Elsevier

Gait & Posture

Volume 42, Issue 4, October 2015, Pages 569-574
Gait & Posture

Correction of gait after derotation osteotomies in cerebral palsy: Are the effects predictable?

https://doi.org/10.1016/j.gaitpost.2015.09.003Get rights and content

Highlights

  • First study showing the effect of tibial osteotomies on knee rotation during gait.

  • The amount of intraoperative derotation of the tibia showed excellent relation.

  • The amount of intraoperative derotation of the femur showed no relation.

  • Best predictor of femoral rotation was the preoperative hip rotation during gait.

  • Strength and passive hip extension did not show any predictive significance.

Abstract

Derotation osteotomies of the femur and tibia are established procedures to improve transverse plane deformities during walking with inwardly pointing knees and in- and out toeing gait. However, effects of femoral derotation osteotomies on gait were reported to be small, and those for the tibia are not known. Therefore, the aim of the study was to show the relation between the amount of intraoperative rotation and the changes during gait for osteotomies at femur and tibia levels, and predict those for the femur from preoperative clinical and gait data.

Forty-four patients with spastic cerebral palsy between 6 and 19 years were included, 33 limbs received rotation only at the femur, 8 only at the tibia and 12 limbs at both levels. Gait analysis and clinical testing was performed pre- and 21.4 (SD = 1.8) months postoperatively.

The amount of intraoperative derotation of the femur showed no significant correlation with the change in hip rotation during walking (R = −0.17, p = 0.25), whereas the rotation of the tibia showed an excellent relationship (R = 0.84, p < 0.001) with the change in knee rotation. Preoperative hip rotation during walking explained only 18% of the variability of the postoperative change in hip rotation during gait. Strength and passive range of motion in hip extension and abduction as well as hip extension or abduction or foot progression during walking did not show any predictive significance.

In conclusion changes of knee rotation during gait is directly predictable from the amount of tibial corrections, contrary the change in hip rotation was not related to the amount of femoral derotation, and prediction was only fair.

Introduction

Excessive internal hip rotation during walking, seen as inwardly pointing knees, is common in children with cerebral palsy (CP) and involves 49% of the patients [1]. It can lead to knocking or rubbing of the knees, lever arm dysfunction [2] and is particularly cosmetically unappealing. Internal hip rotation during walking is often associated with in-toeing gait; these children with internal foot progression angle may be particularly prone to tripping and falling. However, normal or external foot progression angle may also present, and need to be accounted for in a successful therapy that aims to achieve normal alignment of the leg during gait.

The primary reason for internal rotation of the hip is increased anteversion of the femur [3], [4], [5]. Other cause might be flexion contractures of the hip [4] or weakness of external hip rotators [3]. Internal foot progression angle might be the results of internal hip rotation in the transverse plane and might further increase with internal torsional malalignment of the tiba. Normal or external foot progression angle, in the presence of excessive internal hip rotation, may be caused by external torsional malalignment of the tibia; this combination is referred to as miserable malalignment.

Since primary reasons for leg malalignment are torsional deformities of the bones, external derotation osteotomies of the femur (FDO) and tibial rotation osteotomies (TRO) either internal or external are the common procedures to obtain normal leg alignment. FDO was shown to be effective in correcting intoed gait if passive and dynamic hip rotations are both excessively internal as indicated by physical examination and gait analysis [6], [7], [8], [9]. It is not clear why following this recommendation for the correction of femoral anteversion with FDO, 33% of under corrections have been observed [10], and only 60% of the intraoperative amount of correction that aim to improve gait was observed postoperatively during walking [11]. In order to solve this problem, slight overcorrection with more external than internal passive range of motion (PRoM) has been suggested [11], [12]. However, such overcorrections may lead to outward pointing knees during walking [7]. Furthermore, the hip joint allows a considerable range of motion so that muscular weakness, spasticity and compensatory movements [4] might lead to further internal hip rotation following correction of the anteversion. This makes it difficult to plan the appropriate degrees of derotation in FDO to achieve normal leg alignment during gait. Neither age, GMFCS level nor additional soft tissue surgeries in addition to FDO distinguished good from bad responders to FDO [10]. Regarding TRO there is no information in the literature whether there is a correlation between the intraoperative extent of the derotation and the change of knee rotation during walking in patients with CP. Logically, due to the smaller range of motion in the transverse plane at the knee joint, good correlation can be expected between intraoperative amount of derotation and changes during gait. However this correlation is yet to be shown.

Therefore the aim of this study is first, to show the correlation between intraoperative amount of TRO and FDO and the changes during gait. Second, since weak correlations are expected for FDO, we investigate whether predictors from preoperative clinical and gait assessments were able to predict the improvements during walking.

Section snippets

Patients and surgical interventions

Forty-four patients with spastic cerebral palsy GMFCS I-III, with uni- and bilateral involvement between 6 and 19 years volunteered to participate in the study. Participants provided written consent, as approved by the local ethics committee. Of these patients 53 limbs received FDO and/or TRO between 2010 and 2013; patient's details are shown in Table 2. FDO was performed at the distal supracondylar level of the femur using a condylar blade plate fixation. TRO was performed supramalleolar at

Surgical outcome during gait and correlation to the intraoperative derotation

The average external FDO measured intraoperatively was 29.3° (SD = 5.2) [20°, 40°] Functional improvement in mean hip rotation in stance was −18.5° (SD = 10.6) [−42°, 3°] that corresponds to 64% (SD = 39) of the intraoperative amount of FDO. For TRO 15/20 patients received internal, 5/20 external rotation osteotomies. To combine the effects of internal and external TRO in the evaluation, the negative intraoperative TRO and the change of knee rotation were multiplied by (−1) for patients with external

Discussion

For FDO, no significant correlation between intraoperative derotation of the femur and the change in hip rotation during walking was found. Further, the efficiency of FDO on hip rotation during walking cannot be predicted from preoperative muscle strength and PRoM. Only hip rotation observed during gait showed a fair relation. For TRO, the intraoperative rotation of the tibia showed an excellent relationship with the change in knee rotation during walking.

No significant correlation between the

Conclusion

TRO is a predictable procedure that showed the same effects on gait as intraoperatively rotated. Contrary, the effect of FDO cannot be reliably predicted from preoperative measures. Greatest improvements during walking can be expected when the hip is excessively internally rotated. In addition, growth during the period until evaluation likely causes a loss of correction in hip rotation.

Conflict of interest statement

All authors do not have any financial and personal relationships with other people or organizations that inappropriately influence the work performed.

References (21)

There are more references available in the full text version of this article.

Cited by (0)

View full text