well as the pathway for surveillance after surgical interventions. Outcome measures need to be diversified to include patient-reported outcomes related to activities, participation, and quality of life. There is a paucity of information on the impact of hip surveillance on preventing pain and improving health-related quality of life, as well as the optimal timing and frequency of surveillance for children with hemiplegic gait classification of WGH type IV. In addition, whilst there are associations identified between risk factors and the development of hip dislocation, further research is required to understand causal pathways.
Conclusion
This updated review has identified new evidence informing the updated Australian Hip Surveillance Guidelines for Children with Cerebral Palsy 2020 (available at https://www.ausacpdm.org.au/wp-content/uploads/2020/12/200240-Hip-survey-A5-booklet-WEB.pdf). The updated 2020 Guideline now includes revisions of frequency of radiographic surveillance for children at all GMFCS levels, and new information is included regarding the role of other radiological measures relating to risk and the continuation of hip surveillance after skeletal maturity and at transition to adult services.
