With observed success and increased popularity of growth modulation techniques, there has been a trend towards use in progressively younger patients. Younger age at growth modulation increases the likelihood of complete deformity correction and need for implant removal prior to skeletal maturity introducing the risk of rebound deformity. The purpose of this study was to quantify magnitude and identify risk factors for rebound deformity after growth modulation.

We performed a retrospective review of all patients undergoing growth modulation with a tension band plate for coronal plane deformity about the knee with subsequent implant removal. Exclusion criteria included completion epiphysiodesis or osteotomy at implant removal, ongoing growth modulation, and less than one year radiographic follow-up without rebound deformity. Mechanical lateral distal femoral angle (mLDFA), mechanical medial proximal tibial angle (mMPTA), hip-knee-ankle angle (HKA), and mechanical axis station were measured prior to growth modulation, prior to implant removal, and at final follow-up.

Sixty-seven limbs in 45 patients met the inclusion criteria. Mean age at growth modulation was 9.8 years (range 3.4–15.4 years) and mean age at implant removal was 11.4 years (range 5.3–16.4 years). Mean change in HKA after implant removal was 6.9O (range 0O–23 O). Fifty-two percent of patients had greater than 5O rebound and 30% had greater than 10O rebound in HKA after implant removal. Females less than ten years and males less than 12 years at time of growth modulation had greater mean change in HKA after implant removal compared to older patients (8.4O vs 4.7O, p=0.012). Patients with initial deformity greater than 20O degrees had an increased frequency of rebound greater than 10O compared to patients with less severe initial deformity (78% vs 22%, p=0.002).

Rebound deformity after growth modulation is common. Growth modulation at a young age and large initial deformity increases risk of rebound. However, rebound does not occur in all at risk patients, therefore, we caution against routine overcorrection. Patients and their families should be informed about the risk of rebound deformity after growth modulation and the potential for multiple surgical interventions prior to skeletal maturity.

The natural history of gait pattern change in children with spastic diplegia is a transition from toe walking to progressive hip and knee flexion with eventual crouch gait. This has been attributed to the adolescent growth spurt, progressive lever arm dysfunction, and iatrogenic weakening of the soleus with isolated tendo achilles lengthening (TAL). The relative contribution of TAL to the development of crouch gait is uncertain. The purpose of this study was to identify the frequency of crouch gait in spastic diplegic patients with and without history of prior TAL.

Patients with spastic diplegia greater than 10 years of age with instrumented gait analysis were reviewed. Exclusion criteria included diagnosis other than cerebral palsy, prior dorsal root rhizotomy, or incomplete past surgical history. Patients were divided into three groups: Group 1, no prior orthopaedic surgical intervention; Group 2, prior orthopaedic surgery without TAL; Group 3, prior orthopaedic surgery with TAL. Instrumented gait analysis data was analysed. Gait data were analysed using a single randomised limb from each patient.

One hundred and seventy-eight patients were identified: 39 in Group 1, 49 in Group 2, and 90 in Group 3. Mean time from TAL to gait analysis was 7.5 years (range 1.0–14.6 years). Mean age at TAL was 6.3 years (range 1.2–17.5 years). There was no significant difference in age, BMI, walking speed, or cadence between groups. Kinematic analysis showed no significant difference in mean stance phase maximum knee or ankle flexion between groups. There was no significant difference in frequency of increased mid stance knee flexion between groups (Group 1, 53.8%; Group 2 46.9%; Group 3, 43.3%, p=0.546). There was a trend towards increased frequency in excessive stance phase ankle dorsiflexion in Group 3 (60% Group 3 vs 46.2% Group 1, and 40% Group 2, p=0.071). Crouch gait (stance minimum hip flexion > 30, mid stance minimum knee flexion > 200, and stance maximum ankle dorsiflexion > 150) was seen with similar frequency in all groups (Group 1, 23.1%; Group 2, 18.4%; Group 3, 26.7%; p=0.544).

There is a trend towards increased frequency of excessive stance phase ankle dorsiflexion in spastic diplegic patients with prior TAL. However, no significant difference in frequency of crouch gait between patients with and without history of TAL was identified. Crouch gait is part of the natural history of gait pattern change in spastic diplegic patients independent of prior surgical intervention.