Observational study of 180° turn using Inertial Measurement Units in post-stroke ambulatory patients.

OBJECTIVE: Fall is a common complication in post-stroke populations and more likely occurs in left paretic (LP) than in right paretic patients (RP). Since kinematic characteristics of post-stroke patients' turn have not been clearly established, the goal of this study was to analyze with inertial measurement units (IMUs) spontaneous and constraint 180° turn in those patients and in aged matched healthy controls (HC).

MATERIAL/PATIENTS AND METHODS: Seventeen RP patients (43 to 73 years, mean 57.5), 22 LP patients (43 to 63 years, mean 59.6) and 15 healthy controls (36 to 83 years, mean 56.7) were included. All subjects were right handed. They were instructed to turn 180° in a self-selected direction and then in the constraint direction, wearing 3 IMUs on the trunk and on both feet. A synthetic kinematic data representation allowed to observe turning parameters.

RESULTS: Differences were observed in turn duration, number of external steps and mean angular velocity between HC and RP, HC and LP, but not between RP and LP. The spontaneous side of 180° turn was the paretic side for 50% of RP versus 95% of LP. Spontaneous turn to the paretic side was initiated with the paretic foot (i.e. the paretic foot was first in stance phase after turn onset) for 50% of RP versus 90% of LP. Considering that turning to the paretic (and hypoesthesic) side is at high risk of fall, we propose to grade the risk of fall from 1 (low) to 4 (high): turn to the healthy side initiated with the healthy foot (score 1/4), healthy turn with paretic foot (2/4), paretic turn with healthy foot (3/4) and paretic turn with paretic foot (4/4). In recent post-stroke patients (<6 months), none of the RP had a risk 4/4 versus 83% of the LP.

DISCUSSION-CONCLUSION: This original study permitted us to observe spontaneous and constraint 180° turn in post-stroke patients using a synthetic representation with IMUs. Thanks to sensors, the foot on which the 180° turn was initiated could be determined and showed differences between RP and LP. Those results are consistent with fall epidemiology.