Long-term follow-up after multilevel surgery in cerebral palsy

Abstract

Introduction Single-event multilevel surgery (SEMLS) is frequently used to correct pathological gait patterns in children with bilateral spastic cerebral palsy (BSCP) in a single session surgery. However, in-depth long-term evaluation reports of gait outcomes are limited. Therefore, we investigated if SEMLS is able to correct lower extremity joint and pelvic angles during gait towards typically developing gait patterns (TDC) in children with BSCP, and if so, if this effect is durable over a 10-year period. Materials and methods In total 13 children with BSCP GMFCS level II at time of index-surgery between the ages of 7.7–18.2 years at the time of SEMLS were retrospectively recruited. Three-dimensional gait data were captured preoperatively, as well as at short-, mid-, and long-term post-operatively, and used to analyze: movement analysis profile (MAP), gait profile score (GPS), and lower extremity joint and pelvic angles over the course of a gait cycle using statistical parametric mapping. Results In agreement with previous studies, MAP and GPS improved towards TDCs after surgery, as did knee extension during the stance phase (ɳ2 = 0.67; p < 0.001), while knee flexion in the swing phase (ɳ2 = 0.67; p < 0.001) and pelvic tilt over the complete gait cycle (ɳ2 = 0.36; p < 0.001) deteriorated; no differences were observed between follow-ups. However, further surgical interventions were required in 8 out of 13 of the participants to maintain improvements 10 years post-surgery. Conclusions While the overall gait pattern improved, our results showed specific aspects of the gait cycle actually deteriorated post-SEMLS and that a majority of the participants needed additional surgery, supporting previous statements for the use of multilevel surgery rather than SEMLS. The results highlight that the field should not only focus on the overall gait scores when evaluating treatment outcomes but should offer additional long-term follow-up of lower extremity function.