Effect of Hip Reconstructive Surgery on Health-Related Quality of Life of Non-Ambulatory Children with Cerebral Palsy

Transcription

1 1190 COPYRIGHT Ó 2016 BY THE OURNAL OF BONE AND OINT SURGERY, INCORPORATED Effect of Hip Reconstructive Surgery on Health-Relate Quality of Life of Non-Ambulatory Chilren with Cerebral Palsy Rachel DiFazio, PhD, RN, PPCNP-BC, FAAN, Benjamin Shore, MD, MPH, FRCSC, uith A. Vessey, PhD, MBA, RN, FAAN, Patricia E. Miller, MS, an Brian D. Snyer, MD, PhD Investigation performe at Boston Chilren s Hospital, Boston, Massachusetts Backgroun: The primary aim of this stuy was to evaluate the relationship of the migration percentage (a raiographic metric quantifying hip isplacement) in chilren with Gross Motor Function Classification System (GMFCS) level-iv or V cerebral palsy an spastic hip ysplasia to the acetabular inex an the health-relate quality of life (HRQOL) as measure with the Caregiver Priorities an Chil Health Inex of Life with Disabilities (CPCHILD) before an after reconstructive hip surgery. Methos: In a prospective cohort stuy (n = 38), the migration percentage, acetabular inex, an CPCHILD scores were analyze using the Pearson correlation analysis immeiately before reconstructive hip surgery an at 6 weeks an 3, 6, 12, an 24 months after the surgery. Subgroup analysis was use to compare patients who ha a preoperative migration percentage of 50% with those who ha a preoperative migration percentage of <50% an to compare the acetabular inex between patients who ha a pelvic osteotomy an those who ha not. Linear mixe moels were use to analyze changes in the migration percentage, acetabular inex, an CPCHILD scores over time. Results: The preoperative migration percentage negatively correlate with the preoperative CPCHILD score (r = 20.50; p = 0.002). This relationship continue throughout the follow-up perio such that, for each aitional 1% correction in migration percentage, the CPCHILD total score increase by 0.2 point (p < 0.001). There was no correlation between the acetabular inex an CPCHILD total score before or after surgery (p = 0.09 to 0.71). The preoperative CPCHILD total scores iffere between the migration-percentile groups (mean ifference = 13 points; 95% confience interval = 3.3 to 22.8; p = 0.01). However, after hip surgery, the CPCHILD score improve similarly for both groups. Conclusions: These ata support the effectiveness of reconstructive hip surgery for the treatment of spastic hip ysplasia to improve the HRQOL of non-ambulatory chilren with severe cerebral palsy. Level of Evience: Prospective Level IV. See Instructions for Authors for a complete escription of levels of evience. Peer Review: This article was reviewe by the Eitor-in-Chief an one Deputy Eitor, an it unerwent bline review by two or more outsie experts. The Deputy Eitor reviewe each revision of the article, an it unerwent a final review by the Eitor-in-Chief prior to publication. Final corrections an clarifications occurreuring one or more exchanges between the author(s) an copyeitors. Cerebral palsy is a neuroevelopmental conition beginning in early chilhoo an persisting throughout life. Although impaire motor function is its hallmark, chilren with cerebral palsy have varying egrees of functional ability that has been classifie accoring to the Gross Motor Function Classification System (GMFCS) 1. This vali, 5-level orinal graing system is clinically relevant for classifying an preicting motor function of chilren with cerebral palsy 2-4. Non-ambulatory chilren with cerebral palsy (GMFCS level IV or V) represent the severe en of the spectrum, with global impairments in all evelopmental omains 4,5. Spastic hip ysplasia/instability is a common musculoskeletal eformity in chilren with cerebral palsy, secon only to equinus eformity 6,7. The incience is irectly relate to the Disclosure: This stuy was fune by a Boston Chilren s Hospital Department of Patient Services Research Grant. The funing agency ha no role in the esign or conuct of the stuy; collection, management, analysis, or interpretation of the ata; preparation, review, or approval of the manuscript; or ecision to submit the manuscript for publication. The Disclosure of Potential Conflicts of Interest forms are provie with the online version of the article. Bone oint Surg Am. 2016;98:

2 1191 T HE OURNAL OF B ONE &OINT SURGERY BS. ORG TABLE I Demographic an Clinical Characteristics of Patients an Caregivers (N = 38) No. (%)* Patients Male sex 24 (63%) Age (yr) 10.1 ± 3.90 Ethnicity Hispanic or Latino 5 (13%) Not Hispanic or Latino 33 (87%) Race White 32 (84%) Black 2 (5%) Asian 2 (5%) Other or unknown 2 (5%) GMFCS level IV 18 (47%) V 20 (53%) Comorbiities Seizure isorer 23 (61%) Shunte hyrocephalus 5 (13%) Apnea/CPAP/BiPAP 3 (8%) Tracheostomy 1 (3%) Caregivers Male sex 4 (11%) Age (yr) 41.1 ± 7.97 *Except for age, which is given as the mean an stanar eviation. CPAP = continuous positive airway pressure machine an BiPAP = bilevel positive airway pressure machine. isease severity an thus higher in chilren with more severe cerebral palsy 8. In population-base stuies, 35% of chilren with cerebral palsy were foun to be affecte by spastic hip ysplasia 6. Reimers hip migration percentage, a raiographic metric to quantify hip isplacement, measures the extent of hip instability 9. A hip with a migration percentage of 30% is at risk for subluxation 10, an a hip with a migration percentage of 50% is at risk for islocation 11. The acetabular inex, use to etermine the epth of the acetabulum an orientation of the sourcil, quantifies theegreeofacetabularysplasia 12. Hips at risk for subluxation/ islocation with a migration percentage of >30% to 50% are often associate with a ysplastic acetabulum (acetabular inex of >30 ) with marke posterolateral eficiency 13. Untreate spastic hip ysplasia can progress to hip subluxation or islocation, which can become painful an negatively affect walking, staning, sitting, an quality of life 6,14,15. Spastic islocate hips often eteriorate in early aulthoo, leaing to painful egenerative arthritis 11,16. Surgical reconstruction of ysplastic hips in chilren with cerebral palsy has resulte in improvements in pain an mobility Health-relate quality of life (HRQOL) is a patient-focuse multiimensional concept incorporating physical, psychological, an social well-being 21,22. HRQOL metrics provie information on the effects of chronic conitions on a person s life an the appropriateness of interventions esigne to ameliorate those effects 23. Historically, research relate to spastic hip ysplasia in chilren with cerebral palsy has focuse on the classification an prevalence of hip isorers, natural history, surgical inications, surgical techniques, an surgical complications 7,8,11,24. Few stuies have assesse the HRQOL of chilren with severe cerebral palsy with spastic islocate hips 25, an we are not aware of any publishe prospective stuy of the effect of reconstructive hip surgery on the HRQOL of such chilren. The primary aims of this stuy were to (1) evaluate the relationship between the preoperative migration percentage an the chil s preoperative HRQOL as measure by the Caregiver Priorities an Chil Health Inex of Life with Disabilities (CPCHILD) questionnaire, with the hypothesis that increases in the migration percentage woul be associate with ecreases in the CPCHILD scores, an (2) evaluate the relationship between the migration percentage an CPCHILD score following hip reconstructive surgery while ajusting for other clinical factors. The seconary aim was to examine the relationship at baseline (preoperatively) between the migration percentage an acetabular inex. Materials an Methos Institutional review boar approval was obtaine prior to the initiation of this prospective cohort stuy. Eligible participants were recruite from a multiisciplinary cerebral palsy clinic at an acaemic chilren s hospital between February 2011 an November Informe written consent was obtaine from all caregivers. To be eligible for the stuy, a caregiver ha to be fluent in English an tening, at home, a chil with cerebral palsy (GMFCS level IVor V) who was between 3 an 25 years of age an scheule for reconstructive hip surgery to correct severe spastic hip ysplasia. The sample-size calculation was base on the primary hypothesis that there was a negative correlation between the preoperative migration percentage an preoperative CPCHILD total score. For a Pearson correlation analysis, a minimum of 37 subjects at baseline is neee to test for a correlation of >0.4 assuming a 5% significance level an 80% power. The caregivers, who were enrolle at the chil s preoperative visit, complete an informe consent form, a emographic worksheet, an the CPCHILD questionnaire. All patients unerwent a femoral an/or pelvic osteotomy with or without soft-tissue releases by a boar-certifie orthopaeic surgeon specializing in the treatment of neuromuscular orthopaeic isorers. At 6 weeks an at 3, 6, 12, an 24 months postoperatively, the caregivers complete the CPCHILD questionnaire again an raiographic measurements were obtaine. TABLE II Types of Surgery (N = 38) Type of Hip Surgery No. (%) Femoral osteotomy only 14 (37%) Pelvic osteotomy only 1 (3%) Femoral an pelvic osteotomies 23 (61%) Soft-tissue releases 33 (87%) Revision proceure prior to stuy 17 (45%) perio

3 1192 T HE OURNAL OF B ONE &OINT SURGERY BS. ORG TABLE III Demographic, Clinical, an Outcome Characteristics by Migration-Percentage Group Characteristic Baseline Migration Percentage <50% (N = 11) Baseline Migration Percentage 50% (N = 27) P Value Male sex* 7 (64%) 17 (63%) 1.00 Age (yr) 11.1 ± ± GMFCS level* 0.01 IV 9 (82%) 9 (33%) V 2 (18%) 18 (67%) Revision proceure prior to stuy perio* 3 (27%) 14 (52%) 0.28 Comorbiities* Seizure isorer 4 (36%) 19 (70%) 0.07 Shunte hyrocephalus 0 (0%) 5 (19%) 0.30 Apnea/CPAP/BiPAP 0 (0%) 3 (11%) 0.54 Tracheostomy 0 (0%) 1 (4%) 1.00 Migration percentage Baseline 34 (30-40) 100 (78-100) 6 wk 19 (16-22) 2.5 (0-14) mo 28 (6-34) 0 (0-15) mo 27 (24-27) 5 (0-21) yr 11 (0-22) 14.5 (0-30) yr 14.5 (3-32) 11 (0-18) 0.56 CPCHILD total score Baseline 58.9 ± ± wk 48.9 ± ± mo 53.7 ± ± mo 63.2 ± ± yr 62.7 ± ± yr 62.2 ± ± *The values are given as the number with the percentage in parentheses. The values are given as the mean an stanareviation. CPAP = continuous positive airway pressure machine an BiPAP = bilevel positive airway pressure machine. The values are given as the meian an interquartile range. The CPCHILD is a caregiver proxy isease-specific measure of the HRQOL of non-ambulatory chilren with severe cerebral palsy 23. The measure consists of 37 items categorize into 6 omains: (1) activities of aily living (9 items); (2) positioning, transferring, an mobility (8 items); (3) comfort an emotions (9 items); (4) communication an social interaction (7 items); (5) health (3 items); an (6) overall quality of life (1 item). Items are rate on a 6-point orinal scale. A level of assistance moifier base on a 4-point scale is inclue for activity items, an a 3-point level of intensity scale is inclue for symptom items. Scores are reporte for each omain, an the total score ranges from 0 (worst) to 100 (best). CPCHILD is reliable an vali 23 an has emonstrate responsiveness to change following hip reconstructive surgery 26. One investigator reviewe stanarize preoperative an postoperative anteroposterior hip an pelvic raiographs. Stanarize metrics were use to escribe hip ysplasia, an the migration percentage 10 an the acetabular inex 12 were measure on all raiographs. Descriptive statistics were use to examine characteristics of the cohort. Correlations among the migration percentage, acetabular inex, an CPCHILD scores were analyze using a Pearson correlation analysis immeiately before an at 5 time points after surgery. Linear mixe moels were use to analyze the change in CPCHILD score relative to the migration percentages an acetabular inices over time. Piecewise moels were evelope for the CPCHILD scores, given the effect of surgical correction on the migration percentage an acetabular inex. On the basis of prior analysis, HRQOL outcomes were expecte to ecrease seconary to pain an surgical trauma an then improve graually over the ensuing 6 to 12 weeks following surgery 26. Thus, each moel was compose of 3 sections: preoperatively to 6 weeks, 6 weeks to 3 months, an 3 months to 24 months. Correlation structures (unstructure or compoun symmetry) for each moel were chosen on the basis of moel fit, etermine using the likelihoo ratio statistics as well as the Akaike information criterion an Bayesian information criterion. In general, a lower criterion was better, an for each outcome the moels with both a lower Akaike information criterion an a lower Bayesian information criterion were selecte. Subgroup analyses were conucte to compare patients with a baseline migration percentage of <50% to patients with a baseline migration percentage of 50% an to compare acetabular inices between patients who ha a pelvic osteotomy an those who i not. For patients with bilateral hip ysplasia, the more severely affecte hip was use in the analysis since both hips from the same patient are not truly inepenent. Patient an hip characteristics were compare across inex groups. Clinical factors with potential contributing effects, incluing GMFCS level, age, sex, an comorbiities, were also analyze as fixe effects in multivariable linear mixe moels for the CPCHILD score over time.

4 1193 T HE OURNAL OF B ONE &OINT SURGERY BS. ORG TABLE IV Change in CPCHILD Domain an Total Scores from Baseline to 12 Months an from 12 to 24 Months Change from Baseline to 12 Mo (N = 34) Change from 12 to 24 Mo (N = 27) Mean Difference 95% CI P Value Mean Difference 95% CI P Value All patients Activities of aily living Positioning, transferring, an mobility < Comfort an emotions Communication an social interaction Health Quality of life Total score < Baseline migration percentage <50% Activities of aily living Positioning, transferring, an mobility Comfort an emotions Communication an social interaction Health Quality of life Total score Baseline migration percentage 50% Activities of aily living < Positioning, transferring, an mobility < Comfort an emotions Communication an social interaction Health Quality of life Total score < Results Fifty-six patients an caregivers were assesse for eligibility. Fourteen i not meet the inclusion criteria, 3 were lost to follow-up, an 1 caregiver ecline to participate because of ifficulties coping with the emans of caring for the chil. Thus, 38 caregivers an patients were analyze. Patient an caregiver emographics are isplaye in Table I. The average age (an stanareviation) of the patients was 10.1 ± 3.90 years, an the average age of the caregivers was 41.1 ± 7.97 years. Fourteen patients (37%) unerwent only femoral osteotomy, 1 (3%) unerwent only pelvic osteotomy, an 23 (61%) unerwent combine femoral an pelvic osteotomies. Concomitant soft-tissue proceures were performe in 33 patients (87%) (Table II). The overall cohort ha a meian preoperative migration percentage of 81.5% (interquartile range [IQR], 44.5% to 100%) an a meian preoperative acetabular inex of 23 (IQR, 15 to 30 ). Twenty-seven subjects ha a preoperative migration percentage of 50%, with a meian of 100% (IQR, 78% to 100%), while 11 subjects ha a preoperative migration percentage of <50%, with a meian of 34% (IQR, 30% to 40%). No ifferences were ientifie between patient groups with respect to sex (p = 1.00), age (p = 0.23), or comorbiities (p = 0.07 to 1.00) (Table III). Most (82%) of the patients with a migration percentage of <50% ha GMFCS level-iv cerebral palsy, whereas most (67%) of those with a migration percentage of 50% ha GMFCS level V (p = 0.01). The migration percentage, acetabular inex, an CPCHILD scores were examine before an after surgery. The migration percentage an acetabular inex were moerately correlate preoperatively (r = 0.50; p = 0.02) but were much less so after surgery (r = 0.37; p = 0.046), reflecting the high percentage of patients who unerwent concomitant femoral osteotomy an acetabuloplasty. The migration percentage improve by a meian of 65% from baseline to 6 weeks postoperatively (p < 0.001) with no significant changes in migration percentage over the 2-year follow-up perio (p = 0.22 to 0.99). The meian migration percentage was 11% at the 2-year follow-up evaluation (Fig. 1). There was a significant ecline in the CPCHILD total score from baseline to 6 weeks postoperatively (p < 0.001), followe by a steay increase over the ensuing 12 months after surgery (p < 0.001). There was no significant change in the CPCHILD total score after 12 months (p = 0.60). The average CPCHILD total score increase from 49.6 points at baseline to 58.9 points at 12 an 24 months postoperatively (mean

5 1194 T HE OURNAL OF B ONE &OINT SURGERY BS. ORG Fig. 1 Change in migration percentage over time. The squares represent the meian migration percentage at each time point an the I-bars represent the 1st an 3r quartiles aroun each meian. ifference = 9 points; 95% confience interval [CI] = 5.6 to 13.0 points; p < 0.001) (Fig. 2). The CPCHILD scores increase significantly from baseline to 12 months after surgery in all omains except for communication an social interaction (Fig. 2 an Table IV) an increase significantly from baseline to 24 months after surgery in all omains except for activities of aily living, which ecline by an average of 3 points from 12 months to 24 months. The greatest improvement was in the comfort an emotions omain, which increase an average of 11 points from baseline to 24 months (p < 0.001) after surgery. The preoperative migration percentage was negatively correlate with the preoperative CPCHILD total score (r = 20.50; p = 0.002). This relationship continue throughout Fig. 2 Change in CPCHILD total anomain scores over time. The symbols represent the mean CPCHILD total score at each time point an the I-bars represent the upper an lower 95% confience intervals aroun each mean.

6 1195 T HE OURNAL OF B ONE &OINT SURGERY BS. ORG Fig. 3 Relationship between baseline migration percentage an baseline CPCHILD total score. the follow-up perio such that, for each aitional 1% correctioninthemigrationpercentage,thecpchildtotalscore increase by 0.2 point (p < 0.001) (Fig. 3). There was no correlation between the acetabular inex an the CPCHILD total score before or after surgery (p = 0.09 to 0.71). The preoperative CPCHILD total score iffere significantly between the 50% an <50% migration-percentage groups, with a mean ifference of 13 points (95% CI = 3.3 to 22.8 points; p = 0.01). However, after the hip surgery, the CPCHILD total score improve similarly for both groups (Fig. 4). In the Fig. 4 Change in CPCHILD total score over time accoring to migration-percentage (MP) group (<50% or 50%).

7 1196 T HE OURNAL OF B ONE &OINT SURGERY BS. ORG immeiate postoperative perio, there was no ifference in the CPCHILD total scores between groups at either the 6-week (p = 0.12) or the 3-month (p = 0.31) follow-up visit. However, at 6 months, the <50% migration-percentage group ha a significantly higher CPCHILD score (p = 0.02) than the 50% migration-percentage group. At 12 months (p = 0.13) an 24 months (p = 0.40), the migration-percentage groups i not iffer significantly with regar to the CPCHILD total score. In the 24 patients who ha a pelvic osteotomy, there was no correlation between the acetabular inex an the CPCHILD total score at baseline or uring the postoperative follow-up (p = 0.36 to 1.00). The change in the CPCHILD total score was unaffecte by patient age (p = 0.52), sex (p = 0.51), GMFCS level (p = 0.13), surgical proceure (p = 0.32), or number or type of comorbiities. The caregiver s sex an age also ha no effect on the CPCHILD total score. Discussion Our stuy results emonstrate that reconstructive hip surgery to improve hip stability in chilren with spastic hip ysplasia positively influences overall HRQOL. The inverse correlation between migration percentage an HRQOL valiates the clinical assumption that chilren with greater hip isplacement experience a ecrease HRQOL because of limite mobility that complicates toileting, hygiene, transfers, an positioning. Following reconstructive hip surgery, hip stability an HRQOL improve an remaine stable over the ensuing 24 months. No patients unerwent revision surgery uring the stuy perio, emonstrating the short-term stability following reconstructive hip surgery in this population. One of the main goals of reconstructive hip surgery in this population of chilren with cerebral palsy is to provie a pain-free stable hip that allows for seating, transfers, an other activities of aily living. We recognize that the ultimate goal of hip reconstructive surgery is to provie a stable hip into aulthoo without requiring multiple surgical revisions. Our stuy is limite by its follow-up of 2 years, an measuring the longevity of the results of these reconstructive hip proceures is beyon the scope of this stuy. Further longituinal stuy is necessary to recor the longevity of the results of hip reconstructive proceures an measure the changes in HRQOL that occur uring the follow-up perio. Previous work by ung et al. emonstrate a negative association between migration percentage an HRQOL in a group of 34 chilren with GMFCS level-iii, IV, or V cerebral palsy, with a ecrease in the CPCHILD total score of point per 1% increment of migration percentage 25. However, ung et al. i not evaluate the ability of reconstructive hip surgery to moify the relationship between migration percentage an HRQOL. The nee for an timing of surgical intervention for hip ysplasia in non-ambulatory chilren with cerebral palsy have been extensively ebate. Hernon et al. performe a retrospective stuy of 32 chilren (48 hips) with severe cerebral palsy to etermine the appropriate timing of surgery to correct subluxate or islocate hips 17. They foun that the better the migration percentage of the hip preoperatively, the better the final long-term raiographic result, an therefore recommene early surgical intervention to prevent later islocation anegenerative changes 17. Dobson et al. similarly supporte routine screening to etect hip instability an early surgical intervention to prevent late hip islocation an more extensive salvage surgery, which were associate with poorer outcomes 27. Noonan et al. took a ifferent view 28. They evaluate a cohort of 77 institutionalize aults with cerebral palsy to etermine if spastic hip isplacement with osteoarthritis was associate with hip pain an/or ecrease function. They foun no association among raiographic finings of osteoarthritis, hip subluxation, hip islocation, an hip pain. Therefore, they recommene not performing prophylactic hip surgery an intervening only when a patient becomes symptomatic. Their finings, however, may have been inexact because their patients ha multiple caregivers, whose ability to accurately assess pain anysfunction was limite, an no stanarize HRQOL measure was use. Delaying surgery is supporte by the finings of Schmale et al. 29. Their retrospective stuy of reoperation rates an acetabular evelopment in chilren 6 years of age with cerebral palsy who unerwent only bilateral varus erotational osteotomy reveale that 74% of the patients require a reoperation at a mean of 5 years postoperatively. Because of limite evience of acetabular remoeling, the authors recommene postponing reconstructive surgery until the chil nears skeletal maturity. The results of our prospective stuy, which aresse the limitations of earlier work, support the results of previous authors. Similar to ung et al. 25, we foun that chilren with a preoperative migration percentage of 50% ha a significantly lower preoperative CPCHILD score compare with chilren whose preoperative migration percentage was <50% (Table III). We also foun that chilren with a preoperative migration percentage of <50% emonstrate higher postoperative CPCHILD scores than chilren with a preoperative migration percentage of 50%. These results echo those of Hernon et al. 17, who foun that patients with a better preoperative migration percentage ha a better postoperative result. While the final CPCHILD scores in our stuy i not iffer significantly between the 2 groups of chilren stratifie by migration percentage, we believe that our stuy was slightly unerpowere to answer this question concerning this outcome an that a significant ifference might have been ientifie with a larger sample size. The initial ecline in CPCHILD total score at 6 weeks postoperatively followe by progressive improvements over the ensuing 2 years was anticipateue to postoperative immobilization, muscle spasms, an healing surgical incisions that make caregiving challenging initially following surgery. The CPCHILD omain scores improve in all areas except communication an social interaction, which is not surprising as these activities are relatively unaffecte by reconstructive hip surgery.

8 1197 T HE OURNAL OF B ONE &OINT SURGERY BS. ORG The greatest improvement was note in the comfort an emotions omain. This fining is supporte by other researchers. It has been estimate that 25% to 89% of non-ambulatory chilren who have a islocate or subluxate hip emonstrate hip pain 11,13,16,30,31 an retrospective stuies have emonstrate improvements in pain following reconstructive surgery Detection bias may have been a factor in these stuies. We i not fin a correlation between acetabular eficiency an CPCHILD score before or after reconstructive hip surgery. Hägglun et al. note that lateral isplacement of the femoral hea often occurre without raiographic evience of acetabular ysplasia an that acetabular ysplasia was evient only at the later stages of hip lateralization 32. It was uncommon for acetabular ysplasia to occur inepenently of hip lateralization. Therefore, the authors etermine that the acetabular inex was not the best preictive raiographic metric for screening for spastic hip instability. The acetabular eficiency is most often posterolateral in neuromuscular spastic hip isplacement, which is less ientifiable on supine frontal plane raiographs owing to the ifficulties in accurately measuring the acetabular inex. Spencer an Sait suggeste that ynamic examination of the hip at the time of reconstructive hip surgery was most instrumental in efining the nee for pelvic acetabuloplasty 33. Although we initially hypothesize that GMFCS level, age, an comorbiities woul be significant preictors of the migration percentage an the CPCHILD scores, we i not fin significant correlations. However, the stuy was not powere to ientify these interactions. Although our stuy was prospective, it was performe at a single site an thus the results may not be applicable to all non-ambulatory chilren with cerebral palsy. Also, a control group of non-ambulatory chilren with cerebral palsy who ha not unergone hip surgery was not available for comparison. In accorance with institutional stanars, all at-risk patients were offere surgery, an few ecline. Since the cognitive an communication impairments inherent in this population limite their ability to self-report, caregiver perceptions were measure with the CPCHILD questionnaire. As with all proxy measures, the reporte scores may be overestimations or unerestimations an not completely reflecting the chil s HRQOL, especially if the caregivers ha unrealistic expectations. In conclusion, these ata provie support for the association between migration percentage an HRQOL an the effectiveness of reconstructive hip surgery in improving the overall HRQOL of non-ambulatory chilren with severe cerebral palsy. Although aitional research is neee, the migration percentage may be a raiographic proxy for functional limitations an pain associate with hip instability. We believe that these finings can be use by health-care proviers to guie patient eucation, set realistic expectations, an increase family participation in surgical ecision-making. In this era of health-care reform, orthopaeic surgeons are charge with emonstrating improvement in function after their surgical interventions. Reconstructive hip surgery for non-ambulatory chilren with cerebral palsy is costly, both financially an emotionally. The finings from this stuy provie evience that reconstructive hip surgery for the treatment of spastic hip instability reliably improves HRQOL. n Rachel DiFazio, PhD, RN, PPCNP-BC, FAAN 1 Benjamin Shore, MD, MPH, FRCSC 1 uith A. Vessey, PhD, MBA, RN, FAAN 2 Patricia E. Miller, MS 1 Brian D. Snyer, MD, PhD 1 1 Orthopeic Center, Boston Chilren s Hospital, Boston, Massachusetts 2 William F. Connell School of Nursing, Boston College, Chestnut Hill, Massachusetts aress for R. DiFazio: Rachel.Difazio@chilrens.harvar.eu References 1. Palisano R, Rosenbaum P, Walter S, Russell D, Woo E, Galuppi B. Development an reliability of a system to classify gross motor function in chilren with cerebral palsy. Dev Me Chil Neurol Apr;39(4): Rosenbaum PL, Palisano R, Bartlett D, Galuppi BE, Russell D. Development of the Gross Motor Function Classification System for cerebral palsy. Dev Me Chil Neurol Apr;50(4): Epub 2008 Mar Woo E, Rosenbaum P. The gross motor function classification system for cerebral palsy: a stuy of reliability an stability over time. Dev Me Chil Neurol May;42(5): Palisano R, Rosenbaum P, Bartlett D, Livingston MH. Content valiity of the expane an revise Gross Motor Function Classification System. Dev Me Chil Neurol Oct;50(10): Green LB, Hurvitz EA. Cerebral palsy [vii.]. Phys Me Rehabil Clin N Am Nov;18(4):859-82: vii. 6. Soo B, Howar, Boy RN, Rei SM, Lanigan A, Wolfe R, Reihough D, Graham HK. Hip isplacement in cerebral palsy. Bone oint Surg Am an;88(1): Robin, Graham HK, Baker R, Selber P, Simpson P, Symons S, Thomason P. A classification system for hip isease in cerebral palsy. Dev Me Chil Neurol Mar;51(3): Epub 2008 Dec Lonstein E, Beck K. Hip islocation an subluxation in cerebral palsy. Peiatr Orthop Sep-Oct;6(5): Reimers. The stability of the hip in chilren. A raiological stuy of the results of muscle surgery in cerebral palsy. Acta Orthop Scan Suppl. 1980;184(S184): Pons C, Rémy-Néris O, Méée B, Brochar S. Valiity an reliability of raiological methos to assess proximal hip geometry in chilren with cerebral palsy: a systematic review. Dev Me Chil Neurol Dec;55(12): Epub 2013 un Bagg MR, Farber, Miller F. Long-term follow-up of hip subluxation in cerebral palsy patients. Peiatr Orthop an-feb;13(1): Tönnis D. Normal values of the hip joint for the evaluation of X-rays in chilren an aults. Clin Orthop Relat Res Sep;(119): Brunner R, Picar C, Robb. Morphology of the acetabulum in hip islocations cause by cerebral palsy. Peiatr Orthop B ul;6(3): Murray AW, Robb E. The hip in cerebral palsy. Curr Orthop. 2006;20(4): Valencia FG. Management of hip eformities in cerebral palsy. Orthop Clin North Am Oct;41(4): óźwiak M, Harasymczuk P, Koch A, Kotwicki T. Incience an risk factors of hip joint pain in chilren with severe cerebral palsy. Disabil Rehabil. 2011;33(15-16): Epub 2010 Nov Hernon WA, Bolano L, Sullivan A. Hip stabilization in severely involve cerebral palsy patients. Peiatr Orthop an;12(1):68-73.

9 1198 T HE OURNAL OF B ONE &OINT SURGERY BS. ORG 18. Krebs A, Strobl WM, Grill F. Neurogenic hip islocation in cerebral palsy: quality of life an results after hip reconstruction. Chil Orthop Mar;2(2): Epub 2008 Feb Miller F, Girari H, Lipton G, Ponzio R, Klaumann M, Dabney KW. Reconstruction of the ysplastic spastic hip with peri-ilial pelvic an femoral osteotomy followe by immeiate mobilization. Peiatr Orthop Sep-Oct;17(5): Zarrinkalam R, Rice, Brook P, Russo RN. Hip isplacement an overall function in severe cerebral palsy. Peiatr Rehabil Me. 2011;4(3): Taylor RM, Gibson F, Franck LS. A concept analysis of health-relate quality of life in young people with chronic illness. Clin Nurs ul;17(14): Berzon R. Unerstaning an using health relate quality of life instruments within clinical research stuies. In: Stequet M, Hays R, Fayers P, eitors. Quality of life assessment in clinical trials. New York: Oxfor University Press; p Narayanan UG, Fehlings D, Weir S, Knights S, Kiran S, Campbell K. Initial evelopment an valiation of the Caregiver Priorities an Chil Health Inex of Life with Disabilities (CPCHILD). Dev Me Chil Neurol Oct;48(10): Flynn M, Miller F. Management of hip isorers in patients with cerebral palsy. Am Aca Orthop Surg May-un;10(3): ung NH, Pereira B, Nehring I, Brix O, Bernius P, Schroeer SA, Kluger G, Koehler T, Beyerlein A, Weir S, von Kries R, Narayanan UG, Berweck S, Mall V. Does hip isplacement influence health-relate quality of life in chilren with cerebral palsy? Dev Neurorehabil Dec;17(6): Epub 2014 ul Difazio RL, Vessey A, Zurakowski D, Snyer BD. Differences in health-relate quality of life an caregiver buren after hip an spine surgery in non-ambulatory chilren with severe cerebral palsy. Dev Me Chil Neurol Mar;58(3): Epub 2015 Aug Dobson F, Boy RN, Parrott, Nattrass GR, Graham HK. Hip surveillance in chilren with cerebral palsy. Impact on the surgical management of spastic hip isease. Bone oint Surg Br ul;84(5): Noonan K, ones, Pierson, Honkamp N, Leverson G. Hip function in aults with severe cerebral palsy. Bone oint Surg Am Dec;86-A(12): Schmale GA, Eilert RE, Chang F, Seiel K. High reoperation rates after early treatment of the subluxating hip in chilren with spastic cerebral palsy. Peiatr Orthop Sep-Oct;26(5): Cooperman DR, Bartucci E, Dietrick E, Millar EA. Hip islocation in spastic cerebral palsy: long-term consequences. Peiatr Orthop May-un;7(3): Grzegorzewski A, óźwiakm, PawlakM, MorzewskiT, BuchcicP, Mas1ońA. Hip joint pain in chilren with cerebral palsy anevelopmental ysplasia of the hip: why are the ifferences so huge? BMC Musculoskelet Disor. 2014;15(1):96. Epub 2014 Mar Hägglun G, Lauge-Peersen H, Persson M. Raiographic threshol values for hip screening in cerebral palsy. Chil Orthop Mar;1(1):43-7. Epub 2007 Feb Spencer D, Sait MS. The true acetabular inex in chilren with cerebral palsy. Ann R Coll Surg Engl Sep;86(5):371-4.