Secondary Logo

Journal Logo


Abducted Standing in Children With Cerebral Palsy: Effects on Hip Development After 7 Years

Martinsson, Caroline MSc; Himmelmann, Kate MD, PhD

Author Information
doi: 10.1097/PEP.0000000000000789
  • Free


Children with cerebral palsy (CP) who do not walk have a high risk for hip displacement at a young age.1 The spastic muscle forces around the hip force the hip into an adducted, flexed, and inward rotated position, which causes the migration of the femoral head.2

The hip migration percentage (MP) describes the percentage of the femoral head positioned lateral to the clearly delineated lateral acetabular border and is obtained using an anteroposterior pelvic radiograph.3 Children developing typically seldom have an MP exceeding 5%. However, some children with CP exhibit the “gothic arch” deformity,4 which is caused by pressure of the femoral head against the acetabular rim and can affect acetabular border growth. This deformity causes an imprint on the femoral head and, eventually, instability, which often progresses to hip dislocation.

One aim of the management of children with CP is to maintain well-located and pain-free hips.1 This can be achieved through nonsurgical means (eg, by positioning the child in a lying, sitting, or standing position at all times to prevent adduction) and/or by surgery.5 Bilateral adductor-psoas tenotomy (APT) is often recommended when the MP exceeds 40% to 50%.1 However, if this does not decrease the MP, reconstruction involving a varus derotation osteotomy is often performed, sometimes in combination with a pelvic osteotomy.6

In Sweden, almost every child with CP who does not walk performs passive upright standing in a stander or standing shell in a controlled position on a daily basis.7 Standing can be divided in several sessions per day for each child; the total time spent standing is counted.8–10

The hip surveillance program and quality register (CPUP)* includes the entire population of children with CP born in Sweden in 2000 and later.4 The program includes a standardized follow-up including gross motor function classified according to the Gross Motor Function Classification System (GMFCS-E&R),11 passive range of motion (ROM),12 movement abilities, time and frequency of positioning, and other treatments such as hip surgery. A standardized radiology follow-up of the hips is also included in the CPUP.12 For those who have been followed from their second birthday, previous studies have shown that the number of hip dislocations (MP = 100%) can be lowered significantly and that the need for surgery can be reduced by including the standardized follow-up.6 The repeated checks ensure that treatments are given at the optimal time, but much effort is still required to prevent hip displacement. One such method is abducted standing.8–10

A biomechanical test model showed that the magnitude of the force on the hip joint increases with increasing hip abduction and that the direction of the force vectors shifts more medially in the acetabulum, which leaves the acetabular rim unaffected. In this model, the difference between 0° and 30° of abduction was substantial (Figure 1).13 In that study, the authors used a wedge to position the sole of the foot perpendicular to the hip. This function is included in the standers used in Sweden.

Fig. 1.:
Directions for acetabular contact forces at different angles of abduction in terms of the pelvic coordinate system. aThe sole of the foot is placed perpendicular to the leg while the child stands in the stander or standing shell.

The recommendation is to apply up to 30° of hip abduction in each leg while standing14 for 1 to 1.5 h/d, 7 days a week. This procedure has been shown to reduce the MP8–10 in both children with CP who do not walk and those with CP at GMFCS level III, who also stopped scissoring in assisted gait.9,10 An earlier, 1-year follow-up study of children at GMFCS level IV or V showed that standing with 30° of hip abduction bilaterally significantly decreased MP both in children who performed preventive abducted standing and in those after surgery.8 No hip or knee contractures developed during the intervention. All studies of abducted standing have used 1 to 1.5 hours of standing divided into several sessions per day, 7 days a week. The minimum total abducted standing time is considered to be 10 h/wk. Some children have continued to practice abducted standing for up to 7 years, but it is currently unknown whether the prolonged use of abducted standing can preserve its positive effects, which have to date been documented for only 1 year.

The purpose of this study was to investigate the long-term effects of standing at 15° to 30° versus 0° to 10° of hip abduction on MP and ROM, both for preventive purpose and after surgery in children with CP who do not walk. A second aim was to quantify the need for additional hip surgery after the periods covered by this study.


Study Design

This was a longitudinal, retrospective case-control study based on information collected by the CPUP program. The intervention evaluated was the use of a stander or standing shell for 10 h/wk, every week, for 8 months to 7 years. The study group practiced with 15° to 30° of abduction and the control group practiced with 0° to 10° of abduction for each leg.


Participants performed the daily standing in their home and were assisted by parents and supervised by a physical therapist (PT). Data were extracted from the CPUP register. For each child, the parents and the PT confirmed the degree of abduction used.


Inclusion criteria:

  • Children with CP enrolled in CPUP before their second birthday.
  • Aged 3 to 17 years at the time of the final measurement.
  • GMFCS level IV or V.
  • Data available regarding standing hours, hip and knee ROM, and hip radiographs on at least 3 occasions related to surgery including the baseline period and the intervention period.
  • Children using a standing device for a defined period with a frequency of 10 h/wk, preventively and/or after surgery.
  • Standing always with 15° to 30° of abduction for each leg in the study group and with 0° to 10° of abduction for each leg in the control group.

Exclusion criteria (Figure 2):

  • Reconstructive hip surgery as the first register in the database.
  • Dislocation at the time of the first registered hip radiograph.
  • Children practicing abducted standing with more than 10° of abduction but for an undefined period and/or irregular frequency.
  • Period after reconstructive surgery because the change in MP could not be attributed to standing posture.
  • Children with a very different follow-up time compared with study participants were excluded from the control groups.
Fig. 2.:
Flow chart showing the data selection from the CPUP register. Excluded participants are represented as leaves in the tree diagram. The number of children assigned to each group are also shown. APT indicates adductor-psoas tenotomy; GMFCS, Gross Motor Function Classification System.

Study Groups

Invitation letters were sent to families of 35 children identified in the register as using standing with 15° or more of abduction for each leg. Twenty-four signed the informed consent form. Data for 5 children who were deceased were included in the study groups. The first author performed interviews with parents to confirm the degree and frequency of abducted standing. The 29 children were assigned to 1 of 2 groups: group S1, for which APT surgery was followed by abducted standing, and group S2, whose participants did not undergo APT surgery before the abducted standing.

Comparison Groups

Comparison groups were matched for GMFCS level, age, and follow-up time from the baseline start to the end of the intervention period. The first control group (C1) included all children who underwent APT surgery and were well matched with S1. The second control group (C2) included children who did not undergo surgery. This group included 10 children per study participant who were born in the same year and with the closest follow-up times. During this matching process, the outcome data were blinded.

Additional interventions15 such as seating and positioning, goal-directed functional training, and botulinum toxin injections were used at similar frequencies in all groups.


The outcome measures were MP and ROM for hip abduction and knee extension. We recorded additional surgical procedures performed after inclusion. Restricted ROM was defined according to the CPUP as hip abduction of less than 20° and knee extension of 20° and less.4 The frequency of standing was recorded for all children. The period of abducted standing (for S1 and S2) was confirmed by interviewing the parents and PTs.

Data Sources and Measurements

Data had been registered in the CPUP register by the local orthopedic surgeons and PTs using a standardized manual.4 Data were obtained for 3 occasions: (1) at the time of the first radiograph and PT report (start of the baseline); (2) at the start of the intervention period (end of the baseline); and (3) at the end of the intervention period (the last available radiograph and PT report). For each child, we included data for the hip with the highest (worst) MP at the end of the intervention.16

Statistical Analyses

Given that the groups were small and normal distributions were not expected, we chose to use nonparametric tests. Linear multiple regression analysis was used to identify associations between abducted standing 10 h/wk, APT surgery, the 2 interventions combined, and ROM after the intervention. To identify differences between groups, the Mann-Whitney U test was performed using IBM SPSS Statistics (version 22.0, IBM Corp, Armonk, New York).


From the CPUP register, 786 children were identified as suitable participants for the study. We excluded 27 children with no baseline data before surgery and 6 children with a dislocated hip at the time of the first hip radiography (Figure 2). Data about physiotherapy, radiography, and surgery were available in the CPUP register for the baseline and intervention periods for 753 children. Of these, 60 children had practiced standing with 15° or more of abduction for each leg during the study period with a mixed standing posture and low frequency and were therefore excluded. This left the following numbers of participants for each group: S1 after APT surgery (n = 13); S2 without surgery (n = 16); C1 after APT surgery (n = 80); and C2 without surgery (n = 160) (Figure 2).

The age range at inclusion was 0.6 to 16.0 years. The median age was 3.7 (1.4-11.4) years at the start of abducted standing and 4 (1.9-11.7) years at the time of surgery. The median baseline follow-up (from the first recording to surgery or start of abducted standing) was 1.5 (0.8-7.5) years. The median intervention follow-up was 3.5 (0.5-8.7) years.

The equipment to perform abducted standing used in S1 and S2 was 21 standers and 8 standing shells. Every stander (21 children) used the most abducted position with 30° of abduction for each leg. All children achieved 30° of abduction within the first week of use. Three standing shells provided 15° of abduction per leg, 3 had 20° to 23°, and 2 had 28° to 30°.

The median MP before intervention was 43% in S1 and 45% in C1, 27% in S2 and 21% in C2 (Figure 3). The mean reduction in MP during intervention period was 12% for the 2 study groups (S1 and S2). After the intervention period, linear regression analyses showed a significant correlation between the performance of abducted standing 10 h/wk and the change in MP (n = 29 in S1 + S2 and n = 240 in C1 and C2) (P = .001).

Fig. 3.:
Changes in the median MP for the worst hip at first radiograph, before intervention and at last radiograph for all groups. MP indicates migration percentage.

Adductor-psoas tenotomy surgery reduced MP by a median 3.5% in C1 (n = 80) and 18% in S1 (n = 13) (P = .026) (Figure 3). Combined in S1 and C1, surgery decreased MP by 7.5%. Within S1, only 1 child, who started the abducted standing intervention with an MP of 71%, needed a varus derotation osteotomy performed within 1 year. In C1, 26 children (33%) needed another surgery (soft tissue or osteotomy) performed within 2 years, but this difference was not significant.

During the baseline period in the groups whose participants did not undergo surgery (S2 and C2), MP increased faster in S2 (16%) than in C2 (3%) (Figure 3). During the intervention period, the median MP decreased by 7.0% in S2 (n = 16) but increased by 6.5% in C2 (n = 160) (P = .001) (Figure 3). In S2, 2 of 16 children had undergone surgery and in C2, 30 of 160 children had undergone surgery after the intervention period, but this difference was not significant.

During the intervention period, none of the study participants (S1 and S2) developed any new contractures. In the control groups (C1 and C2), the prevalence of hip abduction and knee extension contractures increased from 39 to 80 joints. After surgery in S1 and C1, the difference in ROM was significant for both joints: hip abduction (n = 13 and 80, P = .013) and knee extension (n = 13 and 80, P = .04) (Table).

TABLE - Distribution of GMFCS Level, MP During Study, Frequency of Surgery After Study, and Distribution of Contractures
S1: Study Group After APT Surgery C1: Control Group After APT Surgery S2: Study Group Without Surgery C2: Control Group Without Surgery
Number in groups 13 80 16 160
Male/female 5/8 46/34 8/8 92/68
GMFCS level IV/V 5/8 28/52 7/9 80/80
Age at follow-up start, median (range), y 1,3 (0.8-2.3) 2,2 (0.6-7.8) 1,9 (0.7-11.4) 2,6 (0.6-9.3)
Age at intervention start, median (range), y 3,8 (1.8-7.6) 4,2 (1.8-11.7) 3,7 (1.5-11.4) 4,1 (1.4-12.4)
Age at last radiograph/PT record, median (range), y 7,7 (4.1-12.4) 8,0 (3.3-15.8) 6,5 (3.0-14.8) 7,8 (2.2-16.0)
Degree of abduction in standing 30°: n = 8
20°: n = 3
15°: n = 2

0°-10°: n = 80
30°: n = 15
15°: n = 1

0°-10°: n = 160
MP 1, median (range) 19,5 (0-83) 26 (0-62) 11 (0-31) 18 (0-84)
MP 2, median (range) 43 (20-83) 45 (0-80) 27 (0-44) 21 (0-84)
MP 3, median (range) 25 (5-88) 41,5 (5-100) 20,5 (6-50) 28 (0-100)
Median difference during intervention period (MP 3-MP 2) −18,0 −1,0 −5,5 +6,0
Surgery after study period:
Hip abduction ≤20°
End baseline
1 7 1 15
Hip abduction ≤20°
End study
0 11 1 23
Knee extension ≤ −20°
End baseline
1 4 0 13
Knee extension ≤ −20°
End study
0 12 0 34
Abbreviations: APT, bilateral adductor-psoas tenotomy; GMFCS, Gross Motor Function Classification System; MP 1, first available migration percentage; MP 2, last available migration percentage at end baseline; MP 3, last registered migration percentage; PT, physical therapist.


This study showed that the trend of migrating hip changed in children with CP who performed abducted standing compared with their matched controls regardless of whether this preventive treatment was performed with or without surgery. The effects appeared to last at least 7 years in participants who continued to perform abducted standing.

When measured on hip radiographs obtained before the age of 2 years, an MP of less than 10% is expected for normally developed children. In children with CP, increasing muscle tone and MP suggest the need for abducted standing as a preventive measure. Children in S2, who started with the abducted standing early in life, had a similar increase in MP during the baseline period as those in C1, who were referred for surgery. If they had been treated in other districts, these children probably would have been referred for surgery instead of abducted standing. This negative selection should be kept in mind when considering the results of this study. For S2, the intervention altered the natural course of MP to increase over time, that is, in children in the control group that did not undergo surgery (C2), MP continued to deteriorate. In some regions of Sweden, abducted standing is prescribed postoperatively by the orthopedic surgeon.

As shown in Figure 4, the change in MP was visualized in a coordinate system with the same units on both axes and divided by a diagonal: the upper part of the figure includes all hips with a deterioration in MP during the intervention and the lower part includes hips with an improvement in MP during the same time.3 This presentation allowed us to identify cases with the greatest distance to the diagonal, which highlights those with the largest change.

Fig. 4.:
Comparison of MP in the worst hip before and after intervention for all participants. (A) Comparison of MP in the worst hip in the S1 and C1 groups before and after APT surgery. (B) MP in the worst hip for the S2 and C2 groups without surgery. Open squares indicate control group 1 with 0° to 10° standing abduction. Filled symbols indicate study group 1 as follows: 8 round dots indicate participants who performed 30° of standing abduction for each leg; 3 filled squares indicate participants who performed 20° to 23° of standing abduction for each leg; and 2 filled triangles indicate participants who performed 15° of standing abduction for each leg. MP indicates migration percentage.

Three children in the study group with no surgery had an MP of 0° to 5° before the intervention (Figure 4). These radiographs were obtained before the age of 1 year. At the follow-up, MP had progressed to 23%, 23%, and 50% in these 3 children. The amount of cartilage around the hip joint makes MP a less accurate method for measuring hip development at a very young age.

As seen in Figure 1, the load on the acetabulum increases when abduction increases. We assume that the adductor muscles may not have sufficient force in the abducted position because of the elongated position. In this position, the total body weight rests on the femoral head without affecting the acetabular rim. In addition, stretching of the adductor muscles provides extra loading and exerts extra forces, which press the femoral head into the socket.2

Every user of the abducted stander used it with the maximum abduction the stander could provide in this study. The abduction of the standing shells varied. Standing with only 15° of abduction had the smallest effect on MP (Figure 4A). The two filled triangles represent the children using 15° of abduction. They continued to deteriorate during intervention period (Figure 4B). However, our study included few children using less than 30° of abduction, and further studies are needed to confirm the extent of abduction needed to affect hip migration.

All children using the stander achieved 30° of abduction within the first week of use. At least 2 of the children varied the extent of abduction during periods with increased muscle tone while waiting for the next Botulinum toxin injection into their adductor muscles. The degree of abduction was never less than 25° in each leg. However, it is known that botulinum toxin treatment of the adductor muscles alone cannot improve hip development in children with CP.17

In a previous study8 and in this study, the intervention period did not cause any new contractures in the study participants, whereas in the control groups, 9% developed a new knee or hip contracture. The experience from the CPUP is that a small knee contracture can be the start of asymmetrical posture, which can end in the wind-swept position and by itself is a great risk for continuing hip and spine problems such as hip dislocation and scoliosis.16 This is also why we included only the worst hip in our calculations because asymmetry is common in these children.

Ordinary manual stretching does not affect ROM.15,18 Manual stretching cannot be compared with the effect of stretching during standing because the latter is ongoing for hours every day. In this study, children achieved prolonged stretching of the adductor muscles and hip and knee flexors for 10 h/wk, which is recommended for altering ROM.19 Range of motion remained unchanged in groups S1 and S2. These young children had sufficient ROM, and our goal was to preserve it, which we did for at least 7 years.

Much effort is still needed to prevent hip displacement, and such efforts are included in the CPUP follow-up program. However, our study used only 1 possible intervention. When a child has increased muscle tone and the first hip radiograph shows a high MP, the team may recommend abducted standing as a type of treatment. After this study, we expected undiminished effects in those who had continued abducted standing for 7 years. During the search for the study participants, we found a child who had stopped practicing abducted standing after 4 years. Although the benefits had been maintained for the 4 years, after an additional 3 years, 1 hip had become dislocated.

The data from the CPUP allowed us to evaluate the effects of practicing abducted standing for as long as 7 years, which is a new contribution to research on CP. However, the number of children practicing abducted standing was less because only a few districts in Sweden use this intervention. As a child grows taller and becomes heavier, different standing equipment is needed to manage the effort needed to move the child into and out of the standing position.

In conclusion, practicing standing with 15° to 30° of abduction for 10 h/wk preserves MP and ROM for up to 7 years. This benefit may be a better method for preventing contracture than as a mean of decreasing an already large MP. Abducted standing may also be a method for enhancing the long-term results after surgery.


The authors thank Gunnar Hägglund for wise advice and Arve Opheim for interpretation and analysis of the statistical data.


1. Hägglund G, Lauge-Pederson H, Wagner P. Characteristics of children with hip displacement in cerebral palsy. BMC Musculoskelet Disord. 2007;8:101.
2. Miller F, Slomczykowski M, Cope R, Lipton GE. Computer modeling of the pathomechanics of spastic hip dislocation in children. J Pediatr Orthop. 1999;19(4):486–492.
3. Reimers J. The stability of the hip in children: a radiological study of the results of muscle surgery in cerebral palsy. Acta Orthop Scand. 1980;184(suppl):1–100.
4. Hägglund G. Follow-up program for cerebral palsy.Årsrapport-CPUP-2019-PDF.pdf. Published 2020. Accessed July 6, 2020.
5. Pountney T, Mandy A, Green E, Gard P. Management of hip dislocation with postural management. Child Care Health Dev. 2002;28(2):179–185.
6. Hagglund G, Alriksson-Schmidt A, Lauge-Pedersen H, Rodby-Bousquet E, Wagner P, Westbom L. Prevention of dislocation of the hip in children with cerebral palsy: 20-years results of a population-based prevention programme. Bone Joint J. 2014;96-B(11):1546–1552.
7. Herman D, May R, Vogel L, Johnson J, Henderson R. Quantifying weight-bearing by children with cerebral palsy while in passive standers. Pediatr Phys Ther. 2007;19:283–287.
8. Martinsson C, Himmelmann K. Effect of weight-bearing in abduction and extension on hip stability in children with cerebral palsy. Ped Phys Ther. 2011;23(2):150–157.
9. Macias-Merlo L, Bagur-Calafat C, Girabent-Farres M, Stuberg WA. Standing programs to promote hip flexibility in children with spastic diplegic cerebral palsy. Ped Phys Ther. 2015;27(3):243–249.
10. Macias-Merlo L, Bagur-Calafat C, Girabent-Farres M, Stuberg W. Effects of the standing program with hip abduction on hip acetabular development in children with spastic diplegic cerebral palsy. Disabil Rehabil. 2016;38(11):1075–1081.
11. Palisano RJ, Rosenbaum P, Bartlett D, Livingstone MH. Content validity of the expanded and revised Gross Motor Function Classification System (GMFCS-E&R). Dev Med Child Neurol. 2008;50(10):744–750.
12. Stuberg WA, Fuchs RH, Miedaner JA. Reliability of goniometric measurement of children with cerebral palsy. Dev Med Child Neurol. 1988;30(5):657–666.
13. Yang Z. Effect of Standing Position on Joint loading—A Pilot Study. Stockholm, Sweden: KTH. TRITA-STH. EX; 2013.
14. Paleg GS, Smith BA, Glickman LB. Systematic review and evidence-based clinical recommendations for dosing of pediatric supported standing programs. Ped Phys Ther. 2013;25(3):232–247.
15. Novak I, McIntyre S, Morgan C, et al. A systematic review of interventions for children with cerebral palsy: state of evidence. Dev Med Child Neurol. 2013;55(10):885–910.
16. Hägglund G, Lauge-Pedersen H, Persson-Bunke M, Rodby-Bousquet E. Windswept hip deformity in children with cerebral palsy: a population-based prospective follow-up. J Child Orthop. 2016;10(4):275–279.
17. Willoughby K, Ang S, Thomason P, Graham H. The impact of botulinum toxin A and abduction bracing on long-term hip development in children with cerebral palsy. Dev Med Child Neurol. 2012;54(8):743–747.
18. Katalinic OM, Harvey LA, Herbert RD, Moseley AM, Lannin NA, Schurr K. Stretch for the treatment and prevention of contractures. Cochrane Database Syst Rev. 2010;9(1):CD007455.
19. Pin T, Dyke P, Chan M. The effectiveness of passive stretching in children with cerebral palsy. Dev Med Child Neurol. 2006;48(10):855–862.

*CPUP is a short-form for Cerebral Palsy UPpföljning (follow-up). It consists of a follow-up program for children and adults with cerebral palsy and a quality register where their data are kept.


abducted standing; cerebral palsy; child; hip dislocation; hip displacement; prevention; tenotomy

© 2021 Academy of Pediatric Physical Therapy of the American Physical Therapy Association