Complementary use of statistical parametric mapping and gait profile score to describe walking alterations in multiple sclerosis: a cross-sectional study

Gait analysis is often used to study locomotor alterations in people with multiple sclerosis (PwMS), but the large number of extracted variables challenges the interpretability. In this paper, we analysed gait alterations by combining the Gait Profile Score (GPS), which summarizes kinematic locomotor deviations, and Statistical Parametric Mapping (SPM), which compares kinematics and kinetics over the whole gait cycle. Eleven PwMS and 11 speed-matched Healthy Controls (HC) underwent overground gait analysis. GPS were compared through independent-samples t-tests; sagittal-plane kinematics and power at hip, knee, and ankle were compared through SPM Hotelling’s-T2 and SPM t-tests. Spearman’s correlation coefficients (r) between GPS and clinical outcomes were also calculated. PwMS had higher GPS than HC (PwMS = 8.74 ± 2.13°; HC = 5.01 ± 1.41°;p < 0.001). Multivariate SPM found statistically significant differences at 0–49%, 70–80%, and 93–99% of stride (p < 0.05) and univariate analysis showed reduced ankle dorsiflexion, and lower knee flexion during pre-swing and swing. GPS correlated with Expanded Disability Status Scale (r = 0.65; 95%C.I.[0.04,0.91]; p = 0.04) and 2-Minute Walking Test (r = -0.65; 95%C.I.[-0.91,-0.04]; p = 0.04). GPS in conjunction with SPM revealed multi-joint kinematic alterations on sagittal plane involving distal joint angles, ankle and knee, during the stance phase with no changes at the proximal level. Gait deviations were more pronounced in PwMS with higher disability and walking limitations.


Materials and methods
Gait profile score Introduced by Baker et al., the Gait Profile Score (GPS) is a measure of the overall quality of gait kinematics that combines nine Gait Variable Scores (GVS): pelvic tilt, rotation and obliquity, hip flexion-extension, adduction-abduction and rotation, knee flexion-extension, ankle dorsiflexion and foot progression. 1 Each GVS was here calculated as the root mean square (RMS) difference between a patient's specific timenormalized gait variable and the average kinematic curve obtained from the healthy population across the gait cycle. Finally, the RMS average of all the 9 GVS will then equal the GPS.
Supplementary Table 1 and Supplementary Table 2 below explain how the kinematic variables used to calculate the nine GVS were calculated. Joint angles were calculated as the relative rotation between proximal and distal local reference frames, 2 which are defined by the direction of axes and planes determined by anatomical landmarks, associated to physical markers or calculated from them. 3 Supplementary Table 1. Description of absolute and local reference frames used to determine the kinematic variables.

Functional frame
The vertical upwards Y axis coincides with the gravity line.
The forward X axis corresponds, in the gait trials, to the axis interpolating the posterior superior iliac spines (PSIS_MX) trajectory in the transverse plane (i.e. forward progression direction).
The Z axis is the cross-product between X and Y axes.

Pelvis
The forward-oriented X-axis passes through the midpoint between the posterior superior iliac spines (PSIS) and the midpoint between the anterior superior iliac spines (ASIS).
The upwards Y axis is perpendicular to the PSIS/ASISs plane.
The Z axis, pointing to the right, is the cross-product of X and Y axes.

Thigh
The thigh upwards longitudinal Y axis passes through the hip joint center (defined by the anthropometric measures) and the knee joint center (KJC, defined as midpoint between femur lateral condyle and femur medial condyle).
The forward X axis is perpendicular to the plane identified by the thigh longitudinal axis Y and by the vector defined by the femur condyles.
The thigh Z axis is the cross-product of X and Y axes.

Shank
The upwards longitudinal Y axis passes through the KJC and the ankle joint centre (AJC, midpoint between lateral and medial malleola).
The shank forward X axis is perpendicular to the plane identified by the shank longitudinal axis Y and by the vector defined by malleola.
The shank Z axis is the cross-product of X and Y axes.

Foot
The longitudinal X axis passes through the AJC and the midpoint between the first metatarsal head and fifth metatarsal head.
The upwards Y axis is perpendicular to the plane in the foot longitudinal axis X and the vector defined by metatarsal heads.
The Z axis is the cross product between X and Y axes.

Supplementary Table 2.
Description of the nine kinematics variables used to calculate the Gait Profile Score.

Kinematic variables Description
Pelvic angles Reference frames: absolute and local pelvis.
The pelvic tilt is the pelvis rotation about the absolute transverse axis.
A floating axis is defined as the perpendicular axis to both the absolute transverse axis and the pelvic longitudinal axis. The pelvis obliquity is the angle about the floating axis.
The pelvis rotation is the angle about the longitudinal pelvis axis.
Hip angles Reference frames: local pelvis and local thigh.
The hip flexion angle is the rotation about the pelvis transverse axis.
A floating axis is defined as the perpendicular axis to both the pelvis transverse axis and the thigh longitudinal axis. The hip ab/adduction is the angle about the floating axis.
The hip rotation is the angle about the longitudinal thigh axis.
Knee angles Reference frames: local thigh and local shank.
The knee flexion angle is the rotation about the thigh transverse axis.
Ankle angles Reference frames: local shank and local foot The ankle flexion angle is the rotation about the shank transverse axis. PwMS was assumed to be the same as in HC. According to the noise data on healthy participants from Luciano et al., we assumed a NAMP of 5°, 8° and 5° for hip, knee, and ankle joint kinematics, respectively, and a NFWHM of 30%. 9 Finally, power analysis was conducted for a SPM t-test comparing hip, knee, and ankle angular trajectories between two independent samples, with an allocation ratio of 1:1. Type II error was kept below 0.20 (i.e., statistical power higher than 0.8), while the family-wise error rate (α) was controlled through a Holm-Bonferroni correction in the form: where α was set to 0.05, n was the total number of multiple comparisons (3,