NS03 - Incorporating kinematic analysis of postural instability into the surgical evaluation of patients with movement disorders

Introduction/Purpose: Parkinson’s disease (PD) and normal pressure hydrocephalus (NPH) are neurological disorders characterized by gait and balance problems, which lead to falls. These motor symptoms are evaluated during ON/OFF medication testing and lumbar drain trial (LDT) for PD and NPH patients, respectively. In NPH patients, gait typically improves due to increased stride length and velocity after surgery. However, postural instability (PI) among PD and NPH patients is not well known. While there are no current treatments for PI in PD, our prior work demonstrated that ventriculoperitoneal shunting (VPS) in NPH patients leads to PI improvement.

This study examined prospective quantitative kinematics among PD and NPH patients to determine which deficits were associated with PI and compared to healthy age-matched controls. Furthermore, we wanted to investigate the effect of surgical intervention (Deep Brain Stimulation (DBS) for PD patients, VPS for NPH patients) on PI compared to baseline using quantitative kinematics.

Methods or Case Description: 29 PD patients, 14 NPH patients, and 20 healthy age-matched controls wore inertial measurement units (IMUs) and underwent 10-20 pull-tests of varying intensities performed by a trained clinician. Patients participated in 96 sessions totaling 1377 trials. Patients were evaluated OFF/ON medication (PD), pre-LDT and post-LDT (NPH) and then also again at 12 months post-operatively.

Outcomes: Groups were compared using the relationship between peak COM acceleration (pull intensity) and step length (slope), and the overall step length (y-intercept). At baseline, PD and NPH patients demonstrated similar ability to scale their step length reaction to pull intensity (i.e., no significant difference in slope of OFF meds, pre-LDT, compared to controls). However, the Y-intercept significantly decreased for OFF meds and pre-LDT evaluations compared to controls. NPH patients demonstrated a significant increase in slope but no change in Y-intercept post-LDT compared to pre-LDT (i.e., NPH patients reacted slowly and took small initial steps but appropriately took larger, quicker steps in response to more intense pulls). At follow-up, VPS significantly improved the overall reaction time and step length. Conversely, PD patients showed a significant increase in Y-intercept but a decrease in slope upon taking dopaminergic medication. At baseline, PD patients’ COM velocity reflected mild PI, closer to healthy controls, but did not improve with DBS. Before DBS, PD patients reacted slower and took smaller steps as pull intensity increased, opposite of a healthy response. DBS worsened overall reaction time and step length in PD patients but normalized the modulation of response.

Conclusion: Prospective kinematic evaluation demonstrated postural stability profiles for PD and NPH patients undergoing ON/OFF testing and LDT, respectively, are distinguishable from healthy age-matched controls at baseline and respond differently to intervention. Before VPS, NPH patients showed severe PI, but with normal scaling responses. VPS improves PI and retains normal scaling responses. Before DBS, PD patients have mild PI, but abnormal scaling responses. DBS has a complex relationship with postural response leading to improved scaling responses, but reaction time and initial step length appear worse for lower intensity pulls. This indicates that single, standardized perturbations are not adequate when evaluating PD patients as the results vary with perturbation intensity. This analysis can be incorporated into surgical evaluation of movement disorder patients, and future work should focus on the neural mechanisms of postural instability to develop novel treatments.