Motion analysis and surgical results of anterior transfer of flexor hallucis longus for equinovarus gait in children with hemiplegia

Abstract

Background

Rigid equinovarus foot deformities are seen in patients with cerebral palsy (CP). This retrospective study was undertaken to evaluate flexor hallucis longus tendon (FHL) transfer with gastrocsoleus recession (GSR) using motion analyses and quantitative measurement, and to investigate postoperative complications.

Methods

This study included 10 hemiplegic CP patients who underwent FHL transfer with GSR, and were evaluated by motion analyses consisting of weight distribution in static standing position and three-dimensional gait analysis, both pre and post-operatively. They were assessed in terms of kinematic data, Gait Variable Scores (GVS), and Gait Profile Score (GPS).

Results

The mean age at operation was 7.3 years (range, 4–13 years), and mean follow-up duration was 35 months (range, 25–64 months) post-operatively. Weight distribution at surgical site significantly rose from 34.3% pre-operatively to 47.3% post-operatively, and abnormal asymmetry of weight distribution between surgical site and contralateral site disappeared post-operatively. Maximum ankle dorsiflexion (ADF) at initial contact rose from −20.9° to −6.28°. Similarly, Maximum ADF at both stance and swing phase rose from −13.8° to 17.7° (P = 0.0003), and from −19.5° to 1.35° (P = 0.001), respectively. Although mean GPS decreased from 15.6° pre-operatively to 11.8°, which corresponded to 2.38 times the minimal clinically important difference (MCID = 1.6°), three cases manifested talipes calcaneus at final follow-up.

Conclusion

Although quantitative assessment showed that the potential value of FHL transfer with GSR was to obtain initial heel contact and maintain sufficient clearance from the ground in swing, it also revealed a risk of leading to talipes calcaneus. In the near future, we should establish accurate criteria for determination of transfer site, and consider the possibility of modification of this procedure in order to balance between recurrent equinus and significant talipes calcaneus.

Study design

Clinical comparison between preoperative and postoperative.

Introduction

One important manifestation for spastic hemiplegia patients is equinovarus gait. There are several surgical techniques for this deformity, including Vulpius, Baker, and Strayer. In addition, the prevalence of post-operative recurrence among these patients is greater than with spastic diplegia [[1], [2], [3]].

Rodda et al. [4] modified Winters, Gage and Hicks's classification [5] and established a useful treatment algorithm for spastic hemiplegia. In addition, they recommended hinged ankle-foot-orthoses (AFO) or solid AFO as post-operative treatment to obtain appropriate clearance from the ground at swing phase.

In Japan, there is a culture of removing shoes indoors, and “tatami” flooring, which is difficult to walk on with AFO due to friction and damage to the flooring. Patients who cannot maintain ground clearance such as those with spastic hemiplegia have difficulty walking indoors. Ono et al. [6] first described a surgery for neuromuscular disorders, which consisted of gastrocsoleus recession (GSR) by Baker method, combined with anterior transfer of toe flexors. Thereafter, Hiroshima et al. [7] modified this procedure to transfer of flexor hallucis longus (FHL) alone. They assessed the surgical results by (1) gait stability and heel-to-ball gait, (2) correction of foot deformity, and (3) patient satisfaction, and found good outcomes and effectiveness for neuromuscular disorders including cerebral palsy (CP) by these qualitative assessments. However, this procedure was not widely used because of broad concern regarding transfer of a spastic tendon. Therefore, few reports on the outcomes of gait analysis after this procedure have been published [[8], [9], [10]].

Three-dimensional gait analysis (3DGA) is regarded as the standard modality for evaluating gait disturbance in CP patients, comparing pre-operative and post-operative results. The methods of evaluation have varied, including Gait Deviation Index (GDI) [11], Gait Variable Scores (GVS), and Gait Profile Score (GPS) [12]. Both GVS and GPS have been expressed by Movement Analysis Profile (MAP). Many authors have described assessment for CP patients through not only 3DGA, but also weight distribution including foot pressure rather than simple weight distribution [[13], [14], [15]].

The purpose of this study is to report the outcomes of FHL transfer with GSR for equinovarus gait with hemiplegia; particularly, we would like to reveal the postoperative complications of this old surgical technique using quantitative assessments and consider possible improvements to this procedure.