Starting position effects in the measurement of the postural vertical for pusher behavior.

Pusher behavior (PB) is a severe lateral postural disorder that involves a disturbed subjective postural vertical (SPV) in the frontal plane. SPV is measured by determining the mean value and standard deviation of several trials beginning on both the contralesional- and ipsilesional-tilted positions. However, the postural representation, when passively tilted to the contralesional versus ipsilesional position, is different between patients with and without PB. Therefore, we hypothesized that SPV dependence on the starting position will be influenced by PB. For 53 patients with hemispheric stroke enrolled, SPV was measured using a non-motorized vertical board with eyes closed. The mean value (tilt direction) and standard deviation (variability) were calculated in four trials, each from two positions, with the patient tilted to the contralesional position (SPV-CL condition) and then to the ipsilesional position (SPV-IL condition). Patients were categorized into the non-pusher (n = 29) and pusher (n = 24) groups. In the SPV-CL trials, the tilt direction was significantly tilted contralesionally for the pusher group (- 6.3° ± 1.6°) compared with that for the non-pusher group (- 2.2° ± 1.8°; p < 0.001), with no significant difference in variability between the groups. In the SPV-IL trials, the tilt direction was not significantly different between the groups, but the variability was significantly higher in the pusher group (4.8° ± 2.0°) than in the non-pusher group (2.2° ± 1.3°; p < 0.001). The dependence of tilt direction and variability of SPV on the starting position in patients with PB differed from those noted in patients without PB. These results may help explain this abnormal posture and optimize neurological rehabilitation strategies for PB.