Spatiotemporal scaling changes in gait in a progressive model of Parkinson's disease

Abstract

Objective: Gait dysfunction is one of the most difficult motor signs to treat in patients with Parkinson's disease (PD). Understanding its pathophysiology and developing more effective therapies for parkinsonian gait dysfunction will require preclinical studies that can quantitatively and objectively assess the spatial and temporal features of gait.

Design: We developed a novel system for measuring volitional, naturalistic gait patterns in non-human primates, and then applied the approach to characterize the progression of parkinsonian gait dysfunction across a sequence of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) treatments that allowed for intrasubject comparisons across mild, moderate, and severe stages.

Results: Parkinsonian gait dysfunction was characterized across treatment levels by a slower stride speed, increased time in both the stance and swing phase of the stride cycle, and decreased cadence that progressively worsened with overall parkinsonian severity. In contrast, decreased stride length occurred most notably in the moderate to severe parkinsonian state.

Conclusion: The results suggest that mild parkinsonism in the primate model of PD starts with temporal gait deficits, whereas spatial gait deficits manifest after reaching a more severe parkinsonian state overall. This study provides important context for preclinical studies in non-human primates studying the neurophysiology of and treatments for parkinsonian gait.

Keywords: MPTP; Parkinson's disease; gait; non-human primate; pressure walkway.

Figure 1

mUPDRS sub-score evaluations of cardinal motor signs following MPTP treatments in each subject. Evaluations were conducted in the mild parkinsonian state for two subjects (B, A), moderate parkinsonian state for all subjects, and severe parkinsonian state for two subjects (B, N). mUPDRS sub-scores included evaluation of the most affected side of the body for tremor, akinesia, bradykinesia and rigidity, plus whole-body axial motor signs. The span of each bar represents the sub scores and the height reflects the total additive scores.

Figure 2

Gait testing apparatus and spatiotemporal analysis of gait parameters. (A) The GTA consists of a tunnel capped at each side by end enclosures with sliding doors (not shown). A pressure mat (gray shaded area) recorded location, time, and magnitude of contact pressure (pixelated blue hand and foot prints). Food or juice reward was delivered through hoppers in each end enclosure. (B) Overhead perspective of pixelated calibrated pressure data (color bar, kPa) recorded for a representative pass/trial from right-to-left. Post-processed ‘strike-boxes' placed around each hand or foot print identified contact points and pressure magnitudes by limb (front/hind limbs shown as dashed/solid lines, respectively, and left/right sides shown in green/purple, respectively). Interlimb distance is the length between leading edges of consecutive ipsilateral limb “strike boxes”. (C) Time series of the aggregate pressure data converted into kilogram-force (kg-f) captured within each ‘strike-box' defined temporal gait parameters and stance force profiles for each limb. The representation of the first stance, swing, and stride progression shown from right-to-left (B) is the same instance shown from left-to-right in the time series. Time and distance variables (red) are illustrated for each gait parameter, see Table 2.

Figure 3

Whole-body gait parameter changes with MPTP treatment. Means and 95% CI for cadence (A), gait speed (B), stride speed (C), and stride length (D) for subjects A, B, and N across naïve and parkinsonian (mild, moderate, severe) states. *denotes significance at p < 0.05 (statistical results detailed in Tables S1–S4).

Figure 4

Changes in swing and stance phases between naïve and parkinsonian states. (A) Comparison of swing length (y axis) and swing time (x axis) of the front (top row) and hind (bottom row) limbs. Shaded ellipses represent 95% confidence intervals about the respective data set. Lines intersect at the mean swing time and swing length for the given state. (B) A similar comparison between stance time and stance length of the front (top row) and hind (bottom row) limbs. *denotes a significant difference (p < 0.05) of the variables between states.