Secondary Logo

Journal Logo


Femoral Derotation Osteotomy in Cerebral Palsy: Precise Determination by Tables

Čobeljić, Goran; Djorić, Ivan; Bajin, Zoran; Despot, Borislav

Author Information
Clinical Orthopaedics and Related Research: November 2006 - Volume 452 - Issue - p 216-224
doi: 10.1097/01.blo.0000229276.66570.4c


Medial rotation deformity of the hip in cerebral palsy occurs from increased femoral neck anteversion which develops prenatally and is approximately 40° at birth. After birth anteversion rapidly decreases until the age of 4 years, then decreases at a slower rate and comes to a final physiologic position of approximately 15° at skeletal maturity. Femoral neck anteversion in newborns with cerebral palsy is similar to anteversion in healthy children. However, after birth the anteversion in children with cerebral palsy only slightly decreases, or increases and remains high before and after skeletal maturity. In patients with cerebral palsy and medial rotation deformity of the hip these values are even higher. Neuromuscular imbalance in patients with cerebral palsy inhibits the femoral neck anteversion decrease mechanism resulting in a persistence of fetal ante- version described as femoral torsion, which results in a medial rotation deformity of the hip.2,8,17

Medial rotation deformity of the hip is a problem for ambulatory patients with cerebral palsy because of the inward-pointed knees during gait which can cause clumsy walking, stumbling, falling, and frequent injuries. Knees, ankles, and feet on the side of the deformity are subjected to stress, therefore, causing rapid wear of footwear.5,14,19

Nonoperative treatment has been recommended to correct the medial rotation deformity of the hip in patients who have functional problems, however, it is ineffective, and therefore, surgery often is recommended.7,13,17 Excessive femoral neck anteversion and medial rotation deformity of the hip are thought to be best corrected with the femoral derotation osteotomy.4,11,20 Several methods have been used including osteotomies in the proximal or distal parts of the femur using internal or external fixation. The effects of these methods have been assessed based on radiographically determined femoral neck anteversion,9-12,15,20 rotational hip movements,9-15,20 gait analysis,14 and comparison of symptoms before and after surgery.11,14,19 It is unclear, however, whether attempts to correct the increased femoral neck anteversion have reduced symptoms or the patients' ability to walk.

We investigated if there were important differences between planned and achieved corrections of the increased femoral neck anteversion that would compromise our method. We also investigated whether postoperative values of the femoral neck anteversion were similar to planned corrections in bilateral and unilateral operations or, if not, whether the corrections were more successful in patients with bilateral surgery or in patients with unilateral surgery. Finally, we analyzed what influences our method had on patients' complaints and walking ability.


We retrospectively identified 25 patients who had a femoral derotation osteotomy between 1983 and 2003. We included patients who had preoperative and postoperative femoral neck anteversion measurements, preoperative and postoperative assessment of complaints, and walking ability. Of these 25 patients, 17 (21 osteotomies) were included in our study. Of the remaining eight patients, two had chronic femoral osteitis develop which was treated successfully by bone resections, but both patients had loss of correction. One of these two patients is still being treated, but the other patient has not returned for followups. The remaining six patients were lost to followup. Patients were followed up in 6-month intervals for the first 3 years after surgery and then in 1-year intervals for an average of 11 years (range, 3-20 years).

Among the 17 patients, there were five female and 12 male patients. Osteotomies were performed bilaterally in four patients within 6 months. The average age of the patients was 20 years (range, 9-42 years). Most of the patients (14) were between 11 and 25 years. Ten patients had spastic diplegia, four patients had spastic paraplegia, and three patients had spastic hemiplegia. Preoperatively all patients reported some pain or weakness. The most frequent complaints (14 patients) were knock knees and frequent falls. The least frequent complaint was painful feet (two patients). Nine patients had previous surgery for treatment of equinus deformities. Adductor tenotomies of both hips were performed in three patients before the femoral derotation osteotomies.

Indications for femoral derotation osteotomy included symptomatic medial rotation deformity of the hip, spastic cerebral palsy but with independent gait without walking assists, femoral neck anteversion greater than 45°, and age older than 8 years. We considered unilateral or bilateral hip dislocation, other forms of cerebral palsy, or nonambulatory patients as contraindications for osteotomy.

We used a lateral approach with patients lying in a lateral position. The osteotomy was performed in the intertrochanteric region along with the iliopsoas muscle tenotomy. We added intramuscular tenotomy (lengthening), 3 to 5 cm proximal to the lesser trochanter where the tendon was surrounded by the muscle. This procedure made positioning and fixation easier and allowed better muscle balance with antagonists.3 The fragments were fixed with a nail-plate under compression, or in two patients with plates and screws without compression.

These two patients wore an immobilizing hip spica cast for 6 weeks. During the first 6 weeks of postoperative bed rest, strengthening of the quadriceps muscle was recommended. After 6 weeks, exercises of the hip and knee on the surgically treated side were started. Three months after surgery, the patients were allowed to walk with gradual weightbearing using crutches.

The angle of correction was determined preoperatively by measuring femoral neck anteversion in the deformed hips with computed tomography (CT). Femoral neck anteversion was measured on healthy hips in unilateral deformities. The intention was to achieve approximately symmetric values in both hips. In patients with bilateral deformities, the intention was to achieve approximately symmetric anteversion of 15° to 40° (ideally 25° to 35°), which is considered normal in patients with cerebral palsy.3

The angle of derotation α was determined by simple geometry from the formula:

where α is the angle of derotation; r is the mean radius of the bone; and t is the chord connecting points A and B (in geometry a chord is a straight line connecting the extremities of an arc) (Fig 1).

Fig 1
Fig 1:
The method to determine a planned derotation (t) is shown. The variables α and r are known. The chord t = AB is determined from the geometric formula sin α/2 = (t/2)/r, t = 2rsin α/2. The diameter of the circle equals the mean diameter of the elipsoid cross section of a bone.

From this formula the length of the chord (t) can be derived.

The length of the chord (t) given in millimeters equals the number of millimeters the bone fragments shall be derotated. We compiled tables for the range of correction angles from 1° to 90° and for the average bone diameters (2r) of 10 to 49 mm (Tables 1-4).6

Value of Derotation Variables 2r = 10-29 mm and α = 1-45°
Value of Derotation for Variables 2r = 30-49 mm and α = 1-45°
Value of Derotation for Variables 2r = 10-29 mm and α = 46-90°
Value of Derotation for Variables 2r = 30-49 mm and α = 46-90°

Before surgery, we determined the angle of planned derotation (eg, α = 30°). During surgery, a thin line parallel to the bone axis was marked on the anterior or anterolateral and proximal part of the femur by a saw (Fig 2A). The bone then was cross-cut in the intertrochanteric or trochanteric region perpendicular to the line (Fig 2B). The cross-section of the femur was rarely a circle, but rather an ellipse. We then determined the longest diameter of the ellipse using double compasses, and we read this measurement on a ruler (eg, 22 mm). We determined the diameter of the ellipse, which was perpendicular to the already-measured diameter, using double compasses and we read this measurement on a ruler (eg, 18 mm). Next we calculated the mean value of the diameter which equaled the double radius (eg, 2r = (22 + 18)/2 = 20 mm). We referred to the tables to find the planned angle of derotation (α = 30°) in the first column and the value of the diameter (2r = 20 mm) in the first row of Table 1 for this example. At α = 30° and 2r = 20 mm, t = 5.2 mm is the measure of derotation. Therefore, to correct the increased femoral neck anteversion by femoral derotation osteotomy, the proximal fragment is rotated inward or the distal part rotated outward for the value 5.2 mm, determined by double compasses (Fig 2C).

Fig 2A
Fig 2A:
C. The phases of a femoral derotation osteotomy in the proximal part of a femur are illustrated. (A) A thin line parallel to the bone is marked on the anterior or anterolateral and proximal part of the femur by a saw. (B) The bone is cross-cut in the region of the trochanter minor or intertrochanteric region perpendicular to the line. (C) The variable t is the measure of medial rotation of the proximal fragment.

Femoral neck anteversion was measured on CT scans before surgery and 12 to 18 months after surgery (after removal of the fixation devices). This was done by hand measurement, selecting the same cut on the CT scan before and after surgery for each patient. This was done by two independent observers (ZB and BD) who were not operating surgeons and who were blinded to the results. These measurements were considered important in evaluating results.3,14,15

We identified any complaints related to knock knees, falls, painful knees, ankles and feet, excessive wear of shoes before surgery and during each followup. Each patient's ability to walk was assessed before and after surgery to determine if there was deterioration (the patient required support, crutches, or a walker) or loss of walking ability. Because some patients can adapt to this deformity and function well without surgery, we queried them regarding specific postoperative complaints. 19

Statistical differences were evaluated with a Student's t test and Mann-Whitney U test to compare preoperative and followup values of femoral neck anteversion. Statistical significance was set at p < 0.05. Our data met the assumptions of those tests. The analysis of planned and achieved derotations was performed by Student's t test (coefficient variation [CV], < 30%). Analysis of femoral neck anteversion correction that was achieved was performed using the Mann-Whitney test to compare results in patients with unilateral and bilateral surgery (CV > 30%).


No patient had any signs of gait deterioration after surgery.

The planned and achieved angles of correction were similar. The average planned femoral neck anteversion correction (average values of planned derotation) was 31.9° (range, 20°-45°), whereas the average correction achieved was 32.19° (range, 15°-40°). The least difference between the planned and the achieved angle was 1°, and the greatest was 15°. The average preoperative femoral neck anteversion in 21 hips in which correction was planned was 59.28° (range, 45°-75°) (Table 5).

Postoperatively, values of femoral neck anteversion in bilateral or in unilateral operations were similar to planned corrections (Figs 3 and 4). The average preoperative femoral neck anteversion in eight hips in which bilateral correction was planned was 56.87° (range, 45°-70°), and postoperatively, it was 24° (range, 15°-33°). The average preoperative femoral neck anteversion in 13 hips in which unilateral correction was planned was 60.77° (range, 50°- 75°), and postoperatively, it was 28.23° (range, 15°-47°). The average femoral neck anteversion in the hips that were not treated surgically was 27.3° (range, 20°-38°) (Table 5).

Planned and Achieved Correction of Femoral Neck Anteversion p = 0.825
Fig 3
Fig 3:
Achieved corrections of femoral neck anteversion in patients with bilateral medial rotation deformity of the hip are shown in the histogram
Fig 4
Fig 4:
The corrections of femoral neck anteversion achieved for surgically treated hips and values of femoral neck anteversion in hips not surgically treated in patients with unilateral medial rotation deformity of the hip are shown.

During followup, 13 of 17 patients who had surgery had no complaints, and four patients had minor complaints of pain. Three patients (Patients 2, 9, and 10) had persistent minor pain in their knees on the surgically treated side. Two patients (Patients 8, 10) complained of ankle pain on the surgically treated side. Patient 10 had minor complaints of pain in his knee and ankle (Table 6).

Patient Complaints


Previous studies show medial rotation deformity of the hip in patients with cerebral palsy can be treated successfully with a femoral derotation osteotomy.11,13,14,20 Methods to determine the necessary correction of rotational deformities of the femur and long bones have not been sufficiently precise or simple to perform. Our purpose was to introduce a simple and precise method to perform a femoral derotation osteotomy in patients with cerebral palsy and a medial rotation deformity of the hip. The method was evaluated by comparison of planned and achieved corrections of femoral neck anteversion.

Our study has several limitations. The disadvantage of the method is that the calculation is based on an approximation of the mean diameter of the bone because the bone cross section is an ellipse rather than a circle. Given the perpendicular diameters of the bone are not substantially different in size however, we think this approximation is acceptable. We studied gaits only by observation, and more complex gait analysis would have been useful to detect more subtle differences. Unfortunately, expensive three-dimensional gait analysis techniques are not available to us.

Some earlier methods of femoral derotation osteotomy have drawbacks. Using some techniques, the operator had to stop the surgery, and with pencil and paper, in conditions which were not sterile, calculate the measure of rotation.12 After completing the calculation, the operator went through the sterilizing procedure again and then resumed the surgery. These methods allowed femoral neck anteversion corrections in steps of 15° (15°, 30°, 45°, etc)12 or 10° (10°, 20°, 30°, etc).9 These methods were not widely accepted because they were so complicated.

A traditional method using two Kirschner wires or two Steinmann pins positioned in a planned angle of correction and then rotated to a parallel position11,14,16,19,20 also has some disadvantages. It is difficult to position the wires (pins) in a correct angle, especially if the planned angles are not round values (10°, 20°, 30°) but are between (eg, 17°, 23°, 36°) round values. Checking achieved correction during surgery by rotational hip movements is not precise.

An Ilizarov frame allows precise femoral neck anteversion corrections,10 but experience in patients with spastic cerebral palsy is insufficient.13 Spasms may be difficult to control in patients wearing such a frame, and this apparatus has other drawbacks such as infections around wires and more difficult postoperative care in patients with cerebral palsy.13

We think the necessary angle of correction of anteversion must be determined precisely before surgery. The necessary rotation of the bone fragments is determined during the course of the surgery by the method we have described. The method's simplicity is evident as only two parameters [the mean diameter (2r) of the bone and the angle (α) of planned derotation (correction) of the femoral neck anteversion] determine the value (t) given in millimeters for which the bone fragment shall be derotated (Tables 1-4; Fig 2).

The data suggest we can achieve the desired results (Table 5) and decrease symmetry of the lower extremities (Figs 3 and 4). Preoperative complaints were eliminated in most patients, and substantially reduced in others (Table 6). Residual knee complaints in four patients were related to external rotation of the tibia. This deformity, however, was not substantial and the patients had no additional surgery for the deformity. No patient experienced deterioration of their walking ability.

Our results confirm this method can be used in patients with cerebral palsy who require correction of increased femoral neck anteversion. We think it also can be used for any patient with any rotational deformity of any long bone. The technique, using our tables, is simple to perform, and most patients benefit from the surgery because the planned derotation angles are more easily achieved. Walking ability improved and problems associated with cerebral palsy and medial rotation deformity of the hip were reduced.


1. Beals RK. Developmental changes in the femur and acetabulum in spastic paraplegia and diplegia. Dev Med Child Neurol. 1969;11: 303-313.
2. Bleck EE. Developmental orthopaedics. III: Toddlers. Dev Med Child Neurol. 1982;24:533-555.
3. Bleck EE. Orthopaedic Management in Cerebral Palsy. Philadelphia, PA: JB Lippincott Co; 1987.
4. Bleck EE. Management of the lower extremities in children who have cerebral palsy. J Bone Joint Surg Am. 1990;72:140-144.
5. Brunner R, Krauspe R, Romkes J. Torsion deformities in the lower extremities in patients with infantile cerebral palsy: pathogenesis and therapy. Orthopade. 2000;29:808-813.
6. Djoric I. Metod za preciznu korekciju deformiteta dugih kostiju. Acta Orthop Iug. 1976;1:37-42.
7. Engel GM, Staheli LT. The natural history of torsion and other factors influencing gait in childhood: a study of the angle of gait, tibial torsion, knee angle, hip rotation, and development of the arch in normal children. Clin Orthop Relat Res. 1974;99:12-17.
8. Fabry G, MacEwen GD, Shands AR Jr. Torsion of the femur: a follow-up study in normal and abnormal conditions. J Bone Joint Surg Am. 1973;55:1726-1738.
9. Fait M, Janovec M. Subtrochanterická derotačni osteotomie femuru. Acta Chir Traum Čech. 1976;43:58-61.
10. Herzenberg JE, Smith JD, Paley D. Correcting torsional deformities with Ilizarov's apparatus. Clin Orthop Relat Res. 1994;302:36-41.
11. Hoffer MM, Prietto C, Koffman M. Supracondylar derotational osteotomy of the femur for internal rotation of the thigh in the cerebral palsied child. J Bone Joint Surg Am. 1981;63:389-393.
12. Lasserre G., Saint-Supery J.. Ann Chir Infant. 1966;7:165-170.
13. Moens P, Lammens J, Molenaers G, Fabry G. Femoral derotation for increased hip anteversion: a new surgical technique with a modified Ilizarov frame. J Bone Joint Surg Br. 1995;77:107-109.
14. Ounpuu S, DeLuca P, Davis R, Romness M. Long-term effects of femoral derotation osteotomies: an evaluation using three- dimensional gait analysis. J Pediatr Orthop. 2002;22:139-145.
15. Peterson HA, Klassen RA, McLeod RA, Hoffman AD. The use of computerised tomography in dislocation of the hip and femoral neck anteversion in children. J Bone Joint Surg Br. 1981;63: 198-208.
16. Root L, Siegal T. Osteotomy of the hip in children: posterior approach. J Bone Joint Surg Am. 1980;62:571-575.
17. Samilson RL. Orthopedic surgery of the hips and spine in retarded cerebral palsy patients. Orthop Clin North Am. 1981;12:83-90.
18. Shands AR Jr, Steele MK. Torsion of the femur: a follow-up report on the use of the Dunlap method for its determination. J Bone Joint Surg Am. 1958;40:803-816.
19. Staheli LT. Torsion: treatment indications. Clin Orthop Relat Res. 1989;247:61-66.
20. Tylkowski CM, Rosenthal RK, Simon SR. Proximal femoral osteotomy in cerebral palsy. Clin Orthop Relat Res. 1980;151:183-192.
© 2006 Lippincott Williams & Wilkins, Inc.