Does spinal fusion in the thoraco-lumbar spine modify neck, thorax, thorax-pelvis and pelvis kinematics during gait on the ground?

Abstract

Introduction In the case of an adolescent with Idiopathic Scoliosis (IS), a complex three-dimensional deformity of the spine was observed which affected the structural bones of the pelvis (Holewijn et al. 2017). To evaluate the incidence of this deformity and the influence of spinal fusion (which prevents the further development of spinal deformity), gait analysis was performed under different conditions (such as treadmill versus over-ground; imposed velocity versus free velocity) (Holewijn et al. 2017). However, using a treadmill influences kinematics: Staszkiewicz et al. (2012) revealed differences between walking on a treadmill and walking over-ground in the case of healthy subjects. Walking on a treadmill causes the pelvis to make lower displacements in the transverse plane. Furthermore, this condition has been used to evaluate the influence of spinal fusion in the case of patients with IS (with imposed velocity) (Holewijn et al. 2017). The reduction of pelvic mobility in the transversal plane observed after spinal fusion could be exaggerated and influence the upper segments of the trunk. Therefore, for patients with IS, it seems interesting to evaluate the incidence of spinal fusion on such pelvic and spinal segments during walking overground conditions. It seems that patients with IS use different strategies like changing the pattern of pelvic motion or changing the velocity of walking to minimize incidences of scoliosis (Holewijn et al. 2017). In healthy subjects, Feipel et al. (2001) showed specific correlations between Range of Motion (RoM) of the lumbar spine and speeds. For patients with IS, correlations between the motion of the pelvis and walking velocity could be used to describe the influence of scoliosis and spinal fusion. Therefore, regarding walking over-ground and free speed, two questions are asked: (1) with similar walking speeds before/after spinal fusion, is there a reduction after spinal fusion in the kinematics of four spinal segments (neck, thorax, thoracic-pelvis and pelvis) in patients with IS? (2) Does this surgery reduce the correlation between walking velocity and the kinematics of the pelvis? 2. Methods 2.1. Population Patients with IS (n=17; 14 women and 3 men; mean age: 17.00 years (3.81); Cobb angle before fusion: 49.7degree (12.8); Cobb angle after fusion: 17.24degree (7.5)) enrolled in a clinical trial comparing gait data before and one year after spinal fusion. They were operated by the same orthopedic surgeon. A gait analysis was subject to medical follow-up. Patients gave informed consent form to used anonymous data. They were recruited from patients scheduled for a spinal fusion between January 2013 and December 2016. Asymptomatic subjects (n=17; 10 women and 7 men; 19.63 years (0.81) were our control population. 2.2. Material and method Subjects walk on a 10m level ground walkway, at self-selected speed. From 'Plug-in gait full body' model applied on 31 retro reflective spherical markers (14mm in diameter) captured with motion capture system (Vicon, Oxford Metrics, 100 Hz), gait cycles were defined as strides and kinematic data were explored. RoM on neck angles, thorax angles, lumbar angles (considered as thoracic-pelvis and computed as Holewijn et al. 2017), pelvis angles in each plane and gait velocity were computed for five gait cycles per subject. For each variable, data were averaged for each subject and condition. The sample size relates to the statistical power (90%) computed for RoM of pelvis mobility in coronal plane (patients before fusion and controls). All variables are presented as: mean (standard deviation). The following tests were used to establish the consequences of scoliosis on gait data (Statistica, V13, Dell, USA): Wilcoxon signed rank test for matched pairs (in case of patients), Mann- Whitney test (to compare patients and controls) and Spearman's correlation. The accepted significance level was set at p