Estimating the effect size of surgery to improve walking in children with cerebral palsy from retrospective observational clinical data

Apoorva Rajagopal,Scott L Delp,Łukasz Kidziński,Michael H Schwartz,Jennifer L Hicks,Alec S Mcglaughlin

doi:10.1038/s41598-018-33962-2

Apoorva Rajagopal, Scott L Delp + Show 4 more

Open Access

https://doi.org/10.1038/s41598-018-33962-2

Copy DOI

Abstract

Single-event multilevel surgery (SEMLS) is a standard treatment approach aimed at improving gait for patients with cerebral palsy, but the effect of this approach compared to natural progression without surgical intervention is unclear. In this study, we used retrospective patient history, physical exam, and three-dimensional gait analysis data from 2,333 limbs to build regression models estimating the effect of SEMLS on gait, while controlling for expected natural progression. Post-hoc classifications using the regression model results identified which limbs would exhibit gait within two standard deviations of typical gait at the follow-up visit with or without a SEMLS with 73% and 77% accuracy, respectively. Using these models, we found that, while surgery was expected to have a positive effect on 93% of limbs compared to natural progression, in only 37% of limbs was this expected effect a clinically meaningful improvement. We identified 26% of the non-surgically treated limbs that may have shown a clinically meaningful improvement in gait had they received surgery. Our models suggest that pre-operative physical therapy focused on improving biomechanical characteristics, such as walking speed and strength, may improve likelihood of positive surgical outcomes. These models are shared with the community to use as an evaluation tool when considering whether or not a patient should undergo a SEMLS.

Highlights

Cerebral palsy is a motor disorder characterized by abnormal neurological and musculoskeletal development[1]
Thomason et al.[19] performed a pilot randomized study to examine the effects of a single-event multilevel surgeries (SEMLS), but the study only contained 19 patients, which is likely to be too small a sample given the heterogeneity of patient characteristics in the population. To address these limitations of past research and aid clinical decision-making, we aimed to (i) build and share regression models to predict the effect size of a SEMLS for a limb, quantified as improvement in the Gait Deviation Index (GDI)[20] at a follow-up visit when controlling for expected progression without surgery, and (ii) use these models to infer patient characteristics that are associated with improved outcomes
Our propensity score models to predict the likelihood of a limb receiving a SEMLS conditioned on the first-visit gait analysis data were able to, on average, classify limbs as control or surgery limbs with 67% accuracy, 66% sensitivity, and 71% specificity in the training limbs, and 70% accuracy, 66% sensitivity, and 74% specificity in the testing limbs

Summary

Introduction

Cerebral palsy is a motor disorder characterized by abnormal neurological and musculoskeletal development[1]. These tools can predict positive outcomes with over 70% accuracy, and consistent applications of the tools would likely lead to improvement on historical positive outcome rates These separate models tended to have common outcome predictors, suggesting there are intrinsic patient factors that increase likelihood of positive outcomes, regardless of the specific intervention. Studies predicting the outcome of SEMLS have found that the pre-operative level of the chosen outcome variable is the strongest predictor of post-operative improvement[11,12]. When controlling for this effect, the strongest predictors were Gross Motor Function Classification Scale (GMFCS)[11] and dynamic motor control[12]. There is agreement that the gait of patients from different functional groups evolve differently over time[14,16,17]

Objectives

Methods

Results

Discussion

Conclusion