GUIDANCE study: guided growth of the proximal femur to prevent further hip migration in patients with cerebral palsy—study protocol for a multicentre randomised controlled trial Authors: Renée Anne van Stralen, Merel Charlotte Rosalie Roelen, Sophie Moerman, Melinda Maria Eva Helena Witbreuk, M Adhiambo Witlox, Arno ten Ham, Denise Eygendaal, Max Reijman, Jaap Johannes Tolk Published: 2024 ABSTRACT AbstractIntroductionUp to one-third of patients with cerebral palsy (CP) develop hip migration. Current standard care for early hip migration is bilateral adductor-psoas tenotomy; however, the failure rate is relatively high with 34%–74% of patients with CP requiring secondary hip surgery. Using temporary medial hemiepiphysiodesis of the proximal femur (TMH-PF), the morphology of the hip can be changed. This technique aims to reduce further hip migration and the need for secondary surgical management. Further research is necessary to determine the benefit of TMH-PF in addition to adductor-psoas tenotomy. The hypothesis of this study is that TMH-PF combined with adductor-psoas release decreases the chance of progressive hip migration and the need for secondary hip surgery, compared with adductor-psoas release alone.Methods and analysisThe GUIDANCE study is an open-label multicentre randomised controlled trial. Patients with CP aged between 2 and 8 years, with spastic CP—Gross Motor Function Classification System IV or V, hip abduction ≤40° and hip migration of 30%–50% can be included in this trial. They will be randomised into a control arm (adductor-psoas tenotomy) or an intervention arm (adductor-psoas tenotomy+TMH PH). The primary outcome will be treatment failure at 5-year follow-up. At 2-year follow-up a preliminary analysis will be performed. Secondary outcomes will be differences in patient-reported outcome measures (CPCHILD and CPG pain score), range of motion, radiological measurements including head shaft angle and hip migration percentage and three-dimensional (3D) morphological changes to the proximal femur. Furthermore, an analysis will be performed to identify predictors for treatment failure in both treatment arms.Ethics and disseminationThe GUIDANCE study should provide evidence on the effectiveness of TMH-PF in addition to adductor-psoas tenotomy in children with CP with early hip migration. If beneficial, larger hip reconstructive procedures can be delayed or prevented, providing a distinct benefit for these vulnerable children. The study’s strengths lie in its methodological framework, incorporating randomised allocation and intervention assessment. The main limitation is the inability to blind the treating physician or the researcher for the treatment arm the participant is allocated to. The results of the GUIDANCE study will be presented at scientific meetings and published in international peer-reviewed journals. The aim is to publish the results at 2 years follow-up and 5 years follow-up and to publish the results of the analysis on the 3D morphology of the hip after TMH-PF. Individual de-identified participant data that underlie the results from the GUIDANCE study and the study protocol will be shared if requested.Trial registration numberClinical Trial Registry number: NCT06118736. Registered on 3 November 2023. STRENGTHS AND LIMITATIONS OF THIS STUDY This study is a multicentre randomised controlled trial concerning hip migration in patients with cerebral palsy (CP). This study is a multicentre randomised controlled trial concerning hip migration in patients with cerebral palsy (CP). The GUIDANCE study will be conducted by all healthcare practitioners in Holland providing orthopaedic care for patients with CP, eliminating bias due to treatment location. The GUIDANCE study will be conducted by all healthcare practitioners in Holland providing orthopaedic care for patients with CP, eliminating bias due to treatment location. The standardisation of a protocol for the treatment of hip migration ensures consistency in intervention and assessment; this will provide more robust and valid outcomes. The standardisation of a protocol for the treatment of hip migration ensures consistency in intervention and assessment; this will provide more robust and valid outcomes. There is no option of blinding the researcher and the orthopaedic surgeon to which treatment modality the patient received, which could impact radiographic measurements and the decision for the need for additional surgery. There is no option of blinding the researcher and the orthopaedic surgeon to which treatment modality the patient received, which could impact radiographic measurements and the decision for the need for additional surgery. Introduction Coxa valga is a complex three-dimensional (3D) deformity of the proximal femur caused by altered growth of the proximal femoral physis.13 In patients with cerebral palsy (CP) coxa valga is commonly related to the functional status of the patient, in combination with muscle spasticity and weakness. Coxa valga, contractures and diminished weight bearing can lead to progressive hip migration and subsequent symptomatic dislocation. This can lead to disturbed seating balance or standing abilities, difficulty with perineal care, the development of decubitus ulcers and poor quality of life.49 Up to one-third of children with CP have hip displacement, defined as a migration percentage (MP) of >30%.10 There is a higher incidence reported for patients with an increased Gross Motor function classification system (GMFCS) level. As much as 69.2% of patients classified as GMFCS IV and 89.7% of patients classified as GMFCS V are diagnosed with hip migration,1014 as measured using the Reimers MP.15 The traditional approach to managing hip migration in patients with CP is hip reconstruction consisting of adductor-psoas tenotomy combined with femoral and pelvic osteotomies. This approach is usually reserved for hips with an MP of 40% or more.16 These procedures are associated with significant perioperative morbidity, including pain, significant blood loss, lengthy anaesthetic and functional recovery times.17 Furthermore, these procedures have a relatively high complication rate.18 19 In children under the age of 8 years with an MP of over 30% and abduction less than 30 degrees, an adductor-psoas tenotomy can be considered as an alternative to early bony hip reconstruction. However, the failure rate (defined as progression of migration to >50% or the need for secondary bony surgery) for adductor-psoas tenotomies is relatively high with 34%–74% of patients requiring additional surgery.2024 Guided growth procedures aim to surgically manipulate the growth plate with the goal of correcting deformities. These procedures are often minimally invasive and are well established in the treatment for the gradual correction of angular limb deformities in children.2528 Recent reports have shown that guided growth can also be used to influence growth at the proximal femur in children aged 2–8 years.2935 By placing a screw over the physis on the medial side of the proximal femoral epiphysis, a tether is created that will result in progressive varus of the proximal femur during growth (figure 1). It is understood that this manipulation of the proximal femoral anatomy can alter the course of hip migration and subsequent secondary acetabular dysplasia without performing invasive procedures such as femoral osteotomies.36 37 Furthermore, it is recognised that guided growth procedures of the proximal femoral physis can be carried out as day case procedures, require a shorter operating time than bony reconstructions and allow for immediate weight bearing.29 32 Temporary medial hemiepiphysiodesis of the proximal femur. (a) is the preoperative pelvic radiograph of a 5-year-old boy with cerebral palsy (classified as GMFCS IV). (b) shows the intraoperative image of the right hip, with the arthrogram determining screw position, ruling out screw protrusion and providing information on acetabular morphology. Similarly, (c) depicts the left hip intraoperatively. (d) is the first postoperative pelvic radiograph showing good screw position. GMFCS, Gross Motor Function Classification System. A recent systematic review on this topic32 concludes that this new and minimally invasive procedure has shown to be safe and effective in the modulation of the proximal physis to correct coxa valga deformities. This can prevent progressive subluxation of the hip joint and may prevent the requirement for complex open-hip reconstruction surgery in this vulnerable cohort. Depending on the treatment duration and patient age, physis growing off the screw is a common complication. Patients and caregivers need to be counselled that screw revision can be necessary in about 50% of cases. Invasive pelvic reconstructions and femoral osteotomies may be needed in only 5%–21% of patients initially treated with guided growth.32 It has to be noted that these findings have been only reported in small cohort studies without a control group and with relatively short-term follow-up.3035 Despite these promising preliminary reports, further research is required to assess whether proximal femur-guided growth decreases the risk of further hip migration and the need for further surgery in patients with CP, compared with the current standard of care, which is an adductor-psoas tenotomy. Objectives The primary objective is to determine whether a medial hemiepiphysiodesis of the proximal femur combined with surgical adductor-psoas release decreases the chance of treatment failure, defined as the need for secondary bony hip surgery to the affected hip or progression of hip migration to >50%,20 22 23 38 at 5-year follow-up, compared with adductor-psoas release only in patients with CP at risk for progressive hip migration. The secondary objectives are to evaluate postoperative changes in all participants, following both isolated adductor-psoas tenotomy and adductor-psoas tenotomy combined with TMH-PF. These include but are not limited to, between-group differences in assessing patient-reported outcome measure (PROM), complications, tone regulation, range of motion (ROM) and time-to-failure. These also include changes in radiological assessments like the head shaft angle (HSA), acetabular index (AI) and hip MP, as well as 3D morphological alterations to the proximal femur. Furthermore, analyses will be performed to identify predictors for treatment failure and predictors for (change in) secondary outcome parameters. Methods and analysis This manuscript is written according to the Consolidated Standards for Reporting Trials statement and Standard Protocol Items: Recommendations for Interventional Trials guidelines. Study design and setting This study will be an open-label multicentre randomised controlled trial. All participating hospitals are involved in the orthopaedic surgical care for children with CP in the Netherlands and include the following centres: Amsterdam University Medical Centre (Amsterdam), University Medical Centre Groningen (UMCG), Maastricht University Medical Centre+ (Maastricht), Sint Maartenskliniek (Nijmegen) and Erasmus University Medical Centre (Rotterdam). As the current evidence on the results of TMH-PF is insufficient to clearly define the best treatment for early hip migration in patients with CP, a randomised controlled trial is indicated to determine the superiority of TMH-PF and adductor-psoas tenotomy over adductor-psoas tenotomy to prevent further hip migration and/or the need or bony hip surgery. The results of this study can clarify and confirm the potential added benefit of guided growth of the proximal femur in treating hip migration in patients with CP. Study population Participants Patients eligible for this trial are children with spastic CP aged between 2 and 8 years old. Their functional classification should be GMFCS level IV-V; Children with CP—GMFCS IV and V are non-ambulatory. Children with CP—GMFCS level V lack independence in basic antigravity postural control. Children with CP—GMFCS IV have some self-mobility with limitations. Children are transported or use power mobility outdoors and in the community.39 Inclusion criteria To be eligible to participate in this study, a subject must meet all of the following criteria: Spastic CP Spastic CP GMFCS level IV-V39 GMFCS level IV-V39 Aged 2–8 years Aged 2–8 years Abduction in flexion ≤40° Abduction in flexion ≤40° MP of 30%–50%15 MP of 30%–50%15 HSA > 14540 41 HSA > 14540 41 Exclusion criteria A potential subject who meets any of the following criteria will be excluded from participation in this study: Patients considered not fit for surgery. Patients considered not fit for surgery. History of bony hip surgery to the affected hip. History of bony hip surgery to the affected hip. Severe acetabular dysplasia defined as a gothic arch, an incongruent joint or an AI > 30°, consistent with A2 and A3 acetabular deformity according to Robin and Graham.42 Severe acetabular dysplasia defined as a gothic arch, an incongruent joint or an AI > 30°, consistent with A2 and A3 acetabular deformity according to Robin and Graham.42 Recruitment After being considered eligible by their orthopaedic surgeon, patients will be screened by a researcher. When inclusion and exclusion criteria are met, the patients and their parents will be asked to participate in the study. Written information on the study objective and procedures will be provided and patients and their parents will have the possibility to ask additional questions when necessary. The patient and their parents will be informed that two surgical procedures can be performed and will receive information about the two surgical treatment options available for the patient’s condition and that this study aims to compare the effectiveness of both interventions. During this process, patients and their parents are blinded to the hypothesis of the study to avoid bias and patient preference for one of the two surgeries. If a patient and their parents are willing to participate an informed consent form will be signed by the participants’ parents and baseline measurements registered. Patient and public involvement A diverse group of healthcare experts, patients and caregivers collaborated to identify key research priorities for children needing elective surgery on the lower limbs. One of their top 10 priorities is to establish optimal management strategies for hip displacement in children with CP.43 The GUIDANCE study aims to contribute towards answering this question. Due to the technical nature of the comparison of two surgical techniques in the GUIDANCE study, patients were not involved in developing the study protocol. Since the patients’ and caregivers’ experience and perspective are essential, PROMs are incorporated as a secondary outcome in this study. Randomisation The randomisation procedure will be performed after informed consent has been obtained. Due to the nature of our intervention, it is not possible to blind the participating orthopaedic surgeons performing the surgery, as they are performing the procedure. The orthopaedic surgeon will be in charge of determining the need for additional surgery and therefore classifying the treatment result as a success or a failure. Because the implants are clearly visible in the radiographs, it is impossible to blind the researcher performing the radiographic measurements for this study. Allocation to type of intervention takes place by receiving the consecutive randomisation number and allocation from the coordinating investigator. A computer-generated randomization list will be used (CASTOR EDC V.2023.4.5.0, Amsterdam, The Netherlands). Patients can have one or two hips that fit the inclusion criteria. All eligible hips will be included but randomisation will be performed on a patient level; if a patient has two hips that fit the inclusion criteria, they will both be allocated to the same treatment arm therefore block randomisation with the patient as a block will be performed. We will account for this in the analysis. Patients with bilateral adductor contractures and bilateral coxa valga will undergo bilateral surgery, even if the MP is less than 30%. Only the hips with an MP between 30% and 50% will be included in the study. Pelvic obliquity will not affect the selection. Furthermore, randomisation will be stratified according to the treatment centre and age group (groups 2–5 and 6–8 years of age). The division in the age group is made because different rates of correction can be expected in different age groups with corresponding growth rates. We will use computerised block randomisation with variable block sizes; block sizes 2 to 4. Study procedures and timeline Outcome measurements will take place at different time points as shown in online supplemental table 1 with a total of 10. On the first visit, an eligibility screen will be performed and baseline characteristics will be collected. Patients will receive questionnaires throughout the study. Radiographical imaging and physical examination performed by orthopaedic surgeons will also be performed throughout the study. Complications will be collected at eight time points and reported using the modified Clavien–Dindo system.44 Interventions Theoretical framework for using TMH-PF in hip migration The current standard of treatment in younger children with early hip migration is to perform an adductor-psoas tenotomy.45 The available evidence is heterogeneous regarding the results of isolated soft tissue releases.2024 Treatment success is generally defined as no need for secondary procedures bony surgery to the affected hip and/or no progression of migration beyond 50%.15 The reported rates of successful outcomes vary from 23% to 85%.2024 Reasons for this large variation lie at least partly in differences in population, definition of outcome parameters, length of follow-up and overall organisation of care for patients with CP.24 46 In recent literature, the potential of guided growth of the proximal femur to modify hip migration in patients with CP has been shown. Using TMH-PF the morphology of hips at risk of symptomatic (sub)luxation in CP can be changed, aiming to reduce further hip migration and the need for more invasive surgical treatment modalities.303547 Intervention arm The intervention under investigation is a guided growth procedure of the proximal femur combined with an adductor-psoas tenotomy. The TMH-PH will be performed using a 4.5 or a 6.5/8.0 mm fully threaded cannulated screw system as described by Davids.29 The adductor-psoas tenotomy will be performed through an open procedure. It consists of a tenotomy of the psoas tendon, the adductor longus and brevis and/or gracilis (until abduction is at least 70°). Several papers have been published describing the results of this specific surgical technique.3031 3335 Portinaro et al described the results for their cohort that underwent adductor tenotomies in combination with TMH-PF. They reported treatment failure in 3 of 56 hips (5.3%) at 5-year follow-up.34 Zakrzewski also described a similar surgical intervention in a similar patient population. They evaluated the results of 13 patients who underwent guided growth alone, and 12 patients who underwent a combination of guided growth and adductor tenotomy. Two out of 32 hips (6.25%) had a migration percentage over 50% at the final follow-up.35 Based on these reports, we estimated the risk of treatment failure (defined as the need for a secondary bony procedure of progression of hip migration beyond 50%, at a 5-year follow-up) in the adductor-psoas tenotomy combined with the guided growth treatment arm at 20%. Control arm The control group in this study consists of patients who only undergo an adductor-psoas tenotomy. The adductor-psoas tenotomy will be performed through an open procedure. It consists of a tenotomy of the psoas tendon, the adductor longus and brevis and/or gracilis (until abduction is at least 70°). Results from the Swedish National Register indicate that in a patient population of which 90% are classified as GMFCS IV-V, 43% of hips treated by adductor and psoas tendon release will undergo further surgery (5-year follow-up, mean age 4.9 years, mean MP 47%).20 In another study in a similar population with slightly younger age (mean age 3.9 years, mean follow-up 5 years), a reoperation rate per patient of 77% was reported.21 Shore et al published a mean failure rate of 68% (mean age 4.2 years, follow-up 7 years).22 Higher GMFCS level is known to be associated with a higher risk of treatment failure.22 Five authors describe failure rates ranging from 35% to 81% in patients with CP—GMFCS IV/V.2024 Heterogeneity regarding indication, timing and age of patients are for a large part responsible for the variation in reported treatment success. Based on these reports, the risk of treatment failure in the adductor-psoas tenotomy treatment arm is estimated at 50%. Treatment failure is defined as the need for secondary bony surgery and/or progression of hip migration beyond 50%.2024 Secondary interventions Both groups will be allowed immediate weight bearing and supported standing. They will undergo the same postoperative physiotherapy protocol aimed at maintaining ROM. This is according to routine clinical care for both treatment arms and will consist of prone lying for 20 min three times a day. They will undergo manual passive adductor stretching by their physiotherapist two to three times a week. A removable abduction splint (broomsticks or swash) is fitted in maximum comfortable abduction. Participants will wear this orthosis during the day and night outside of the physiotherapy sessions, transport and personal hygiene. Another secondary intervention that can be performed in both groups is repeated soft tissue release. The indications for repeated soft tissue releases (adductor-psoas tenotomy) are abduction <3° and no indication for bony surgery. Another possible secondary intervention is screw revision. This can only be performed in the intervention arm. The indication for screw revision is <1 thread of the screw remaining in the epiphysis and HSA >100°. Outcomes Primary outcome measure The primary outcome is treatment failure. This will be defined as: need for secondary bony hip surgery to the affected hip OR progression to hip migration >50%, at a 5-year follow-up. The indication for secondary bony surgery has been discussed during a consensus meeting with all participating orthopaedic surgeons. The indication will be defined as: MP >50% at a minimum of 1 year after surgery, an MP increase of >10% in 1 year during follow-up or an increase in AI to >30° or an increase more than 5° during follow-up. Participant characteristics At baseline, several participant characteristics will be obtained: age, gender, weight, height, GMFCS classification, MP, head-shaft angle, AI and use of tone regulation (Baclofen, Trihexyfenidyl (Artane), Baclofen pump, Selective Dorsal Rhizotomy, for example). Secondary outcome measures Radiographic imaging Standardised conventional radiographs, anteroposterior (AP) and frog leg lateral pelvic radiographs will be performed according to the standardised protocol as defined by the American Academy for Cerebral Palsy and Developmental Medicine (AACPDM).48 On these images, the following measurements will be performed: MP (%)This measurement calculates the percentage of the femoral head covered by the acetabulum. This is used to determine how far the femoral head has migrated out of acetabular socket and is a measure to indicate hip (in)stability.15 MP (%) This measurement calculates the percentage of the femoral head covered by the acetabulum. This is used to determine how far the femoral head has migrated out of acetabular socket and is a measure to indicate hip (in)stability.15 This measurement calculates the percentage of the femoral head covered by the acetabulum. This is used to determine how far the femoral head has migrated out of acetabular socket and is a measure to indicate hip (in)stability.15 HSA in degreesThis measurement is used to assess the orientation of the femoral head relative to the femoral shaft. There is a correlation between higher values and hip displacement.40 41 HSA in degrees This measurement is used to assess the orientation of the femoral head relative to the femoral shaft. There is a correlation between higher values and hip displacement.40 41 This measurement is used to assess the orientation of the femoral head relative to the femoral shaft. There is a correlation between higher values and hip displacement.40 41 AI in degreesThis measurement reflects the inclination of the acetabulum and is important in assessing hip joint development.49 AI in degrees This measurement reflects the inclination of the acetabulum and is important in assessing hip joint development.49 This measurement reflects the inclination of the acetabulum and is important in assessing hip joint development.49 Melbourne hip classification42 Melbourne hip classification42 Pelvic obliquity50 Pelvic obliquity50 Positioning of the hemiepiphysiodesis screw in the intervention group31 Positioning of the hemiepiphysiodesis screw in the intervention group31 Mose circle51 and alpha angle52 to assess femoral head sphericity Mose circle51 and alpha angle52 to assess femoral head sphericity The MP has demonstrated excellent reliability across multiple studies, with intrarater and interrater intraclass correlation coefficients all exceeding 0.93.53 54 Similarly, the AI has exhibited good reliability, with intrarater and interrater intra-class correlation coefficients (ICCs) of 0.84 and 0.79, respectively.54 Furthermore, the head-shaft angle has shown excellent interrater and intrarater reliability.55 In summary, the MP, AI, and head-shaft angle are reliable metrics for monitoring hip migration in children with CP. CT-scans of the hip will be performed in the intervention group only. On these scans, the following measurements will be performed: Positioning of the hemiepiphysiodesis screwCheck for protrusion of the screw into the joint and number of threads across the physis.31 56 57Assess positioning of the screw in the physis as a percentage in both the sagittal and coronal plane.31Screw position is calculated by using the percentage of the physis that lies medially to screw on AP radiographs and by using the percentage of physis that lies posteriorly to the screw on Lauenstein radiographs, with a range from 0 to 100 (with 0 being medially/posteriorly and 100 being laterally/anteriorly). Based on these two measurements, the distance of the screw to the centre of the physis and the location where the screw penetrates the physis can be determined. Positioning of the hemiepiphysiodesis screw Check for protrusion of the screw into the joint and number of threads across the physis.31 56 57 Check for protrusion of the screw into the joint and number of threads across the physis.31 56 57 Assess positioning of the screw in the physis as a percentage in both the sagittal and coronal plane.31Screw position is calculated by using the percentage of the physis that lies medially to screw on AP radiographs and by using the percentage of physis that lies posteriorly to the screw on Lauenstein radiographs, with a range from 0 to 100 (with 0 being medially/posteriorly and 100 being laterally/anteriorly). Based on these two measurements, the distance of the screw to the centre of the physis and the location where the screw penetrates the physis can be determined. Assess positioning of the screw in the physis as a percentage in both the sagittal and coronal plane.31 Screw position is calculated by using the percentage of the physis that lies medially to screw on AP radiographs and by using the percentage of physis that lies posteriorly to the screw on Lauenstein radiographs, with a range from 0 to 100 (with 0 being medially/posteriorly and 100 being laterally/anteriorly). Based on these two measurements, the distance of the screw to the centre of the physis and the location where the screw penetrates the physis can be determined. Screw position is calculated by using the percentage of the physis that lies medially to screw on AP radiographs and by using the percentage of physis that lies posteriorly to the screw on Lauenstein radiographs, with a range from 0 to 100 (with 0 being medially/posteriorly and 100 being laterally/anteriorly). Based on these two measurements, the distance of the screw to the centre of the physis and the location where the screw penetrates the physis can be determined. 3D proximal femoral shape analysis regarding (change in) sphericity, congruency and head–neck orientation. 3D proximal femoral shape analysis regarding (change in) sphericity, congruency and head–neck orientation. The direction of migration of the hip and acetabular morphology will be determined according to the method described by Brunner et al58 The direction of migration of the hip and acetabular morphology will be determined according to the method described by Brunner et al58 Sphericity of the femoral head will be assessed. Sphericity of the femoral head will be assessed. The CT protocol for the hip was performed without the use of contrast, using the following parameters: a collimation of 144 mm × 0.40 mm, Manual CareDose4D and a Care keV IQ level set at 80 keV, optimised specifically for bone and calcium imaging. The scan was conducted with a tube voltage of 120 kV, a reference mAs of 66, a pitch of 1.50 and a rotation time of 0.25 s. Images were reconstructed in the axial, coronal and sagittal planes. The axial reformat was performed with slice thicknesses of 1 mm and 0.4 mm, while both coronal and sagittal reformats were created with a slice thickness of 0.8 mm. All reformats employed a Br72 kernel and Iterative Reconstruction at level Q3. Additionally, 3D volume rendering with a slice thickness of 0.8 mm was also generated. On average, patients will receive a radiation dose of approximately 0.1, 0.2, and 0.3 mSv due to the CT scans at the age of roughly 5, 7 and 10 years, respectively. In total, the estimated radiation dose will be 0.6 mSv All radiographic measurements will be performed by the surgeon and the independent researcher. Clinical parameters A physical examination will be performed with the patient supine. ROM will be assessed and abduction in flexion will be evaluated with the hips in 90° of flexion. Questionnaires Dutch translation of the CP—Child questionnaire Parents will be asked to fill out the CP—Child.59 60 The questionnaire takes approximately 20 to 30 min to complete. The CPCHILD is a parent-completed questionnaire and was developed to help clinicians identify areas that are impairing a child’s quality of life, determine what is most important to the child and caregiver and monitor the child’s progress in a way that is meaningful to the family. The CPCHILD consists of 37 items distributed among seven sections. Each item is rated on a 6- or 7-point scale, and for some items, in addition, a 4-point level of assistance modifier. Transformed scores from 0 (worst) to 100 (best) are calculated for a total CPCHILD score, as well as for each of the six sub-scales or domains.59 60 CPG pain score (Checklist Pijn Gedrag, CPG versie2.5 NL Boerlage, Van Dijk, Erasmus MC 2010) A CPG pain score is based on 10 specific points of observation on the patients’ behaviour and then these observations are translated to a score ranging between 0 and 10. When a CPG score of 5 or higher is documented, the patient could potentially be in pain.61 Secondary surgical interventions other than bony procedures The number of screw revisions and repeated soft tissue releases, including indications and timing will be recorded. Sample size and power calculations We assessed the required sample size to answer our primary research question. Based on previous reports we hypothesised a failure rate of 50% for adductor-psoas tenotomy (group A) and a failure rate of 20% for TMH-PF (group B). To determine the correct sample size for this research, three different sample sizes: 70, 80 and 90, and three different ICC values: 0.6, 0.7 and 0.8, were taken into account. For each setting, we simulated 500 datasets from a mixed-effects logistic regression model with a random intercept term. The variance of the random intercepts, sigma_bˆ2, was obtained from the ICC using the equation sigma_bˆ2 = (ICC × piˆ2/3)/(1−ICC). The model’s coefficients were set to achieve a marginal probability for group A equal to 0.5 (50% failure rate) and for group B equal to 0.2 (20% failure rate). Each simulated dataset was analysed with a random intercept mixed effects logistic regression, and the estimated coefficients were transformed to the corresponding coefficients with a marginal interpretation. The proportion of times the coefficient for the log OR of B versus A was statistically significant at the level of 5% was the estimated power for the specific design setting. Statistical analysis General descriptive statistics will be performed on participant characteristics and baseline measurements. Distribution of all variables will be tested by the Shapiro–Wilk test. For the normally distributed variables, parametric tests will be used. For not normally distributed variables, non-parametric tests will be used. Participants will be analysed according to the intention-to-treat (as randomised) principle following the original group allocation. A secondary as-treated analysis will also be performed. Primary outcome The primary outcome, treatment failure, will be analysed after 5 years follow-up. After 2 years follow-up a preliminary analysis will be performed. The analysis that will be used is a Generalised Estimating Equations approach to account for the clustering of both hips per participant. The working correlation matrix will be exchangeable and the link function the logit. The dependent variable will be treatment failure. Variables that are known to be associated with hip migration, based on previous studies or based on a strong clinical rationale, will be considered as covariates. The considered covariates are ROM, GMFCS classification, age, head-shaft angle, AI and tone management. Statistical significance is set at p = 0.05. Secondary outcomes For the secondary outcomes, the between-group differences in change between baseline and follow-up time points will be analysed using linear mixed-effect models. This accounts for radiographic parameters, such as MP, head-shaft angle and 3D shape of the proximal femur, age, PROMs, such as CPCHILD and CPG pain score, ROM and complication rates according to the modified Clavien–Dindo system.44 A similar regression analysis will be performed to identify predictors for treatment failure, (change in) secondary outcome measures and changes in 2D and 3D hip morphology. Time to failure will be analysed using the stratified cox proportional hazards model. Linear mixed effect models using failure as an outcome will also be performed on a subgroup, namely <4 years old or >4 years old. Statistical analyses will be done using R software. Data management All data will be handled confidentially and anonymised in compliance with the Dutch Personal Data Protection Act (‘Wet Bescherming Persoonsgegevens’). Questionnaires are collected digitally and the patient study data will be stored and coded using data management software (CASTOR EDC V. 2023.4.5.0, Amsterdam, The Netherlands). Each participant receives an anonymized study number used for all documentation, study reports and publications. The key for this study number will be handled by an independent researcher. All data will be stored during the study period, and when the study is finished, the research files are stored for 15 years at the Erasmus MC research archive. Data monitoring The Medical Research Ethics Committee deemed the study as a ‘low-risk’ study. Therefore, no Data Monitoring and Safety Committee was recommended at the time of ethical approval. However, the study will be monitored by an independent monitoring board. The Medical Research Ethics Committee will be informed yearly on the inclusion rate, adverse events and study results. Ethics and dissemination The GUIDANCE study is the first randomised controlled trial to evaluate the effectiveness of TMH-PF in addition to an adductor-psoas tenotomy in non-ambulant children with CP. Recent studies indicate positive outcomes, report few complications and show a decrease in MP.3035 These findings suggest a potential reduction in treatment failure, potentially leading to fewer secondary surgeries and decreased morbidity. The study’s strengths lie in its methodological framework, incorporating randomised allocation and intervention assessment. This is particularly beneficial given the reliance of prior research on retrospective cohort designs, which are susceptible to selection bias. Such bias can hold significant implications within this specific patient population. Additionally, the multicentre approach involving all hospitals involved in orthopaedic care for patients with CP in the Netherlands, along with a consensus on outcome measures, enhances the study’s applicability and generalizability. Furthermore, this technique is applicable in other countries as well and could be incorporated into other surveillance guidelines. This robust methodology not only supports the reliability of the findings but also provides a solid foundation for future research, making it particularly valuable for further investigations in the Netherlands. This study design has its limitations. First, the treating surgeon and the researcher cannot be blinded to the allocated treatment of the participant. The surgeon must be aware of the surgery performed and can clearly see the presence or absence of the screw on follow-up radiographs. The same applies to the researcher who is measuring the follow-up radiographs, as these images clearly indicate which treatment arm the participant was allocated to. Because the definition of failure has been clearly established and both the treating surgeon and the researcher will measure and evaluate radiographs, it is expected that this will not impact the overall primary outcome. Second, the subject under study has been prioritised by patient representatives but the research protocol as such was developed without. As outlined in the methods section, including the experiences and perspectives of participants and their caregivers is essential for this study, hence why PROMs are a valuable secondary outcome parameter. Ethical approval and consent to participate This study will be conducted according to the principles of the Declaration of Helsinki (64th version, October 2013) and was approved by the Medical Research Ethics Committee of Erasmus Medical Centre in Rotterdam (reference number: MEC-2023–0714). The trial is registered in the Clinical Trial Registry: number NCT06118736. In all participating hospitals, the local responsible authorities approved the conduct of the study. In case of important protocol modifications (eg, new or cancelling of a recruiting hospital, change in outcomes) this will be communicated towards the Medical Research Ethics Committee. The ethical approval process included a radiation exposure risk assessment (online supplemental file 2). Trial status Study inclusion is ongoing and has started in March 2024. Preliminary results are anticipated mid-2029 and final results are anticipated late-2032. Dissemination The results of the GUIDANCE study will be presented at scientific meetings and published in international peer-reviewed journals. The aim is to publish the results at 2 years follow-up and at 5 years follow-up and to publish the results of the analysis on the 3D morphology of the hip after TMH-PF. Summaries will be provided to the funders of the study as well as participants and their parents/guardians. Individual de-identified participant data that underlie the results from the GUIDANCE study and the study protocol will be shared if requested. Data will be available beginning 12 months and ending 5 years following publication of this paper. Data will be available for researchers who provide a methodologically sound scientific proposal, which has been approved by an ethical committee. Proof of the latter should be provided. Analyses should achieve the aims as reported in the approved proposal. Proposals for data should be directed to the corresponding author. supplementary material