Hip surgery and radiology reporting for children with cerebral palsy prior to initiation of a hip surveillance program

Abstract

PURPOSE:

Hip displacement impacts quality of life for many children with cerebral palsy (CP). While early detection can help avoid dislocation and late-stage surgery, formalized surveillance programs are not ubiquitous. This study aimed to examine: 1) surgical practices around pediatric hip displacement for children with CP in a region without formalized hip surveillance; and 2) utility of MP compared to traditional radiology reporting for quantifying displacement.

METHODS:

A retrospective chart review examined hip displacement surgeries performed on children with CP between 2007–2016. Surgeries were classified as preventative, reconstructive, or salvage. Pre- and post-operative migration percentage (MP) was calculated for available radiographs using a mobile application and compared using Wilcoxon Signed Ranks test. MPs were also compared with descriptions in the corresponding radiology reports using directed and conventional content analyses.

RESULTS:

Data from 67 children (115 surgical hips) were included. Primary surgery types included preventative (63.5% hips), reconstructive (36.5%), or salvage (0%). For the 92 hips with both radiology reports and radiographs available, reports contained a range of descriptors that inconsistently reflected the retrospectively-calculated MPs.

CONCLUSION:

Current radiology reporting practices do not appear to effectively describe hip displacement for children with CP. Therefore, standardized reporting of MP is recommended.

Keywords

Cerebral palsy hip subluxation hip dislocation hip surgery surveillance program migration percentage radiology reporting

1. Introduction

Cerebral palsy (CP) is the most common motor impairment in childhood with an overall prevalence estimated at 2.11/1000 live births [1]. Hip displacement is the second most common orthopedic deformity in children with CP [2]. There is a negative correlation between the severity of displacement, measured by migration percentage (MP) and quality of life (particularly in the domains of health status, communication, and social participation) [3]. By the time hip displacement is clinically recognized, degenerative changes may already require extensive surgeries [3]. Early detection of hip subluxation may reduce invasive surgical interventions, prevent dislocation, and improve outcomes for these children [3]. In addition to potentially alleviating patient morbidity, such outcomes may also benefit healthcare systems by reducing surgical costs [4, 5, 6].

Hip surveillance is the process of identifying and monitoring progressive hip displacement through screening radiographs and clinical examinations at standardized intervals, determined by motor impairment (i.e., higher Gross Motor Function Classification System (GMFCS) [7]. levels, e.g., Levels IV and V, require more frequent surveillance). Surveillance aims to facilitate referral to an orthopedic surgeon at the first signs of hip displacement (e.g., pain, reduced range of motion, radiological evidence of displacement). Reimer’s Hip MP [8] – a radiological measurement of the femoral head’s containment within the acetabulum [9] – is a key measurement to guide clinical decision making and prognosis for hip displacement [10]. Hips with MP $>$ 30–33% have been described as being subluxated and require increased observation and orthopedic referral, as almost all hips with MP $>$ 40% progress to dislocation without intervention [11]. However, to our knowledge, no study has examined MP reporting in practice and how effectively traditional (i.e., non-standardized) radiology reporting conveys the information needed to support hip surveillance in this population.

Hip surveillance programs have been implemented in several regions worldwide and, in some cases, have virtually eliminated hip dislocation and the need for late-stage salvage surgeries [3]. Despite the success of hip surveillance programs, they have not been universally implemented; for example, no formal monitoring program existed in the Canadian province of Saskatchewan prior to 2017. The purpose of this retrospective chart review was to understand pre-surveillance surgical practices for hip displacement in the province. A secondary aim was to examine the utility of Reimer’s Hip MP in comparison to traditional radiology reporting for the purposes of hip surveillance.

2. Methods

A retrospective chart review was conducted to describe the type and number of primary and subsequent hip surgeries performed on children with CP and hip displacement in Saskatchewan from 2007 to 2016. This study was reviewed and approved on ethical grounds, with a waiver of informed consent granted by the research ethics board of the former Regina Qu’Appelle Health Region (REB 19–29).

Patients were identified via discharge abstract database coding (ICD 10 diagnostic codes G80.0, G80.1, G80.2, G80.3, G80.4, G80.8, G80.9) from the province’s two urban hospital sites that provide surgical care for children with CP and hip displacement. Inclusion criteria were: $<$ 18 years old on the date of primary surgery; diagnosis of CP; and admitted to hospital for hip displacement surgery between January 1, 2007 and December 31, 2016.

Data were collected from electronic and paper medical charts, and entered into REDCap [12] by one investigator at each site. To improve methodological consistency, the senior author completed collection for the initial five files with each of the two investigators at their respective sites. The data collection tool included: home health region, sex, age at surgery, GMFCS level, CP type (topographical and movement disorder), surgical purpose, pre- and post-operative descriptive radiology report findings, and number of subsequent surgeries (i.e., to manage ongoing subluxation and/or remove surgical hardware). Surgical purpose was recorded as either preventative (i.e., lengthening of hip adductors or iliopsoas, or obturator neurectomy), reconstructive (i.e., femoral varus derotation osteotomy or pelvic osteotomy), or salvage (i.e., arthrodesis, femoral head resection with valgus osteotomy, proximal femoral resection with soft-tissue interposition, or replacement arthroplasty) [13].

MP was retrospectively calculated for all available radiographs (pre- and/or post-operative) by one investigator using a mobile phone and HipScreen Application (www.hipscreen.org, Shriners Hospitals for Children, Sacramento, California). This application uses the camera and touchscreen of a mobile device to allow users to position the image of the hip on an MP ruler and calculate the amount of displacement. It has excellent inter-rater (ICC 0.972 to 0.989) and intra-rater reliability (ICC 0.976 to 0.989) for determining MP in children with CP [14].

2.1 Data analysis

Hip surgery and MP data were examined for normality. Parametric (normally-distributed) data are presented as frequencies, mean, and percentages. Non-parametric data are presented as median and interquartile range. Wilcoxon Signed Ranks Test examined differences between pre- and post-operative MPs. Right and left hips were separated in the analysis due to dependence of data. Significance was set at $p<$ 0.05. Analyses were conducted using SPSS statistical package version 23 (SPSS Inc, Chicago, IL).

One investigator performed a combination of conventional and directed content analysis [15] to code descriptive data for radiology report findings for those with a corresponding radiology image available. Coding was done using NVivo Version 12 (QSR International Pty Ltd), and key words were inductively grouped into categories and preliminary themes to describe the content of the reports. As well, key words and phrases describing the extent of displacement were compared with the MP for the corresponding radiograph and grouped according to whether the MP was 0–29% (low risk of dislocation) or $\geqslant$ 30%, with values of 30–99% indicating moderate/high risk of dislocation and $\geqslant$ 100% indicating frank/complete dislocation. Two investigators then reviewed the data, discussed the codes, and agreed on themes and coding structure.

Table 1
Demographics ( $n=$ 67 patients)

CP type – topographical classification
	Frequency	Percent
Urban site
Site 1	34	50.7
Site 2	33	49.3
Sex
Male	39	58.2
Female	28	41.8
GMFCS level
I	5	7.5
II	3	4.5
III	23	34.3
IV	12	17.9
V	23	34.3
Missing	1	1.5
Hemiplegia	3	4.5
Diplegia	19	28.4
Quadriplegia	45	67.2
CP type – movement disorder classification
Spastic	61	91.0
Athetoid	4	6.0
Mixed	2	3.0

3. Results

A total of 87 charts were identified. Twenty patients were excluded as they did not meet inclusion criteria: fifteen had primary hip surgeries outside the study time frame, three underwent surgeries unrelated to hip displacement, one had no documented diagnosis of CP, and one chart was improperly purged.

Data from 67 patients (115 surgical hips) were analyzed (Table 1). Mean age at the time of primary surgery was 8.67 $\pm$ 3.8 years. Higher GMFCS level was associated with an increased frequency of surgery (i.e., 52.2% of surgeries were performed on children in GMFCS levels IV and V, compared to 12.0% in levels I and II). Primary surgery was categorized as preventative for 73 (63.5%) hips and reconstructive for 42 (36.5%) hips, and no salvage surgeries were recorded (Table 2). Forty-eight (71.6%) patients received bilateral procedures while only 19 (28.4%) patients received unilateral procedures. Forty-three subsequent surgeries, of which the majority were unilateral (25, 58.1%), were conducted on 29 (43.3%) patients for a total of 61 surgical hips. No salvage procedures were performed. Thirty-eight (56.7%) children did not have any additional procedures, 18 (26.9%) received only one additional procedure, 9 (13.4%) received two additional procedures, one (1.5%) patient received three additional procedures, and one (1.5%) patient received four additional procedures.

Table 2
Primary and subsequent surgeries

Surgery time-point	Purpose	$n$	%
Primary	Preventative	73	63.5
	Soft tissue release (not specified)	9	12.3
	Lengthening hip adductors	37	50.7
	Lengthening hip adductors and iliopsoas	27	37.0
	Reconstructive	42	36.5
	Reconstruction (not specified)	7	16.7
	Femoral varus derotation osteotomy	20	47.6
	Pelvic osteotomy (reshaping, e.g., dega osteotomy)	15	35.7
	Salvage	0	0.0
	Total primary surgical hips	115	100
Subsequent	Hardware removal	33	54.1
	Preventative	9	14.8
	Reconstructive	18	29.5
	Salvage	0	0.0
	Post-operative infection debridement	1	1.6
	Total secondary surgical hips	61	100

3.1 Radiology report data and migration percentage

MP measurements for the 92 hips (41 children) with both radiographs and radiology reports available were: $\geqslant$ 100% in seven hips (6 children; 9.0%) at primary surgery (i.e., 5 unilateral and 1 bilateral) and in two hips (1 child) post-operatively, $\geqslant$ 30% for 47 hips, and 0–29% for 35 hips. MP decreased from pre- to post-operative time-points for both right ( $Z=$ 4.0, $p<$ 0.001) and left ( $Z=$ 3.61, $p<$ 0.001) hips, as illustrated by the percentage of hips with MPs $\geqslant$ 30% (Fig. 1). For the right-sided surgical hips, MP was $\geqslant$ 30% in 60% of pre-operative (median $=$ 60; IQR $=$ 30) and 19% of post-operative (median $=$ 42.5; IQR $=$ 30) hips. For the left-sided surgical hips, MP was $\geqslant$ 30% in 56% of pre-operative (median $=$ 50; IQR $=$ 30) and 24% of post-operative (median $=$ 47.5; IQR $=$ 60) hips.

Figure 1.

Percentage of hips with pre- and post-operative MP of 0–29% and $\geqslant$ 30%. (a) Right hip. (b) Left hip.

Radiologists described several aspects of the hip radiographs in their reports, including the shape and spatial orientation of the femoral head, neck, and/or acetabulum. The degree of displacement was described for 78/92 hips, and varied descriptors inconsistently reflected the MP (Table 3); these included: subluxation (34), dislocation (17), displacement (4), and migration (1). As well, radiologists often used qualifiers to describe the extent of displacement (e.g., “minimal”, “mild”, “moderate”, “severe”) or highlight a change from a previous radiograph. Some reports described the direction of displacement (primarily “superior-lateral”). MP was documented for two hips (one radiograph), and the degree of “coverage” was described as an absolute distance in four hips. Hip alignment was described as normal for 22 hips, of which 18 had an MP $=$ 0–29%.

Table 3

Most common descriptors used in radiology reports to describe hip displacement and the number and range of MPs corresponding to those terms

Descriptor	Corresponding MPs ( $n$ )	MP range	$n$
Displacement	30, 50, 90, 100	30–100%	4
Dislocation		20–120%	17
Partial	40
Complete	110
Frank	90, 120 (2)
Not otherwise specified	20, 40, 60, 70, 75, 80 (2), 90 (3), 100, 120
Subluxation		10–130%	34
Minimal	20, 40
Mild	10, 30
Slight	10
Moderate	60 (2)
Pronounced	70
Significant	70
Completely	110, 130
Progressive	50
Improved	20, 70
Similar	10, 90
Not otherwise specified	20, 30, 40 (4), 45, 50 (3), 60 (6), 75, 80
Migration	45	45%	1
Normal alignment (e.g., well-aligned, in good position, appropriately covered, well- positioned, no subluxation, no dislocation)	0 (5), 10 (7), 15 (2), 20 (4), 30 (2), 40, 60	0–60%	22
Total descriptors coded		0–130%	78
No descriptor used			14
Total $n$ with radiology report and radiograph available			92

When the radiology report descriptors were compared with the MPs and grouped according to whether the MP was 0–29%, 30–99%, or $\geqslant$ 100%, three themes emerged:

1) MP was high enough to indicate clinical concern (i.e., $\geqslant$ 30%), but descriptors understated or did not objectively quantify the extent of displacement. For example, for two hips (MP $=$ 60% and 30%), radiology reports indicated “femoral heads appear to articulate with the acetabula bilaterally.” For another hip with MP $=$ 30%, the report stated, “right acetabulum is relatively well developed and there is no subluxation at the right hip.” A hip with MP $=$ 75% showed “evidence of superior subluxation.” Another child’s MPs were 20% and 120%, and the report read “bilateral hip dislocation, greater on the left.” A hip with MP $=$ 70% was described as having “improved coverage.” Of the hips with MP $=$ 40%, descriptions included “Hip is in” and “Minimal subluxation,” and another received no comment. Four hips with $\geqslant$ 30% were described as normal. Although hip surveillance guidelines advise referral to an orthopedic surgeon for MPs of 30%, descriptions of hips with MP $=$ 30% included “mild subluxation” and “no subluxation.”

2) MPs that were not high enough to cause clinical concern ( $<$ 30%) were described inconsistently in radiology reports. Hips with MPs of 10–29% were sometimes described as being normal and sometimes described as subluxated. For example, descriptions for hips with MP $=$ 10% included: “subluxation no longer present,” “in good position,” “slight lateral subluxation,” and “no subluxation.” Descriptions for hips with MP $=$ 20% included: “undercoverage,” “adequate coverage,” “in good position,” “minimal subluxation,” and “superolaterally positioned femoral head relative to acetabulum.”

3) MP and report descriptors generally agreed. Agreement could only be confirmed quantitatively where the radiologist estimated or calculated the MP, or in extreme cases where retrospectively-calculated MP was a) 30–99% and the radiologist reported the hip was normal (18/92 radiographs), or b) $\geqslant$ 100% and reported as completely dislocated (5/92 radiographs). Five of seven dislocated hips (i.e., MP $\geqslant$ 100%) were identified as such (i.e., “complete” or “frank” dislocation, or “completely subluxed”). The other two hips with MP $\geqslant$ 100% were described as “dislocated”; however, 15 other hips with MPs ranging from 20–90% were also described as “dislocated,” suggesting non-specific use of the term “dislocation” that does not facilitate identification of hips that are truly dislocated.

4. Discussion

This study explored hip subluxation surgery in children with CP over a ten-year period in a Canadian province without a formalized hip surveillance program. During this time, 67 patients underwent primary hip surgery, consisting of mostly bilateral, preventative procedures. A majority of these procedures were performed on children with high GMFCS levels. Although no salvage surgeries were performed, seven hips (in 6 children) were dislocated at the time of primary surgery and two remained dislocated post-operatively. Radiology reports demonstrated that MP was reported for only one radiograph, and a range of descriptors was used to subjectively describe hip displacement, without meaningful relationship to the MP.

Several studies have documented surgical practices pre-and post-hip surveillance implementation. In Australia and Sweden, salvage procedures were discontinued shortly after implementation of surveillance practices [3, 4, 7, 8, 9, 10, 11] due to early identification and reduced incidence of dislocation. A more recent exploration of surgical care for children with CP GMFCS levels III-V, compared Sweden (a center with a surveillance program) and Norway (a center without surveillance) [16]; while the number of surgeries was similar between sites, the age at primary surgery was younger (5.7 vs. 7.6 years) and fewer number of dislocations occurred (1 vs 18) in Sweden. These findings imply that implementation of hip surveillance programs may reduce the incidence of hip dislocation and therefore the numbers of subsequent salvage surgeries.

Although a reduction in salvage surgeries has been described as a potential outcome of hip surveillance, none of the patients underwent salvage surgeries during our study period. It is possible that individual surgeon philosophy of care may have affected these outcomes. While no salvage surgeries were performed, six children (9.0%) had dislocation within our surgical based cohort, which is much higher than the 2.5% of children reported in a population-based study from Scotland (a country with a hip surveillance program) [17] and similar to the 9% dislocation rate observed in Sweden prior to implementation of hip surveillance [4]. These discrepancies in results may be due to sample size differences in studies. Although salvage surgeries can reduce pain, they are associated with complications [10]; therefore, it is also possible that patients/families of children with dislocation declined the procedure, the child’s overall health status precluded surgery, or surgeons recommended alternate approaches. Furthermore, children in this sample were older at primary surgery than Elkamil and colleagues [16] reported in Sweden and Norway. Both age at primary surgery and incidence of dislocation may be important variables to monitor as surveillance programs are implemented.

Hip surveillance requires a multidisciplinary approach. Radiologists play a key role in successfully measuring and tracking hip displacement to inform clinical decision making. The term displacement has been used to refer to hips with MP $>$ 30–33% and dislocation to MP $>$ 100% [18]; however, we found that radiologists often used these and other terms interchangeably in their reports, and terminology did not reliably correspond to MP. For example, “displacement” referred to hips with MP $=$ 30–100%, “dislocation” for MP $=$ 20–120%, and “subluxation” for MP $=$ 10–130%. The results of this study support implementation of MP as a key component of standardized radiographic reporting to facilitate hip surveillance program efficacy. This may be achieved successfully through implementation of structured department-wide reporting systems amongst radiologists [19].

4.1 Limitations

Due to the province’s small population, our sample size was limited and these results should be compared to data from larger populations with caution. Given that this study examined a surgically-treated cohort, the true incidence of dislocation in children with CP in the province’s population was likely much higher than the data indicated. As well, patients lost to follow-up or accessing surgical interventions outside the study area were not reflected. Additionally, due to the long study period and retrospective design, changes in health care processes and technology may have impacted the data available. For example, only 92 hip radiology reports had radiographs available for MP measurement. Lastly, standardized radiographic patient positioning was likely not used, although it is recommended for accurate MP measurement. While positioning has not been shown to significantly impact the validity of results [20], the extent to which this may have affected calculation of MP in this study is unknown.

Future research is planned to evaluate the impact of implementing a CP hip surveillance program in this region. Additional studies may help to understand the role of surgeon/family preference in surgical decision making or radiologists’ perspectives in incorporating MP as part of a hip surveillance program.

5. Conclusion

This study described a variety of preventative and reconstructive procedures performed for hip displacement in a center without a formal hip surveillance program for children with CP. A significant number of hips had progressed to dislocation at the time of primary surgery, suggesting the potential value of implementing a hip surveillance program in this region. Radiology reporting practices did not consistently and effectively communicate the degree of hip displacement; therefore, routine reporting of MP is recommended to help inform care for children undergoing hip surveillance. This information may improve standardization of hip surveillance and ultimately outcomes for children with CP.

Footnotes

Acknowledgments

The authors would like to thank Dr. Matthew Pechey for his help with data collection along with the Health Information Management Services of the Former Regina Qu’Appelle Health Region for their support.

Conflict of interest

The authors have no conflicts of interest to disclose.

References

Oskoui

, Coutinho

, Dykeman

, Jett

, Pringsheim

. An update on the prevalence of cerebral palsy: A systematic review and meta-analysis. Dev Med Child Neurol. 2013; 55(6): 509-519. doi: 10.1111/dmcn.12080.

Cornell

. The hip in cerebral palsy. Dev Med Child Neurol. 1995; 37(1): 3-18. doi: 10.1111/j.1469-8749.1995.tb11928.x.

Dobson

, Boyd

, Parrott

, Nattrass

, Graham

. Hip surveillance in children with cerebral palsy. Impact on the surgical management of spastic hip disease. J Bone Joint Surg Br. 2002; 84(5): 720-726. doi: 10.1302/0301-620x.84b5.12398.

Hgglund

, Alriksson-Schmidt

, Lauge-Pedersen

, Rodby-Bousquet

, Wagner

, Westbom

. Prevention of dislocation of the hip in children with cerebral palsy: 20-year results of a population-based prevention programme. Bone Joint J. 2014; 96-B(11): 1546-1552. doi: 10.1302/0301-620X.96B11.34385.

Kentish

, Wynter

, Snape

, Boyd

. Five-year outcome of state-wide hip surveillance of children and adolescents with cerebral palsy. J Pediatr Rehabil Med. 2011; 4(3): 205-217. doi: 10.3233/PRM-2011-0176.

Wynter

, Gibson

, Willoughby

, Love

, Kentish

, Thomason

, et al. Australian hip surveillance guidelines for children with cerebral palsy: 5-year review. Dev Med Child Neurol. 2015; 57(9): 808-820. doi: 10.1111/dmcn.12754.

Palisano

, Rosenbaum

, Walter

, Russell

, Wood

, Galuppi

. Development and reliability of a system to classify gross motor function in children with cerebral palsy. Dev Med Child Neurol. 1997; 39(4): 214-223. doi: 10.1111/j.1469-8749.1997.tb07414.x.

Reimers

. The stability of the hip in children. A radiological study of the results of muscle surgery in cerebral palsy. Acta Orthop Scand Suppl. 1980; 184: 1-100. doi: 10.3109/ort.1980.51.suppl-184.01.

Scrutton

, Baird

, Smeeton

. Hip dysplasia in bilateral cerebral palsy: Incidence and natural history in children aged 18 months to 5 years. Dev Med Child Neurol. 2001; 43(9): 586-600. doi: 10.1017/s0012162201001086.

10.

Shore

, Spence

, Graham

. The role for hip surveillance in children with cerebral palsy. Curr Rev Musculoskelet Med. 2012; 5(2): 126-134. doi: 10.1007/s12178-012-9120.

11.

Hgglund

, Lauge-Pedersen

, Persson

. Radiographic threshold values for hip screening in cerebral palsy. J Child Orthop. 2007; 1(1): 43-47. doi: 10.1007/s11832-007-0012-x.

12.

Harris

, Taylor

, Thielke

, Payne

, Gonzalez

, Conde

. Research electronic data capture (REDCap) – a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009; 42(2): 377-381. doi: 10.1016/j.jbi.2008.08.010.

13.

Miller

, Mayson

, Mulpuri

, O’ Donnell

. Developing a province-wide hip surveillance program for children with cerebral palsy: From evidence to consensus to program implementation: A mini-review. J Pediatr Orthop B. 2020; 29(6): 517-522. doi: 10.1097/BPB.0000000000000707.

14.

Kulkarni

, Davids

, Boyles

, Cung

, Bagley

. Reliability and efficiency of three methods of calculating migration percentage on radiographs for hip surveillance in children with cerebral palsy. J Child Orthop. 2018; 12(2): 145-151. doi: 10.1302/1863-2548.12.170189.

15.

Mc Hugh

. Interrater reliability: The kappa statistic. Biochem Med (Zagreb). 2012; 22(3): 276-282.

16.

Elkamil

, Andersen

, Hgglund

, Lamvik

, Skranes

, Vik

. Prevalence of hip dislocation among children with cerebral palsy in regions with and without a surveillance programme: A cross sectional study in Sweden and Norway. BMC Musculoskelet Disord. 2011; 12: 284. doi: 10.1186/1471-2474-12-284.

17.

Bugler

, Gaston

, Robb

. Hip displacement in children with cerebral palsy in Scotland: A total population study. J Child Orthop. 2018; 12(6): 635-639. doi: 10.1302/1863-2548.12.180106.

18.

Robb

, Hgglund

. Hip surveillance and management of the displaced hip in cerebral palsy. J Child Orthop. 2013; 7(5): 407-413. doi: 10.1007/s11832-013-0515-6.

19.

Larson

, Towbin

, Pryor

, Donnelly

. Improving consistency in radiology reporting through the use of department-wide standardized structured reporting. Radiology. 2013; 267(1): 240-250. doi: 10.1148/radiol.12121502.

20.

Kinch

, Campbell

, Maclean

JGB

, Read

, Barker