Improvements in Trunk Sway Observed for Stance and Gait Tasks during Recovery from an Acute Unilateral Peripheral Vestibular Deficit

Objectives: Our aim was to track improvements in postural control during recovery from an acute unilateral peripheral vestibular deficit (UVL), presumably due to vestibular neuritis, and to determine if recovery rates were different for stance and gait tasks. Postural control was quantified using simple measurements of trunk sway: amplitudes of trunk sway angle and angular velocity, in the roll and pitch directions as well as task duration, were examined for a battery of stance and gait tasks. These measures were collected at the onset of the deficit and then 3 weeks and 3 months later. Study Design: A repeated-measures design was used for UVL subjects and age-matched healthy controls. Stance tasks involved standing on 1 or 2 legs with eyes open or closed. Gait tasks consisted of tandem gait, walking normally with eyes closed, or with the head rotating or head pitching, walking up and down stairs and walking over a series of low barriers. Stance and tandem gait tasks were repeated using a foam support surface instead of a normal floor. Patients: Twenty-eight patients with acute UVL were examined. Main Outcome Measures: The range of trunk sway angular displacement and angular velocity in the pitch and roll directions was measured for each task in addition to task duration. The measures were compared with those of normal subjects. Results: The amplitudes of pitch trunk sway for 2-legged stance tasks with eyes closed underwent the greatest reduction 3 weeks after UVL onset. At 3 months, trunk sway was almost normal for all 2-legged stance tasks. One-legged stance tasks with eyes open showed a similar but slower improvement. Stance time without a fall showed a very rapid improvement for 1-legged tasks but was still shorter than that of normal subjects at 3 months. Trunk sway for the simple gait tasks was within normal range at 3 months; however, task duration was still longer than normal. More complex gait tasks, such as walking 8 tandem steps on foam or walking up and down stairs, showed no improvement in trunk roll sway at 3 months. A mix of variables from mainly gait tasks best identified a balance deficit due to UVL, with complex gait tasks becoming more important for identification purposes as compensation progressed. The accuracy of UVL identification with durations alone was 75% of the accuracy with combined trunk sway and duration measures. Conclusions: These data suggest that recovery of normal trunk control during the compensation process for unilate