Secondary Logo

Journal Logo

Guided Growth: Novel Applications in the Hip, Knee, and Ankle

Guided Growth

Novel Applications in the Hip, Knee, and Ankle

Bouchard, Maryse MD, MSc, FRCSC

Author Information
Journal of Pediatric Orthopaedics: September 2017 - Volume 37 - Issue - p S32-S36
doi: 10.1097/BPO.0000000000001022
  • Free


Hemiepiphyseodesis for angular deformities in the growing child is effective and avoids the complications of osteotomies and gradual deformity correction with external fixation. This article reviews applications of screw hemiepiphyseodesis for coronal and sagittal plane deformities in the lower extremity. Indications, technical pearls, potential complications, and the literature review are discussed.


Screw hemiepiphyseodesis of the medial malleolus was first described by Stevens and Belle1 in 1997 as a single vertical screw centered in the midcoronal plane of the malleolus to correct ankle valgus. Medial malleolus hemiepiphyseodesis is indicated for ankle valgus ≥5 to 8 degrees in patients with open physes.1–3 In published series on this technique, patients have a mean age of 10 to 11 years at time of surgery.1–4 Most surgeons used a single 4.0 or 4.5 mm cannulated fully threaded screw and achieved a total correction of 9.7 to 12 degrees effectively correcting the deformity.1,2 Mean correction rate was 0.37 to 0.65 degrees/month (Fig. 1).1,3–5

Medial malleolus screw hemiepiphyseodesis: 5-year-old girl with right limb hypoplasia. Correction of ankle valgus with intentional slight overcorrection in 30 months (X-rays courtesy of Vincent Mosca, MD, Seattle, WA).

Some papers noted on varying rates of correction based on the patient’s condition.1,4 Myelomeningocele patients corrected the slowest, whereas children with cerebral palsy and clubfoot corrected the fastest.4 Rupprecht et al2 found no difference in correction rate between medical conditions, but adolescents corrected faster than children under 10 years. Davids et al5 reported that fully threaded screws corrected faster than partially threaded screws though this was not statistically significant.

Stevens et al3 compared hemiepiphyseodesis techniques. They reported that the tension band plate construct was as effective as screw hemiepiphyseodesis in correcting ankle valgus, but had less risk of hardware failure or joint penetration, and easier hardware removal (Fig. 2). Skin breakdown was more common with the plate construct.3

Broken medial malleolus hemiepiphyseodesis screw: 12-year-old boy with multiple hereditary exostosis (X-ray courtesy of Peter Stevens, MD, Salt Lake City, UT).

There are few complications reported with screw hemiepiphyseodesis of the medial malleolus. Intraoperative fluoroscopy is needed to avoid placing the screw too laterally, risking penetration into the ankle joint and erosion of the medial dome of the talus. Screw fracture and migration are reported.1,3 Sagittal plane deformity can occur if the screw is not central in the medial malleolus.5 No papers have reported physeal arrest from this technique.

A rebound effect (recurrence of valgus) after hardware removal before skeletal maturity is commonly reported. Rebound rates vary from 0.28 to 0.6 degrees/month.5 Stevens et al3 recommend overcorrecting to 5 degrees of varus if screw is removed while the physis is open.


There is no description in the literature of screw epiphyseodesis of the fibula for deformity correction, however there are anecdotal reports. In conditions such as achondroplasia and hypochondroplasia where the fibula is long relative to the tibia, ankle varus, and painful prominence of the distal and/or proximal fibula can result. Some surgeons have placed epiphyseodesis screws in the distal and/or proximal fibula to prevent/correct these deformities (personal oral communication). Children with skeletal dysplasias grow more slowly and do not have the typical adolescent growth spurt. If fibular epiphyseodesis screws are considered, the procedure should be performed when the child is young and the deformity mild to maximize the effectiveness of guided growth (Fig. 3A). The surgeon should monitor for potential screw breakage, migration, and joint penetration (Fig. 3B).

A, Distal fibular screw hemiepiphyseodesis: 5-year-old girl with achondroplasia with painful knees and ankles, tibia vara, and ankle varus. Improved alignment in 3 years. B, Proximal screw migration (arrow) of distal fibular hemiepiphyseodesis screw (X-rays courtesy of Klane White, MD, Seattle, WA). Postop indicates postoperative.


In 1998, Metaizeau et al6 first described and popularized the use of percutaneous transphyseal screws in the knee for correcting leg length inequality and angular deformity. Screws in the proximal tibia and distal femur are used to correct valgus, varus, and procurvatum of the knee with excellent results. One or 2 cannulated 4.5 to 7.3 mm fully threaded screws are placed percutaneously on the convex side of the deformity, as close to the periphery of the physis as possible.7,8

In studies of screw hemiepiphyseodesis for angular deformities of the knee, mean patient age at time of procedure is 12 to 14 years.7–9 Typically, screws are placed in the proximal tibia in a retrograde manner, and antegrade in the distal femur.7–9 The reported rate of correction in the femur is 0.73 to 75 degrees/month and mean total correction is 6.91 to 7.6 degrees.7,8 The tibia corrects more slowly at 0.37 degrees/month with a mean total correction 3.88 degrees.7

Khoury et al7 reviewed the effectiveness of screw hemiepiphyseodesis in the knees of 30 patients (60 physes) for angular deformity and found the best results in patients with Blount disease, and that tibial varus corrected faster than tibial valgus. Sanchez Mesa and Yamhure8 performed a prospective study of 100 knees in 52 patients with pathologic genu valgum who had antegrade medial distal femoral screw hemiepiphyseodesis. They reported no overcorrection. Undercorrection only occurred in female patients older than 15 years as physeal closure occurred soon after surgery. There were no major complications. Minor complications included hardware irritation, hemarthrosis, and superficial infection.7–9

Screw hemiepiphyseodesis for guided growth is preferred by some surgeons to avoid the known complications of staples (loss of fixation, hardware failure, and physeal arrest) and the diminished accuracy of drill hemiepiphyseodesis.8,9 No physeal arrests with the screw technique have been reported.

In 2007, Stevens9 described the use of a tension band plate construct in the knee for angular deformity correction, deeming it accurate and effective. Although the different techniques of hemiepiphyseodesis in the knee have not been directly compared, Stevens et al10 reported a greater total deformity correction with staples (9 to 10 degrees in the femur) than the reported correction with screws, and considered the plate construct technically less demanding to perform than screw hemiepiphyseodesis.9 Rebound is reported with all 3 techniques.8–10

Hemiepiphyseodesis for fixed flexion deformities (FFD) of the knee in children with conditions such as cerebral palsy, arthrogryposis, and spina bifida is also effective and reduces the need for extensive soft tissue releases and osteotomies.11,12 Only hemiepiphyseodesis using staple and tension band plate techniques have been published,11–13 although there are surgeons using screw hemiepiphyseodesis (personal communication). This technique is reserved for patients with a >10 degrees FFD with >12 months of growth remaining.11,13

MacWilliams and colleagues reported on 4 children with anterior distal femoral tension band plates for FFD who showed improvements in gait, contractures, and pain. Patella alta occurred but was not found to be clinically significant.14 Klatt and Stevens11 reviewed 18 patients (29 knees) with >10 degrees FFD (mean 23 degrees) who had tension band plate hemiepiphyseodesis and achieved a mean correction of 15 degrees. Undercorrection occurred in 1 child who had >12 months of growth remaining.11 Kramer and Stevens12 reported on stapling for FFDs in 28 children (47 knees) with a mean FFD of 10 to 25 degrees. Correction of 0 to 11 degrees was achieved. There was 1 case of staple extrusion but no other complications and no physeal arrest. None of the patients required osteotomies.12

At our institution since February 2014, we have performed 13 cases of screw hemiepiphyseodesis for FFD: 12 cases in the femur (Fig. 4) and 1 in the tibia (Figs. 5A, B). Patients ranged from 6 to 14 years of age. Most children had myelomeningocele, Larsen syndrome, and arthrogryposis. In 1 child, the femoral screws bent during correction, but there have been no complications or revisions, and adequate correction was achieved in all cases (personal communication).

A, Anterior distal femur screw hemiepiphyseodesis: 10-year-old boy with Pierre Robin syndrome and 11 degrees FFD bilaterally with improved alignment in 17 months. Screws removed successfully on February 25, 2016. Note mild bending of screws. B, Anterior distal femur screw hemiepiphyseodesis: intraoperative wire placement (X-rays courtesy of Vincent Mosca, MD, Seattle, WA).
A, Anterior proximal tibia screw hemiepiphyseodesis: 6-year 4-month-old boy with Larsen syndrome and knee fixed flexion deformities. Screws placed July 10, 2014 and near full correction achieved at latest follow-up within 6 months. B, Anterior proximal tibia screw hemiepiphyseodesis: intraoperative wire placement (X-rays courtesy of Vincent Mosca, MD, Seattle, WA).


To prevent or minimize deformity associated with greater trochanteric (GT) overgrowth, surgeons have attempted to slow or halt growth of the GT apophysis. In 1967, Langenskiold and Selenius13 first described a Phemister hemiepiphyseodesis of the lateral GT for overgrowth in the setting of avascular necrosis of the hip. In 1980, Gage and Cary reviewed 15 patients who had the same procedure. They reported worse results in children with Legg-Calve-Perthes (LCP) disease than those with neonatal avascular necrosis or developmental hip dysplasia. Outcomes were improved if done before 8 years of age.15 Matan et al16 found the Phemister hemiepiphyseodesis of the GT effective in preventing deformity in 5 to 10 year olds with LCP if performed with a proximal femoral varus osteotomy. Van Tongel and Fabry reinforced the importance of age at the time of procedure. Their group had a mean age of 10.6 years and radiographic GT overgrowth present at the time of surgery. No patient had improvement in deformity.17 Stevens et al18 described hemiepiphyseodesis of the GT with a tension band plate and concomitant soft tissue releases in 12 patients with LCP (mean age: 7.3 y) with good results. One child later required an osteotomy and 4 had resulting leg length discrepancies (Fig. 6).

Inferomedial proximal femoral screw hemiepiphyseodesis: girl presenting at 6 months old with fixed L hip dislocation, medial open reduction (MOR) performed at 7 months, and development of subsequent type II lateral growth arrest at the age of 6 years. Screw placement at the age 8 with plan to remove at age 11 years (X-rays courtesy of Leo Donnan, MD, Melbourne, Australia).

Although not yet published, some surgeons perform screw epiphyseodesis of the GT considering it less invasive or technically demanding than the above methods, with good results and no complications other than the rare screw migration (personal communication).

Screw hemiepiphyseodesis in the inferomedial femoral neck has been described for correction of coxa valga secondary to lateral type II growth arrest of the proximal femoral physis in the setting of developmental hip dysplasia.19,20 A single large-diameter fully threaded cannulated screw is placed percutaneously across the inferomedial proximal femoral physis to preserve superolateral growth.19,20 Torode and Young reviewed 11 patients aged between 5 and 14 years with prior medial open reduction of the hip. Significant improvements in femoral and acetabular radiographic anatomy were seen after screw hemiepiphyseodesis. Four patients required advancement or revision of the screw.20 McGillion and Clarke19 described full correction of coxa valga in 8/10 patients with this technique. Mean age was 12.3 years. This method has no major complications reported, but a postoperative computed tomography is recommended to ensure the screw has not penetrated into the joint (personal communication).


Combined with tension band plates and staples, screw hemiepiphyseodesis offers many options for guided growth to manage pediatric lower extremity deformity. It can obviate the need for invasive, complex reconstructive surgeries, decreasing risks, and lowering costs. Screw hemiepiphyseodesis is becoming more widely used and the present literature, although limited, suggests it is effective and safe. Complications are infrequent and minor, mostly consisting of hardware irritation, migration, or breakage. No physeal arrest has been reported with screw hemiepiphyseodesis. Innovation of new applications and techniques of guided growth should be encouraged and published.


1. Stevens PM, Belle RM. Screw epiphyseodesis for valgus ankle. J Pediatr Orthop. 1997;17:9–12.
2. Rupprecht M, Spiro AS, Breyer S, et al. Growth modulation with a medial malleolar screw for ankle valgus deformity: 79 children with 125 affected ankles followed until correction or physeal closure. Acta Orthop. 2015;86:611–615.
3. Stevens PM, Kennedy JM, Hung M. Guided growth for ankle valgus. J Pediatr Orthop. 2011;31:878–883.
4. Chang FM, Ma J, Pan Z, et al. Rate of correction and recurrence of ankle valgus in children using a transphyseal medial malleolar screw. J Pediatr Orthop. 2015;35:589–592.
5. Davids JR, Valadie AL, Ferguson RL, et al. Surgical management of ankle valgus in children: use of a transphyseal medial malleolar screw. J Pediatr Orthop. 1997;17:3–8.
6. Metaizeau JP, Wong-Chung J, Bertrand H, et al. Percutaneous epiphyseodesis using transphyseal screws. J Pediatr Orthop. 1998;18:363–369.
7. Khoury JG, Tavares JO, McConnell S, et al. Results of screw epiphysiodesis for the treatment of limb length discrepancy and angular deformity. J Pediatr Orthop. 2007;27:623–628.
8. Sanchez Mesa PA, Yamhure FH. Percutaneous hemi-epiphysiodesis using transphyseal cannulated screws for genu valgum in adolescents. J Child Orthop. 2009;3:397–403.
9. Stevens PM. Guided growth for angular correction: a preliminary series using a tension band plate. J Pediatr Orthop. 2007;27:253–259.
10. Stevens PM, Maguire M, Dales MD, et al. Physeal stapling for idiopathic genu valgum. J Pediatr Orthop. 1999;19:645–649.
11. Klatt J, Stevens PM. Guided growth for fixed knee flexion deformity. J Pediatr Orthop. 2008;28:626–631.
12. Kramer A, Stevens PM. Anterior femoral stapling. J Pediatr Orthop. 2001;21:804–807.
13. Langenskiold A, Selenius P. Epiphyseodesis of the greater trochanter. Acta Orthop Scand. 1967;38:199–219.
14. MacWilliams BA, Harjinder B, Stevens PM. Guided growth for correction of knee flexion deformities: a series of four cases. Strategies Trauma Limb Reconstr. 2011;6:83–90.
15. Gage JR, Cary JM. The effects of trochanteric epiphyseodesis on growth of the proximal end of the femur following necrosis of the capital femoral epiphysis. J Bone Joint Surg Am. 1980;62:785–794.
16. Matan AJ, Stevens PM, Smith JT, et al. Combination trochanteric arrest and intertrochanteric osteotomy for Perthes’ disease. J Pediatr Orthop. 1996;16:10–14.
17. Van Tongel A, Fabry G. Epiphysiodesis of the greater trochanter in Legg-Calve-Perthes disease: the importance of timing. Acta Orthop Belg. 2006;72:309–313.
18. Stevens PM, Anderson LA, Gililland JM, et al. Guided growth of the trochanteric apophysis combined with soft tissue release for Legg-Calve-Perthes. Strategies Trauma Limb Reconstr. 2014;9:37–43.
19. McGillion S, Clarke NMP. Lateral growth arrest of the proximal femoral physis: a new technique for serial radiological observation. J Child Orthop. 2011;5:201–207.
20. Torode IP, Young JL. Caput valgum associated with developmental dysplasia of the hip: management by transphyseal screw fixation. J Child Orthop. 2015;9:371–379.

screw hemiepiphyseodesis; guided growth; valgus; varus

Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.