430
Original article
Radiological outcome of reconstructive hip surgery in children with gross motor function classification system IV and V cerebral palsy Shiran Zhanga, Nichola C. Wilsona,b, Anna H. Mackeyb and Ngaire Susan Stotta,b Hip subluxation is common in children with cerebral palsy (CP). The aim of this study was to describe the radiological outcome of reconstructive hip surgery in children with CP, gross motor function classification system (GMFCS) level IV and V, and determine whether the GMFCS level plays a predictive role in outcome. This was a retrospective cohort study conducted at a tertiary-level pediatric hospital with a CP hip surveillance program. Of 110 children with GMFCS IV and V CP registered for hip surveillance, 45 underwent reconstructive hip surgery between 1997 and 2009, defined as varus derotational proximal femoral osteotomy with or without additional pelvic osteotomy. Eleven children were excluded because of lack of 12-month follow-up (n = 10) or missing clinical records (n = 1). Thus, 21 GMFCS IV children (median age 6 years at surgery) and 13 GMFCS V children (median age 5 years at surgery), who underwent 58 index surgeries, were included in the study. Clinical records and radiology were reviewed. The two surgical groups were femoral osteotomy (varus derotational femoral osteotomy with an AO blade plate or femoral locking plate fixation), or femoral ostetotomy with additional pelvic osteotomy. Reimer’s migration percentage (MP) was calculated from anteroposterior pelvis radiographs to determine the outcome for each hip independently. Failure was defined as
Introduction Cerebral palsy (CP) is the most common cause of disability in children, affecting 2–2.5 in 1000 live births per year in developed countries [1,2]. Hip subluxation and dislocation is highly prevalent in children with CP and causes significant morbidity, including pain, difficulties with positioning and perineal hygiene, pelvic obliquity, and possibly increased risk of scoliosis [3–6]. Hip subluxation results from imbalanced muscular forces acting on the hip, coupled with failure of normal bone development because of lack of weight bearing. The incidence of subluxation increases with the severity of CP and occurs in parallel with reduced ambulatory ability [5,7]. The gross motor function classification system (GMFCS) is a reliable and stable five-level scale that ranks ambulatory ability and musculoskeletal function in children with CP [8]. Children with CP functioning at GMFCS IV and V levels are considered nonambulatory, although some children who are GMFCS level IV can take some steps with support from another person. Children who function at GMFCS level V are wheelchair dependent and are medically frail, with a Swedish study reporting 1060-152X © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
MP of greater than 60% or further operation on the hip. Reconstructive surgeries were performed for 58 hips with a median preoperative MP of 55%. There were 15 failures at a median of 62 months, including nine failures in 35 GMFCS IV hips and six failures in 23 GMFCS V hips. Overall, GMFCS V hips tended to fail earlier, (hazard ratio 2.3) with a median time to failure of 78 and 39 months for GMFCS IV and V hips, respectively. Combined femoral and pelvic osteotomies had the lowest failure rates in both groups of patients. The GMFCS classification may have some predictive value for outcomes following reconstructive hip surgery, with surgery for GMFCS V hips tending to fail earlier. J Pediatr Orthop B 23:430–434 © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Journal of Pediatric Orthopaedics B 2014, 23:430–434 Keywords: cerebral palsy, gross motor function classification system, hip surgery, radiological outcome a Department of Surgery, Faculty of Medical and Health Sciences, The University of Auckland and bStarship Children’s Hospital, Auckland, New Zealand
Correspondence to Ngaire Susan Stott, MBChB, FRACS, PhD, Faculty of Medical and Health Sciences, The University of Auckland, Private Bag 92019, Auckland Mail Centre, Auckland, 1142 New Zealand Tel: + 64 9 3737599 x82861; fax: + 64 9 3677159; e-mail: [email protected]
only a 60% survival rate into adulthood [9,10]. Data from the Victorian Cerebral Palsy Register show that 69 and 90% of children with CP between the ages of 6 and 14 years, who function at GMFCS IV or V levels, respectively, develop hip subluxation or dislocation [7]. Management of hip subluxation is multidisciplinary and includes hip surveillance, nonoperative, and operative interventions. Operative interventions for identified hip subluxation may be soft tissue release (adductor muscle release) or bony procedures, usually femoral or pelvic osteotomies. Reconstructive hip surgery refers to femoral and pelvic osteotomies that reduce the subluxed or dislocated hip by correcting osseous deformities, with additional open reductions when the hip cannot be reduced closed. Salvage procedures such as femoral head resection are reserved for longstanding painful dislocations in older children [11]. A recent study describing outcomes following soft tissue surgery showed that the GMFCS level is associated with outcome, with higher rates of further surgery in GMFCS level IV or V children [12]. The predictive value of the DOI: 10.1097/BPB.0000000000000075
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Radiological outcomes GMFCS IV and V Zhang et al. 431
GMFCS following reconstructive hip surgery is yet to be studied. This study aims to describe the radiological outcome of reconstructive hip surgery in children with nonambulatory CP and determine whether the GMFCS level plays a predictive role in outcome in this group of children.
Methods Study design
This is a retrospective cohort study of children with CP, GMFCS IV and V, registered on a tertiary referral service hip surveillance program, who had undergone reconstructive hip surgery between 1997 and 2009. Children on the surveillance program who were GMFCS III or less were not part of this study as only three of 38 children had needed reconstructive hip surgery. Children who had not had at least 12 months of postoperative follow-up were excluded. The study received ethics approval from the regional and institutional review boards. Data collection
Demographic and clinical data were retrieved from the hospital’s Clinical Record Information System. This included the GMFCS level as determined by the examining physician or physiotherapist, date of birth, sex, and the operation details. For the purpose of this study, the surgeries were classified as either femoral osteotomy (varus derotational proximal femoral osteotomy with an AO blade plate or femoral locking plate fixation) or combined femoral and pelvic osteotomy, with or without additional soft tissue release and open reduction. Soft tissue release for femoral osteotomy was sequential release of adductor longus, gracilis, and up to 50% of adductor brevis to achieve 50° of abduction in extension and thus facilitate the varus correction. If the range of hip abduction was satisfactory under anesthesia, no adductor release was performed. In three hips that were subjected to additional pelvic osteotomy, a psoas lengthening at the pelvis brim as well as the sequential adductor lengthening as above were performed. No obturator neurectomies or hamstring releases were performed in this group of children. Radiographic measurements
Pelvic radiographs were retrieved from the hospital’s Picture Archiving and Communication System, which archives all radiographs taken from 2002 onwards. Radiographs taken before 2002 were in hardcopy and digitized onto Picture Archiving and Communication System before analysis. Of the radiological measures, the migration percentage (MP) is the only measurement that correlates consistently with hip displacement [13]. In this study, the outcome of the index surgery for each hip was considered independently. The index surgery was defined as the first procedure performed on that hip irrespective of whether the
contralateral hip had been operated on. A surgery was deemed failed if MP reached 60% or if further surgery was performed on that hip. MP was measured on standard anteroposterior pelvic radiographs as described by Reimers [14]. Measurements were taken on digitized radiographs by the same observer to avoid interobserver error. Radiographs of inadequate quality were excluded. The duration of follow-up was defined as from the time between the index procedure to the time of the last follow-up radiograph. The preoperative and all subsequent postoperative radiographs were reviewed. Statistical analysis
Hips were grouped by demographic and clinical characteristics. Cox regression was used to generate hazard ratios (HR) comparing outcomes in different groups. An HR ratio greater than 1 indicates a higher failure rate in that group and HR less than 1 indicates a lower failure rate. Statistical significance was defined as P value less than 0.05. All statistical analyses were carried out on StatsDirect 2.7.8 (StatsDirect Ltd, Cheshire, UK).
Results The hip surveillance program registered 110 GMFCS IV and V children, 45 of whom had undergone reconstructive hip surgery. Ten children were excluded as the duration of radiological follow-up was less than 12 months. One further child was excluded because of missing documentation of her surgery (performed overseas). The preoperative radiographs for five hips could not be retrieved, but the MP measures recorded in the radiograph reports were available and thus used as the preoperative baseline. This led to a final sample of 34 children, with a median age at surgery of 5 years (range 1–13), as described in Table 1. All children had bilateral CP and the majority had a diagnosis of spastic quadriplegia. Twenty-four children had both hips operated on; thus, 58 index surgeries were included in the analysis (Tables 2 and 3). All except two hips were operated on at our own institution. Reconstructive hip surgery was performed to correct existing hip migration of at least one hip, with additional pelvic osteotomy performed on the most severely subluxated hips. The decision to proceed with pelvic osteotomy was usually made intraoperatively when femoral osteotomy alone failed to provide adequate coverage and stability. Forty-three of the 58 hips had an adductor release to facilitate varus correction. The Table 1
Participant demographics
GMFCS IV GMFCS V Total
Number of children
Sex (boys : girls)
Median age at first surgery (years)
21 13 34
16 : 5 5:8 21 : 13
6 (1–13) 5 (2–12) 5 (1–13)
GMFCS, gross motor function classification system.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
432 Journal of Pediatric Orthopaedics B 2014, Vol 23 No 5
Table 2
Summary of index procedures by GMFCS level Number of hips
Median age in years at time of surgery (range)
Median preoperative MP (%)
Median postoperative MP (%)
Number of failures
Median months to failure (range)
Median months duration of follow-up (range)
35 23 58
6 (1–13) 5 (2–12) 5 (1–13)
48 (8–100) 66 (12–100) 55 (12–100)
0 (0–34) 0 (0–57) 0 (0–57)
9 6 15
78 (36–105) 39 (1–82) 62 (1–105)
65 (14–157) 50 (14–159) 62.5 (14–159)
GMFCS IV GMFCS V Total
GMFCS, gross motor function classification system; MP, migration percentage.
Table 3
Summary of index procedures by type of procedure Number of hips
Median age in years at time of surgery (range)
Femoral osteotomy (n = 39) GMFCS IV 23 5 GMFCS V 16 3.5 Subtotal 39 5 Femoral and pelvic osteotomy (n = 19) GMFCS IV 12 10 GMFCS V 7 5 Subtotal 19 9
Median preoperative MP (range) (%)
Median postoperative MP (range) (%)
Number of failures
Median months to failure (range)
Median months duration of follow-up (range)
(1–11) (2–9) (1–13)
47 (8–95) 52 (12–100) 48 (8–100)
0 (0–34) 0 (0–57) 0 (0–57)
8 5 13
58 (36–105) 62 (13–82) 62 (13–105)
87 (19–140) 54.5 (14–159) 70 (14–159)
(1–13) (2–12) (1–13)
55 (38–100) 72 (50–100) 69 (38–100)*
0 (0–18) 0 (0–0) 0 (0–18)
1 1a 2
Not applicable Not applicable 45 (1–89)
58.5 (14–157) 39 (15–61) 55 (14–157)
GMFCS, gross motor function classification system; MP, migration percentage. Failure because of technical error at 1 month. *Significantly higher than hips subjected to isolated femoral osteotomy (P = 0.02). a
median duration of follow-up across the group was 62.5 months (range 14–159 months). The median preoperative MP was 55% (12–100%), with no difference between GMFCS IV hips and GMFCS V hips (48 and 66%, respectively, P = 0.15). There was a trend for children who were GMFCS level V to be younger at the time of the index surgery (P = 0.06) Failure rates
Fifteen of the 58 reconstructive hip surgeries (26%) failed over the period of follow-up; nine of the failures were GMFCS IV hips and six were GMFCS V hips. Six hips were subjected to further surgery because of resubluxation with MP equal to or greater than 60%; a further five hips resubluxated to an MP of 60%, but did not undergo any further surgery during the time frame of the study. Of the remaining hips that resubluxated, three subsequently underwent salvage surgery, including femoral head resection, to correct soft tissue contractures but not to reduce the dislocated hip. For the final hip, the index procedure failed at 1 month and further surgery was performed. The median times to failure for GMFCS IV and V hips were 78 and 39 months, respectively, as shown in Table 2. This difference was not statistically significant (P = 0.17) because of the small sample size. The HR for the difference in time to failure between the groups was 2.3, indicating a shorter time to failure in GMFCS V hips, but this had a broad 95% confidence interval of 0.78–6.77, P = 0.13. The combined femoral and pelvic osteotomy approach yielded considerably better results compared with isolated femoral osteotomy, despite the hips being
significantly more displaced preoperatively (Table 3). Hips that were subjected to combined femoral and pelvic osteotomy had significantly higher median preoperative MP compared with hips that were subjected to femoral osteotomy alone (69 and 48%, respectively, P = 0.02). Of 19 hips that were subjected to a combined procedure, only two failed. Of these two failures, one represented a technical error, with reoperation at 1 month from the index procedure (which was judged not to have adequately reduced the hip). When this hip is excluded from analysis, the proportion of failures in the combined femoral and pelvic osteotomy group was significantly less than in the isolated femoral osteotomy group (Fisher exact test, P < 0.05).
Discussion This study aimed to describe the radiological outcome of reconstructive hip surgery in children with nonambulatory CP and determine whether GMFCS plays a role in predicting outcome. We found that a quarter of the hips that were subjected to reconstructive surgery had resubluxated or required further corrective surgery during the median 5-year follow-up period of the study. Hip surgeries in children who functioned at GMFCS V level did not have an overall higher failure rate, but did tend to fail sooner than those surgeries in children who functioned at a GMFCS IV level. For both groups of children, the best radiological outcomes were found after combined femoral and pelvic osteotomies. Although the relationship between GMFCS level and time to failure was not statistically significant because of the small sample size, the data suggested that GMFCS
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Radiological outcomes GMFCS IV and V Zhang et al. 433
may have some value in predicting the outcome of reconstructive hip surgery in children with CP. The trend toward a shorter time to failure in children who are GMFCS level V may reflect differences in weight bearing and bone remodeling between these groups. Our result differs from a previous review of 79 femoral osteotomies performed on 65 children, which found that ambulatory status had no prognostic importance for residual hip displacement at an average of 5 years of follow-up [11]. However, it is unclear in that study how ambulatory ability was defined. Better radiological outcomes were achieved with combined femoral and pelvic osteotomy compared with femoral osteotomy alone, despite these hips having significantly higher MP preoperatively. This was true for both GMFCS IV and V hips. Both groups achieved excellent radiological results following combined surgery, with only one failure in each group during follow-up. These results are consistent with previous reports. In one study of 52 hips in children with spastic CP, the authors found significantly lower MP at the final follow-up for combined osteotomies compared with femoral osteotomy alone, and no child subjected to the combined approach required further surgery, whereas 25% who underwent femoral osteotomy alone did [15]. Similarly, another study found that for more severely affected hips, the combined approach had a significantly lower rate of redislocation [16]. More recently, Huh and colleagues compared the outcome of 116 hips of mostly nonambulatory children who underwent either femoral osteotomy alone or femoral osteotomy combined with open reduction or pelvic osteotomy. They found that, despite the higher preoperative MP in the latter group, the results were as good as femoral osteotomy alone [17]. However, none of these studies assessed their results by the GMFCS level of the child. The high rate of redislocation and further surgery following isolated proximal femoral osteotomies highlights the need for ongoing hip surveillance for these children postoperatively and raises the question as to whether an isolated femoral osteotomy is the most appropriate index procedure. Although radiological outcome is an important measure of technical achievement, it is not clear whether a child with CP, GMFCS level V, whose radiographs show a final MP at skeletal maturity of 35% after femoral surgery will be worse off as an adult than a child whose radiographs show a final MP of 15%. To answer this question, a prospective, longitudinal, multisurgeon study is needed with a suite of outcome measures that assess technical outcomes but also outcomes of importance to the family and the child such as the presence of pain, seating, perineal hygiene, etc.
surveillance is to offer hip surgery for radiological changes, rather than problems identified by the child or the family. As a result, successful surgeries may not provide early functional benefits to the child or family, only reassurance of the reduced risks of a later painful dislocation. Good concordance has been suggested between radiological and clinical outcomes in this patient population [18], but further studies are needed. Second, we included two hips that were operated on at another center in the Auckland region as we had full documentation of the operative procedures and access to all radiographs. Finally, this is a retrospective review over a 15-year period with variability in duration of follow-up and subsequently not all hips were followed to skeletal maturity. Although our hip surveillance program has a good capture of children with CP within the region, it excludes children who may have died or removed from follow-up by parental choice. Thus, some cases of hip subluxation may have been missed from the review. Conclusion
We found that combined femoral and pelvic osteotomies have the most predictable radiological outcome for hip subluxation in children with CP who function at GMFCS level IV or V. When failure did occur, this appeared to be earlier for the hips of children who function at GMFCS level V but greater numbers are needed to be confident of this finding. Further work is needed to increase the numbers of patients studied and assess not only the radiological outcome but also functional and clinical outcomes in a prospective manner, as well as including children who function at GMFCS III levels.
Acknowledgements The researchers acknowledge the statistical support from the Starship Children’s Research Centre, which is supported by the Starship Foundation and the Auckland District Health Board. N.C.W. received funding from the Health Research Council of New Zealand and the New Zealand Orthopaedic Association Trust. Conflicts of interest
There are no conflicts of interest.
References 1 2 3 4
5
This study had several limitations. The first is that we examined only the radiological outcome and did not address functional outcomes. However, the goal of hip
6
Kerr Graham H, Selber P. Musculoskeletal aspects of cerebral palsy. J Bone Joint Surg Br 2003; 85:157–166. Access Economics Pty Ltd. The economic impact of cerebral palsy in Australia in 2007. Report. Australia: Cerebral Palsy Australia; 2008. Pritchett JW. The untreated unstable hip in severe cerebral palsy. Clin Orthop Relat Res 1983; 173:169–172. Cooke PH, Cole WG, Carey RP. Dislocation of the hip in cerebral palsy. Natural history and predictability. J Bone Joint Surg Br 1989; 71:441–446. Lonstein JE, Beck K. Hip dislocation and subluxation in cerebral palsy. J Pediatr Orthop 1986; 6:521–526. Moreau M, Drummond DS, Rogala E, Ashworth A, Porter T. Natural history of the dislocated hip in spastic cerebral palsy. Dev Med Child Neurol 1979; 21:749–753.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
434 Journal of Pediatric Orthopaedics B 2014, Vol 23 No 5
7 8
9
10
11 12
13
Soo B, Howard JJ, Boyd RN, Reid SM, Lanigan A, Wolfe R, et al. Hip displacement in cerebral palsy. J Bone Joint Surg Am 2006; 88:121–129. Wood E, Rosenbaum P. The gross motor function classification system for cerebral palsy: a study of reliability and stability over time. Dev Med Child Neurol 2000; 42:292–296. Palisano R, Rosenbaum P, Walter S, Russell D, Wood E, Galuppi B. Development and reliability of a system to classify gross motor function in children with cerebral palsy. Dev Med Child Neurol 1997; 39:214–223. Westbom L, Bergstrand L, Wagner P, Nordmark E. Survival at 19 years of age in a total population of children and young people with cerebral palsy. Dev Med Child Neurol 2011; 53:808–814. Valencia FG. Management of hip deformities in cerebral palsy. Orthop Clin North Am 2010; 41:549–559. Shore BJ, Yu X, Desai S, Selber P, Wolfe R, Graham HK. Adductor surgery to prevent hip displacement in children with cerebral palsy: the predictive role of the Gross Motor Function Classification System. J Bone Joint Surg Am 2012; 94:326–334. Pons C, Rémy-Néris O, Médée B, Brochard S. Validity and reliability of radiological methods to assess proximal hip geometry in children with
14
15
16
17
18
cerebral palsy: a systematic review. Dev Med Child Neurol 2013; 55:1089–1102. Reimers J. 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. Al-Ghadir M, Masquijo JJ, Guerra LA, Willis B. Combined femoral and pelvic osteotomies versus femoral osteotomy alone in the treatment of hip dysplasia in children with cerebral palsy. J Pediatr Orthop 2009; 29:779–783. Oh CW, Presedo A, Dabney KW, Miller F. Factors affecting femoral varus osteotomy in cerebral palsy: a long-term result over 10 years. J Pediatr Orthop B 2007; 16:23–30. Huh K, Rethlefsen SA, Wren TA, Kay RM. Surgical management of hip subluxation and dislocation in children with cerebral palsy: isolated VDRO or combined surgery? J Pediatr Orthop 2011; 31:858–863. Zenios M, Hannan M, Zafar S, Henry A, Galasko CS, Khan T. Clinical and radiological outcome of combined femoral and Chiari osteotomies for subluxed or dislocated hips secondary to neuromuscular conditions: a minimum of 10-year follow-up. Musculoskelet Surg 2012; 96:101–106.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Original article
Radiological outcome of reconstructive hip surgery in children with gross motor function classification system IV and V cerebral palsy Shiran Zhanga, Nichola C. Wilsona,b, Anna H. Mackeyb and Ngaire Susan Stotta,b Hip subluxation is common in children with cerebral palsy (CP). The aim of this study was to describe the radiological outcome of reconstructive hip surgery in children with CP, gross motor function classification system (GMFCS) level IV and V, and determine whether the GMFCS level plays a predictive role in outcome. This was a retrospective cohort study conducted at a tertiary-level pediatric hospital with a CP hip surveillance program. Of 110 children with GMFCS IV and V CP registered for hip surveillance, 45 underwent reconstructive hip surgery between 1997 and 2009, defined as varus derotational proximal femoral osteotomy with or without additional pelvic osteotomy. Eleven children were excluded because of lack of 12-month follow-up (n = 10) or missing clinical records (n = 1). Thus, 21 GMFCS IV children (median age 6 years at surgery) and 13 GMFCS V children (median age 5 years at surgery), who underwent 58 index surgeries, were included in the study. Clinical records and radiology were reviewed. The two surgical groups were femoral osteotomy (varus derotational femoral osteotomy with an AO blade plate or femoral locking plate fixation), or femoral ostetotomy with additional pelvic osteotomy. Reimer’s migration percentage (MP) was calculated from anteroposterior pelvis radiographs to determine the outcome for each hip independently. Failure was defined as
Introduction Cerebral palsy (CP) is the most common cause of disability in children, affecting 2–2.5 in 1000 live births per year in developed countries [1,2]. Hip subluxation and dislocation is highly prevalent in children with CP and causes significant morbidity, including pain, difficulties with positioning and perineal hygiene, pelvic obliquity, and possibly increased risk of scoliosis [3–6]. Hip subluxation results from imbalanced muscular forces acting on the hip, coupled with failure of normal bone development because of lack of weight bearing. The incidence of subluxation increases with the severity of CP and occurs in parallel with reduced ambulatory ability [5,7]. The gross motor function classification system (GMFCS) is a reliable and stable five-level scale that ranks ambulatory ability and musculoskeletal function in children with CP [8]. Children with CP functioning at GMFCS IV and V levels are considered nonambulatory, although some children who are GMFCS level IV can take some steps with support from another person. Children who function at GMFCS level V are wheelchair dependent and are medically frail, with a Swedish study reporting 1060-152X © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
MP of greater than 60% or further operation on the hip. Reconstructive surgeries were performed for 58 hips with a median preoperative MP of 55%. There were 15 failures at a median of 62 months, including nine failures in 35 GMFCS IV hips and six failures in 23 GMFCS V hips. Overall, GMFCS V hips tended to fail earlier, (hazard ratio 2.3) with a median time to failure of 78 and 39 months for GMFCS IV and V hips, respectively. Combined femoral and pelvic osteotomies had the lowest failure rates in both groups of patients. The GMFCS classification may have some predictive value for outcomes following reconstructive hip surgery, with surgery for GMFCS V hips tending to fail earlier. J Pediatr Orthop B 23:430–434 © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Journal of Pediatric Orthopaedics B 2014, 23:430–434 Keywords: cerebral palsy, gross motor function classification system, hip surgery, radiological outcome a Department of Surgery, Faculty of Medical and Health Sciences, The University of Auckland and bStarship Children’s Hospital, Auckland, New Zealand
Correspondence to Ngaire Susan Stott, MBChB, FRACS, PhD, Faculty of Medical and Health Sciences, The University of Auckland, Private Bag 92019, Auckland Mail Centre, Auckland, 1142 New Zealand Tel: + 64 9 3737599 x82861; fax: + 64 9 3677159; e-mail: [email protected]
only a 60% survival rate into adulthood [9,10]. Data from the Victorian Cerebral Palsy Register show that 69 and 90% of children with CP between the ages of 6 and 14 years, who function at GMFCS IV or V levels, respectively, develop hip subluxation or dislocation [7]. Management of hip subluxation is multidisciplinary and includes hip surveillance, nonoperative, and operative interventions. Operative interventions for identified hip subluxation may be soft tissue release (adductor muscle release) or bony procedures, usually femoral or pelvic osteotomies. Reconstructive hip surgery refers to femoral and pelvic osteotomies that reduce the subluxed or dislocated hip by correcting osseous deformities, with additional open reductions when the hip cannot be reduced closed. Salvage procedures such as femoral head resection are reserved for longstanding painful dislocations in older children [11]. A recent study describing outcomes following soft tissue surgery showed that the GMFCS level is associated with outcome, with higher rates of further surgery in GMFCS level IV or V children [12]. The predictive value of the DOI: 10.1097/BPB.0000000000000075
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Radiological outcomes GMFCS IV and V Zhang et al. 431
GMFCS following reconstructive hip surgery is yet to be studied. This study aims to describe the radiological outcome of reconstructive hip surgery in children with nonambulatory CP and determine whether the GMFCS level plays a predictive role in outcome in this group of children.
Methods Study design
This is a retrospective cohort study of children with CP, GMFCS IV and V, registered on a tertiary referral service hip surveillance program, who had undergone reconstructive hip surgery between 1997 and 2009. Children on the surveillance program who were GMFCS III or less were not part of this study as only three of 38 children had needed reconstructive hip surgery. Children who had not had at least 12 months of postoperative follow-up were excluded. The study received ethics approval from the regional and institutional review boards. Data collection
Demographic and clinical data were retrieved from the hospital’s Clinical Record Information System. This included the GMFCS level as determined by the examining physician or physiotherapist, date of birth, sex, and the operation details. For the purpose of this study, the surgeries were classified as either femoral osteotomy (varus derotational proximal femoral osteotomy with an AO blade plate or femoral locking plate fixation) or combined femoral and pelvic osteotomy, with or without additional soft tissue release and open reduction. Soft tissue release for femoral osteotomy was sequential release of adductor longus, gracilis, and up to 50% of adductor brevis to achieve 50° of abduction in extension and thus facilitate the varus correction. If the range of hip abduction was satisfactory under anesthesia, no adductor release was performed. In three hips that were subjected to additional pelvic osteotomy, a psoas lengthening at the pelvis brim as well as the sequential adductor lengthening as above were performed. No obturator neurectomies or hamstring releases were performed in this group of children. Radiographic measurements
Pelvic radiographs were retrieved from the hospital’s Picture Archiving and Communication System, which archives all radiographs taken from 2002 onwards. Radiographs taken before 2002 were in hardcopy and digitized onto Picture Archiving and Communication System before analysis. Of the radiological measures, the migration percentage (MP) is the only measurement that correlates consistently with hip displacement [13]. In this study, the outcome of the index surgery for each hip was considered independently. The index surgery was defined as the first procedure performed on that hip irrespective of whether the
contralateral hip had been operated on. A surgery was deemed failed if MP reached 60% or if further surgery was performed on that hip. MP was measured on standard anteroposterior pelvic radiographs as described by Reimers [14]. Measurements were taken on digitized radiographs by the same observer to avoid interobserver error. Radiographs of inadequate quality were excluded. The duration of follow-up was defined as from the time between the index procedure to the time of the last follow-up radiograph. The preoperative and all subsequent postoperative radiographs were reviewed. Statistical analysis
Hips were grouped by demographic and clinical characteristics. Cox regression was used to generate hazard ratios (HR) comparing outcomes in different groups. An HR ratio greater than 1 indicates a higher failure rate in that group and HR less than 1 indicates a lower failure rate. Statistical significance was defined as P value less than 0.05. All statistical analyses were carried out on StatsDirect 2.7.8 (StatsDirect Ltd, Cheshire, UK).
Results The hip surveillance program registered 110 GMFCS IV and V children, 45 of whom had undergone reconstructive hip surgery. Ten children were excluded as the duration of radiological follow-up was less than 12 months. One further child was excluded because of missing documentation of her surgery (performed overseas). The preoperative radiographs for five hips could not be retrieved, but the MP measures recorded in the radiograph reports were available and thus used as the preoperative baseline. This led to a final sample of 34 children, with a median age at surgery of 5 years (range 1–13), as described in Table 1. All children had bilateral CP and the majority had a diagnosis of spastic quadriplegia. Twenty-four children had both hips operated on; thus, 58 index surgeries were included in the analysis (Tables 2 and 3). All except two hips were operated on at our own institution. Reconstructive hip surgery was performed to correct existing hip migration of at least one hip, with additional pelvic osteotomy performed on the most severely subluxated hips. The decision to proceed with pelvic osteotomy was usually made intraoperatively when femoral osteotomy alone failed to provide adequate coverage and stability. Forty-three of the 58 hips had an adductor release to facilitate varus correction. The Table 1
Participant demographics
GMFCS IV GMFCS V Total
Number of children
Sex (boys : girls)
Median age at first surgery (years)
21 13 34
16 : 5 5:8 21 : 13
6 (1–13) 5 (2–12) 5 (1–13)
GMFCS, gross motor function classification system.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
432 Journal of Pediatric Orthopaedics B 2014, Vol 23 No 5
Table 2
Summary of index procedures by GMFCS level Number of hips
Median age in years at time of surgery (range)
Median preoperative MP (%)
Median postoperative MP (%)
Number of failures
Median months to failure (range)
Median months duration of follow-up (range)
35 23 58
6 (1–13) 5 (2–12) 5 (1–13)
48 (8–100) 66 (12–100) 55 (12–100)
0 (0–34) 0 (0–57) 0 (0–57)
9 6 15
78 (36–105) 39 (1–82) 62 (1–105)
65 (14–157) 50 (14–159) 62.5 (14–159)
GMFCS IV GMFCS V Total
GMFCS, gross motor function classification system; MP, migration percentage.
Table 3
Summary of index procedures by type of procedure Number of hips
Median age in years at time of surgery (range)
Femoral osteotomy (n = 39) GMFCS IV 23 5 GMFCS V 16 3.5 Subtotal 39 5 Femoral and pelvic osteotomy (n = 19) GMFCS IV 12 10 GMFCS V 7 5 Subtotal 19 9
Median preoperative MP (range) (%)
Median postoperative MP (range) (%)
Number of failures
Median months to failure (range)
Median months duration of follow-up (range)
(1–11) (2–9) (1–13)
47 (8–95) 52 (12–100) 48 (8–100)
0 (0–34) 0 (0–57) 0 (0–57)
8 5 13
58 (36–105) 62 (13–82) 62 (13–105)
87 (19–140) 54.5 (14–159) 70 (14–159)
(1–13) (2–12) (1–13)
55 (38–100) 72 (50–100) 69 (38–100)*
0 (0–18) 0 (0–0) 0 (0–18)
1 1a 2
Not applicable Not applicable 45 (1–89)
58.5 (14–157) 39 (15–61) 55 (14–157)
GMFCS, gross motor function classification system; MP, migration percentage. Failure because of technical error at 1 month. *Significantly higher than hips subjected to isolated femoral osteotomy (P = 0.02). a
median duration of follow-up across the group was 62.5 months (range 14–159 months). The median preoperative MP was 55% (12–100%), with no difference between GMFCS IV hips and GMFCS V hips (48 and 66%, respectively, P = 0.15). There was a trend for children who were GMFCS level V to be younger at the time of the index surgery (P = 0.06) Failure rates
Fifteen of the 58 reconstructive hip surgeries (26%) failed over the period of follow-up; nine of the failures were GMFCS IV hips and six were GMFCS V hips. Six hips were subjected to further surgery because of resubluxation with MP equal to or greater than 60%; a further five hips resubluxated to an MP of 60%, but did not undergo any further surgery during the time frame of the study. Of the remaining hips that resubluxated, three subsequently underwent salvage surgery, including femoral head resection, to correct soft tissue contractures but not to reduce the dislocated hip. For the final hip, the index procedure failed at 1 month and further surgery was performed. The median times to failure for GMFCS IV and V hips were 78 and 39 months, respectively, as shown in Table 2. This difference was not statistically significant (P = 0.17) because of the small sample size. The HR for the difference in time to failure between the groups was 2.3, indicating a shorter time to failure in GMFCS V hips, but this had a broad 95% confidence interval of 0.78–6.77, P = 0.13. The combined femoral and pelvic osteotomy approach yielded considerably better results compared with isolated femoral osteotomy, despite the hips being
significantly more displaced preoperatively (Table 3). Hips that were subjected to combined femoral and pelvic osteotomy had significantly higher median preoperative MP compared with hips that were subjected to femoral osteotomy alone (69 and 48%, respectively, P = 0.02). Of 19 hips that were subjected to a combined procedure, only two failed. Of these two failures, one represented a technical error, with reoperation at 1 month from the index procedure (which was judged not to have adequately reduced the hip). When this hip is excluded from analysis, the proportion of failures in the combined femoral and pelvic osteotomy group was significantly less than in the isolated femoral osteotomy group (Fisher exact test, P < 0.05).
Discussion This study aimed to describe the radiological outcome of reconstructive hip surgery in children with nonambulatory CP and determine whether GMFCS plays a role in predicting outcome. We found that a quarter of the hips that were subjected to reconstructive surgery had resubluxated or required further corrective surgery during the median 5-year follow-up period of the study. Hip surgeries in children who functioned at GMFCS V level did not have an overall higher failure rate, but did tend to fail sooner than those surgeries in children who functioned at a GMFCS IV level. For both groups of children, the best radiological outcomes were found after combined femoral and pelvic osteotomies. Although the relationship between GMFCS level and time to failure was not statistically significant because of the small sample size, the data suggested that GMFCS
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Radiological outcomes GMFCS IV and V Zhang et al. 433
may have some value in predicting the outcome of reconstructive hip surgery in children with CP. The trend toward a shorter time to failure in children who are GMFCS level V may reflect differences in weight bearing and bone remodeling between these groups. Our result differs from a previous review of 79 femoral osteotomies performed on 65 children, which found that ambulatory status had no prognostic importance for residual hip displacement at an average of 5 years of follow-up [11]. However, it is unclear in that study how ambulatory ability was defined. Better radiological outcomes were achieved with combined femoral and pelvic osteotomy compared with femoral osteotomy alone, despite these hips having significantly higher MP preoperatively. This was true for both GMFCS IV and V hips. Both groups achieved excellent radiological results following combined surgery, with only one failure in each group during follow-up. These results are consistent with previous reports. In one study of 52 hips in children with spastic CP, the authors found significantly lower MP at the final follow-up for combined osteotomies compared with femoral osteotomy alone, and no child subjected to the combined approach required further surgery, whereas 25% who underwent femoral osteotomy alone did [15]. Similarly, another study found that for more severely affected hips, the combined approach had a significantly lower rate of redislocation [16]. More recently, Huh and colleagues compared the outcome of 116 hips of mostly nonambulatory children who underwent either femoral osteotomy alone or femoral osteotomy combined with open reduction or pelvic osteotomy. They found that, despite the higher preoperative MP in the latter group, the results were as good as femoral osteotomy alone [17]. However, none of these studies assessed their results by the GMFCS level of the child. The high rate of redislocation and further surgery following isolated proximal femoral osteotomies highlights the need for ongoing hip surveillance for these children postoperatively and raises the question as to whether an isolated femoral osteotomy is the most appropriate index procedure. Although radiological outcome is an important measure of technical achievement, it is not clear whether a child with CP, GMFCS level V, whose radiographs show a final MP at skeletal maturity of 35% after femoral surgery will be worse off as an adult than a child whose radiographs show a final MP of 15%. To answer this question, a prospective, longitudinal, multisurgeon study is needed with a suite of outcome measures that assess technical outcomes but also outcomes of importance to the family and the child such as the presence of pain, seating, perineal hygiene, etc.
surveillance is to offer hip surgery for radiological changes, rather than problems identified by the child or the family. As a result, successful surgeries may not provide early functional benefits to the child or family, only reassurance of the reduced risks of a later painful dislocation. Good concordance has been suggested between radiological and clinical outcomes in this patient population [18], but further studies are needed. Second, we included two hips that were operated on at another center in the Auckland region as we had full documentation of the operative procedures and access to all radiographs. Finally, this is a retrospective review over a 15-year period with variability in duration of follow-up and subsequently not all hips were followed to skeletal maturity. Although our hip surveillance program has a good capture of children with CP within the region, it excludes children who may have died or removed from follow-up by parental choice. Thus, some cases of hip subluxation may have been missed from the review. Conclusion
We found that combined femoral and pelvic osteotomies have the most predictable radiological outcome for hip subluxation in children with CP who function at GMFCS level IV or V. When failure did occur, this appeared to be earlier for the hips of children who function at GMFCS level V but greater numbers are needed to be confident of this finding. Further work is needed to increase the numbers of patients studied and assess not only the radiological outcome but also functional and clinical outcomes in a prospective manner, as well as including children who function at GMFCS III levels.
Acknowledgements The researchers acknowledge the statistical support from the Starship Children’s Research Centre, which is supported by the Starship Foundation and the Auckland District Health Board. N.C.W. received funding from the Health Research Council of New Zealand and the New Zealand Orthopaedic Association Trust. Conflicts of interest
There are no conflicts of interest.
References 1 2 3 4
5
This study had several limitations. The first is that we examined only the radiological outcome and did not address functional outcomes. However, the goal of hip
6
Kerr Graham H, Selber P. Musculoskeletal aspects of cerebral palsy. J Bone Joint Surg Br 2003; 85:157–166. Access Economics Pty Ltd. The economic impact of cerebral palsy in Australia in 2007. Report. Australia: Cerebral Palsy Australia; 2008. Pritchett JW. The untreated unstable hip in severe cerebral palsy. Clin Orthop Relat Res 1983; 173:169–172. Cooke PH, Cole WG, Carey RP. Dislocation of the hip in cerebral palsy. Natural history and predictability. J Bone Joint Surg Br 1989; 71:441–446. Lonstein JE, Beck K. Hip dislocation and subluxation in cerebral palsy. J Pediatr Orthop 1986; 6:521–526. Moreau M, Drummond DS, Rogala E, Ashworth A, Porter T. Natural history of the dislocated hip in spastic cerebral palsy. Dev Med Child Neurol 1979; 21:749–753.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
434 Journal of Pediatric Orthopaedics B 2014, Vol 23 No 5
7 8
9
10
11 12
13
Soo B, Howard JJ, Boyd RN, Reid SM, Lanigan A, Wolfe R, et al. Hip displacement in cerebral palsy. J Bone Joint Surg Am 2006; 88:121–129. Wood E, Rosenbaum P. The gross motor function classification system for cerebral palsy: a study of reliability and stability over time. Dev Med Child Neurol 2000; 42:292–296. Palisano R, Rosenbaum P, Walter S, Russell D, Wood E, Galuppi B. Development and reliability of a system to classify gross motor function in children with cerebral palsy. Dev Med Child Neurol 1997; 39:214–223. Westbom L, Bergstrand L, Wagner P, Nordmark E. Survival at 19 years of age in a total population of children and young people with cerebral palsy. Dev Med Child Neurol 2011; 53:808–814. Valencia FG. Management of hip deformities in cerebral palsy. Orthop Clin North Am 2010; 41:549–559. Shore BJ, Yu X, Desai S, Selber P, Wolfe R, Graham HK. Adductor surgery to prevent hip displacement in children with cerebral palsy: the predictive role of the Gross Motor Function Classification System. J Bone Joint Surg Am 2012; 94:326–334. Pons C, Rémy-Néris O, Médée B, Brochard S. Validity and reliability of radiological methods to assess proximal hip geometry in children with
14
15
16
17
18
cerebral palsy: a systematic review. Dev Med Child Neurol 2013; 55:1089–1102. Reimers J. 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. Al-Ghadir M, Masquijo JJ, Guerra LA, Willis B. Combined femoral and pelvic osteotomies versus femoral osteotomy alone in the treatment of hip dysplasia in children with cerebral palsy. J Pediatr Orthop 2009; 29:779–783. Oh CW, Presedo A, Dabney KW, Miller F. Factors affecting femoral varus osteotomy in cerebral palsy: a long-term result over 10 years. J Pediatr Orthop B 2007; 16:23–30. Huh K, Rethlefsen SA, Wren TA, Kay RM. Surgical management of hip subluxation and dislocation in children with cerebral palsy: isolated VDRO or combined surgery? J Pediatr Orthop 2011; 31:858–863. Zenios M, Hannan M, Zafar S, Henry A, Galasko CS, Khan T. Clinical and radiological outcome of combined femoral and Chiari osteotomies for subluxed or dislocated hips secondary to neuromuscular conditions: a minimum of 10-year follow-up. Musculoskelet Surg 2012; 96:101–106.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
