We studied the prevalence of severe crouch gait over a 15-year period in a defined population of children with spastic diplegia and Gross Motor Function Classification System levels II and III, to determine if there had been a decrease following changes to the management of equinus gait. These changes were replacing observational with three-dimensional gait analysis, replacing single level with multilevel surgery, and replacing gastrocsoleus lengthening with gastrocnemius recession. Of 464 children and adolescents with spastic diplegia who underwent three-dimensional gait analysis, 27 had severe crouch gait. Seventeen of these had been managed by isolated lengthening of the gastrocsoleus. Following changes in the management of equinus gait, the prevalence of severe crouch gait decreased from 25% and stabilised at a significantly lower rate, fluctuating between 0% and 4% annually (p < 0.001).

We conclude that severe crouch gait in this population was precipitated by isolated lengthening of the gastrocsoleus. These findings may be relevant to other surgical populations, as severe crouch gait may be a useful way to monitor the quality of the surgical management of abnormal gait in children with cerebral palsy and spastic diplegia.

Introduction and Aims: We studied the outcome of single event multilevel surgery (SEMLS) for the correction of severe crouch gait in spastic diplegia, over a five-year period. It was unknown if gait correction post-SEMLS could be sustained at skeletal maturity.

Method: This was a prospective cohort study, utilising validated outcome measures. Presenting symptoms were increasingly abnormal gait, anterior knee pain, patellar fractures and fatigue. SEMLS was based on pre-operative gait analysis: mean of seven procedures (range 5–10), including lengthening of contracted muscle-tendon units (particularly hamstrings and psoas), as well as rotational osteotomies and bony stabilisation procedures to correct lever arm dysfunction. Post-operatively subjects wore Ground Reaction Ankle Foot Orthoses (GRAFOs) and received a community-based rehabilitation program. Post-operative changes were evaluated at five years: technical outcome by 3D kinematics and functional outcome by mobility status. Outcomes were analysed with linear regression with robust standard errors.

Results: Eleven children with spastic diplegic cerebral palsy fulfilled the criteria for ‘severe crouch gait’, defined as knee flexion > 30 degrees and ankle dorsiflexion > 15 degrees throughout stance. Ten of 11 subjects had previous Tendo Achilles lengthening. Mean age pre-operatively was 12 years one month (range 8–16) and at follow-up 17 years 10 months (range 16–21). All subjects regained pre-operative mobility levels with improved gait pattern, relief of knee pain and healing of patellar fractures. There was a significant decrease in dependence on assistive devices. Pre- versus five years post-operative kinematics showed clinically and statistically significant increases in knee extension and decreases in ankle dorsiflexion. Improvements were seen in knee extension initial contact (p< 0.001, 95% CI 15°, 31°); maximum knee extension (p< 0.001, 95% CI 16°, 37°), ankle dorsiflexion (p< 0.001, 95% CI 8°, 18°) and plantarflexion 3rd rocker in stance (p=0.03, 95% CI 1°, 17°); knee excursion (p=0.003, 95% CI –24°, −6°), and peak knee flexion timing (p=0.02, 95% CI 2%, 20%).

Conclusion: Multilevel surgery for severe crouch gait in spastic diplegia results in consistently marked improvements in dynamic knee and ankle function, but not at the hip and pelvic levels. The results are durable in most patients, after five years and after reaching skeletal maturity.

We describe the results of a prospective study of 28 children with spastic diplegia and in-toed gait, who had bilateral femoral derotation osteotomies undertaken at either the proximal intertrochanteric or the distal supracondylar level of the femur. Preoperative clinical evaluation and three-dimensional movement analysis determined any additional soft-tissue surgery.

Distal osteotomy was faster with significantly lower blood loss than proximal osteotomy. The children in the distal group achieved independent walking earlier than those in the proximal group (6.9 ± 1.3 v 10.7 ± 1.7 weeks; p < 0.001). Transverse plane kinematics demonstrated clinically significant improvements in rotation of the hip and the foot progression angle in both groups. Correction of rotation of the hip was from 17 ± 11° internal to 3 ± 9.5° external in the proximal group and from 9 ± 14° internal to 4 ± 12.4° external in the distal group. Correction of the foot progression angle was from a mean of 10.0 ± 17.3° internal to 13.0 ± 11.8° external in the proximal group (p < 0.001) compared with a mean of 7.0 ± 19.4° internal to 10.0 ± 12.2° external in the distal group (p < 0.001). Femoral derotation osteotomy at both levels gives comparable excellent correction of rotation of the hip and foot progression angles in children with spastic diplegia.