Hip surveillance for patients with cerebral palsy in the United States

1 History of hip care in children with cerebral palsy

Hip subluxation and dislocation in children with cerebral palsy (CP) has been recognized to occur predominantly (89%) in the non-ambulatory population, although it can occur in ambulatory patients [1]. Hips are typically normal at birth and tend to sublux over time [2]. Hip subluxation is thought to be due to abnormal muscle forces (including muscle weakness and spasticity), altered bony anatomy, and lack of weightbearing [2, 3].

There are many ways to measure hip dysplasia and subluxation on radiographs. Reimers’ migration percentage, which measures the percent of the proximal femoral epiphysis outside the lateral edge of the acetabulum (or Perkin’s line), has been shown to be the most reproducible and consistent measure of hip subluxation in children with CP [4, 5]. Radiographs should be obtained with standardized positioning of children with CP for a supine anteroposterior (AP) pelvis radiograph with the patellae upward, a bump under the legs (especially if there is any hip flexion contracture and/or anterior pelvic tilt), and the legs in neutral abduction/adduction [6]. Children with a migration percentage above 30% are considered to have a “hip at risk” [7, 8].

Treatment for hip subluxation depends on the child’s age and severity of subluxation. For younger children with mild to moderate hip subluxation (<50%), preventative (soft tissue) surgery can improve hip coverage and development [9–11]. In older children or children with more severe subluxation, reconstructive surgery with proximal femoral varus derotational osteotomies are required, and acetabular osteotomies are often added to address residual acetabular dysplasia [9, 10]. However, in children and adults with long-standing subluxation/dislocation with deformation of the femoral head and acetabulum, salvage surgeries may be recommended, such as proximal femoral resections or valgus osteotomies [12–14]. These salvage options may improve range of motion and pain; however, many patients continue to have pain after salvage [12, 15].

Controversy exists in the literature on whether or not children who are affected by severe CP with significant hip pathology have pain [15–18]. These reports have several methodological problems: the majority of patients had difficulty in communication due to their neuromuscular condition, many of the studies were done at long-term care facilities with inconsistent care givers, and there is a lack of discussion about what level of pain is considered to interfere with function [15, 16]. However, more recent studies demonstrate improvements in Health Related Quality of Life (HRQoL), hip morphology, pain and salvage surgeries when a proactive approach is undertaken to manage hip pathology [18–20].

Current thought is that children with CP have less pain, improved HRQoL and improved future hip morphology when their hips are treated with appropriate surgical techniques for the level of hip subluxation, prior to the development of hip dislocation and proximal femoral deformity.

2 Hip surveillance guidelines

True hip surveillance requires a population-based approach to healthcare in which all individuals with a condition or at risk for that condition are identified and then followed over time. In the case of hip surveillance and CP, all patients with CP or at risk for CP would need to be identified in order to apply hip surveillance strategies [21]. That type of population-based healthcare is simply not a reality in the United States today. However, several models exist internationally for how hip surveillance can be applied to a population and how it can make a difference in overall outcomes of hip pathology in patients with CP.

Each of these international models varies slightly in terms of the specific radiographic schedule to obtain radiographs. However, they are all based on the knowledge of the variability of hip pathology with Gross Motor Functional Classification System (GMFCS) levels and take into account previous measurements of migration percentage as well as age to determine the specific hip surveillance schedule [21, 22].

2.1 Australian guidelines

The Australian standards of care for hip surveillance are probably the most recognized international model today. This model was developed by a national working group between 2006 and 2008, implemented originally in 2008, and revised in 2014 to its current form [23]. The majority of hip surveillance in Australia occurs at localized, community physical therapy units that perform comprehensive physical examinations along with radiographic surveillance. Surveillance begins between 12 and 24 months of age for all children identified with CP or for those at risk for developing CP regardless of the projected GMFCS level. After the initial assessment, the surveillance schedule varies according to the GMFCS level, and all patients are referred for orthopedic evaluation in the presence of hip pain, unstable (increasing) migration percentage, or migration percentage above 30% [24].

GMFCS I: All patients receive an initial screening radiograph upon entry into the program. These patients receive repeat clinical evaluations, including screening for obvious limb length inequalities (such as an abnormal Allis and/or Galeazzi sign) without radiographic exam at age 3 and age 5. If the GMFCS level of I is confirmed and the patient has normal hips, then they are discharged from hip surveillance unless the patient has a hemiplegic Winters and Gage (W&G) type IV (patients with hip internal rotation, knee flexion, and ankle flexion) CP. In that case, because those patients have an increased risk for late dislocation, they are continued to be followed until the W&G classification is confirmed between age of five and 10. If it is confirmed to be a type IV, then repeating clinical radiographs are obtained at age 10 and very every year after until skeletal maturity.

GMFCS II: After the initial radiograph, repeat clinical and radiographic examination occurs 12 months after the initiation of the surveillance. If the migration percentage at that time is stable, they are then repeated again at age 4 to 5 years and then at 8 to 10 years of age. If at any time the hip migration percentage is unstable, then those radiographs are repeated yearly until the hip has stabilized or is treated. If the migration percentage is stable at 10 years of age, those patients are then discharged from surveillance. If the migration percentage is unstable or more than 30%, those patients continue to have radiographic exams until skeletal maturity.

GMFCS III, IV, and V: Those patients will receive radiographic evaluation every 6 months until a stable migration percentage is obtained and then every 12 months thereafter. At age 7 years, if the migration percentage is stable then the radiographic exam is discontinued until prepuberty; radiographs will then recommence every 12 months instead until skeletal maturity. At any time if the radiograph shows a migration percentage that is greater than 30% or unstable, then radiographs are obtained every 6 months. For patients at the GMFCS V level, radiographs are obtained yearly regardless of migration percentage. For all of these patients, surveillance continues yearly after skeletal maturity in the presence of progressive scoliosis, pelvic obliquity, or an abnormal migration percentage.

2.2 Scandinavian model –CPUP

The oldest hip surveillance program was established in southern Sweden in 1994 and now crosses over 4 countries in Scandinavia (Sweden, Norway, Denmark, and Iceland) through the CPUP program [25]. Like the Australian program, the heart of hip surveillance occurs at localized therapy centers in the communities. Physical exams of all children with a diagnosis of CP and at risk for CP are performed every 6 months until the age of 6 years and then annual exams thereafter. Radiographic evaluation in the form of AP pelvis accompanies this physical exam, and the radiographic schedule is determined by GMFCS level, similar (but slightly different) to that of Australia. Once an abnormal radiograph is obtained (either migration percentage more than 30% or more or unstable), referral to the orthopedic surgeon is initiated. If the migration percentage is less than 33% at skeletal maturity (triradiate cartilage closure) then screening is discontinued.

2.3 British Columbia model

The only formalized hip surveillance program that is a population-based program in North America was established in British Columbia in 2012 [26]. This model came about when the local clinicians were concerned about their relatively poor access to orthopedic providers and delays in transfers and referrals for children who were at risk for severe hip dysplasia. In this model, clinical exams begin at diagnosis for all children identified with CP or those at risk, and then follow-up evaluations are scheduled according GMFCS level. In the British Columbia system, referral to an orthopedic consultant for preventive surgical consideration occurs when the migration percentage is greater than 30%, hip abduction is less than 30 degrees, or there is any reported worsening in function and/or increase in hip pain [26].

2.4 American Academy of Cerebral Palsy and Developmental Medicine (AACPDM) hip surveillance care pathways

In an effort to try to provide a clinical practice guideline for its North American members, in 2014 AACPDM employed an evidence-based consensus group to determine the organization’s recommendations for hip surveillance. A significant literature review was undertaken with a focus mostly on the Scandinavian and Australian guidelines. A care pathway was published on the organization’s website and is widely used by AACPDM members today. This care pathway recommends both clinical exam schedules based on age and GMFCS levels as well as radiographic examinations which match closely to the Scandinavian guidelines [8]. In addition, the care pathway reviews the state of the evidence regarding hip surveillance and provides practical guidelines of how to obtain appropriate, consistent, standardized AP pelvic radiographs, even in the setting of persistent common lower extremity deformities, such as excessive femoral anteversion and hip flexion contracture. This too is very similar to the work done in Australia and Scandinavia.

2.5 Pediatric Orthopedic Society of North America (POSNA) hip screening

Much progress has been made in the last 10 years in North America regarding the dissemination of the importance of hip surveillance across subspecialties. Efforts at improving hip surveillance in orthopedic, physiatry, neurology, and early intervention clinics are being expanded throughout the network, including an effort for radiologists in North America to provide migration percentage measurements as a standard radiographic report. However, the majority of the current state of hip surveillance in the United States is primarily through orthopedic offices. Pediatric orthopedic surgeons have reported a knowledge of the importance of hip surveillance but also have demonstrated a lack of knowledge of the specific guidelines discussed here [27]. This stems from the fact that there are still a large percentage of pediatric orthopedic surgeons in United States who care for some children with CP but not in a dedicated CP center. As such, many are members of POSNA but are not members of AACPDM. This has led a group of CP orthopedic specialists inside of POSNA to try to bridge the knowledge gap of their members by coming up with simplified guidelines for monitoring hip subluxation in children with CP. That effort was started in 2016 and was published in 2020 in an effort to provide easy to use, but standardized, radiographic surveillance schedules for busy pediatric orthopedic surgical practices [28].

3 Sequela of guidelines

The current expert opinion by most healthcare providers that specialize in CP is that reduced hips are likely to provide improved quality of life when compared to dislocated hips. The classic literature is unclear on whether dislocated hips are painful or not. Previous reports in the literature report 25% –50% of patients with dislocated hips have hip pain; however, many of those reports were primarily in institutionalized adult patients. Modern literature reports a significant percentage of adults with CP complaining of pain and that a high percentage of those pain complaints are localized to the hip. Newer disease-specific outcome instruments such as the Caregiver Priorities and Child Health Index of Life with Disabilities have been shown to be responsive to detecting a decrease in HRQoL with progressive hip dysplasia and subluxation [19]. Furthermore, those studies have similarly shown an improvement in quality of life after hip reconstruction [20]. This evidence provides indirect support of the importance of hip surveillance and early detection of hip pathology in patients with CP.

There is some quantitative, direct evidence to suggest that hip surveillance does make a difference in the care of children with CP. In 2005, Hagglund et al. showed their first results of 10 years of a hip surveillance program in Sweden [29], which demonstrated that they were able to quantify the number of children throughout Sweden with dislocated hips and showed that, through the hip surveillance program, that number decreased. This finding is important for several reasons: first, it demonstrated the efficacy of the population-based program to actually measure the number of dislocated hips that presumably would go on to be painful for those patients; second, the authors proved that hip surveillance and responsive orthopedic care can decrease the number of dislocated hips; and finally, the ultimate end result was proven: a decrease in hip dislocation and a decreased need for salvage surgery. Along those lines, Dobson et al. reported on the first 3 years of the orthopedic clinic based on the hip surveillance program in Australian [30]. They similarly showed complete elimination of hip dislocation and salvage surgery, although with an increase in preventive surgery.

Today, as part of the Cerebral Palsy Research Network (CPRN), significant efforts are being made to ensure that children with CP are being appropriately monitored for hip dysplasia. Goals for the next steps are to subsequently measure the effects of quality improvement interventions to increase the percentage of hips that are being surveilled in United States, and then ultimately to measure the primary outcome which is to decrease the number of dislocated hips and the need for hip salvage surgeries.

3.1 Current state of hip surveillance

The CPRN is a multi-center collaborative effort to improve the care of children with CP through evidence-based initiatives. Although guidelines for hip surveillance exist, there are many barriers to implementation in the United States. Not all centers in the United States have a centralized location for children with CP to be cared for and evaluated. Often children are seen in multiple locations and clinics, including primary care, physical medicine and rehabilitation, pulmonology, neurology, and orthopedic surgery. This can lead to fragmentation of care and absence of hip surveillance as a part of standard well-being/screening protocols.

As mentioned previously, one of the initiatives of CPRN is implementing hip surveillance guidelines. To establish a baseline of hip surveillance, CPRN surveyed the membership of their network with 21/36 centers (58%) responding. Four centers had multiple people respond, so these responses were used for intra-center consistency.

Within the CPRN, most children with CP enter the healthcare environment through physical medicine and rehabilitation (66%). The majority of hip surveillance was performed by orthopedic surgeons (56%) and physical medicine and rehabilitation specialists (44%). Most centers (71%) are using the guidelines from AACPDM; however, 10% were not following any guidelines to direct hip surveillance⁸. Most distributed the guidelines via word of mouth (66%) with no formal education plan (62%).

The hip exam, radiograph orders and measurements were completed by orthopedic surgeons (100%) and physical medicine and rehabilitation specialists (86/81% /52%). The consistency of measuring and recording of hip migration percentage by radiologists was poor. Less than half of the centers referred children for hip subluxation greater than 30%, which is recommended [8]. Most centers (57%) did not have quality measures in place for positioning the child in radiology.

For those centers with more than one response (4/25, 16%), there was disagreement on what or if guidelines were used to direct hip surveillance. The standard method of hip migration measurement and when to refer to orthopedic surgery for hip subluxation was unknown.

4 Conclusions

Hip surveillance for CP in the United States is highly variable, even among pediatric centers that have CP programs. Through clear guidelines and active surveillance, hip subluxation can be treated appropriately, resulting in improved HRQoL, hip mobility and congruency and decreased painful hip dislocations and subluxation. Quality initiatives, such as those championed by the CPRN, can improve the effectiveness of radiographic screening and ultimately improve the outcomes of children with CP hip pathology.