[A randomized controlled trial of a treadmill training with the perturbation to improve the balance performance in the community dwelling elderly subjects]

Abstract

We had developed a gait perturbation system based on a treadmill composed of two separate walking belts. The servomotors in this treadmill are controlled by a personal computer. A falling perturbation can be generated by the sudden and random deceleration of the walking belts. The purpose of this study was to determine, in a cohort of community dwelling elderly subjects, whether our treadmill training with the perturbation would improve balance performance. The 29 participants aged over 65 years were randomly assigned to receive either perturbed treadmill training or ordinary treadmill training. The single leg standing time with the eyes opened and closed, the functional reach test (FR), the timed up and go test (TUG), the 10 m maximum gait speed and the latency of the bilateral tibialis anterior muscle (LTA) were measured at the pre training period and after the 4-week training period. These indexes were measured again one month after the training period. There were no significant differences between the perturbed training subjects and the ordinary training subjects in these indexes in the pre-training period. Following training period, the perturbed training subjects were improved in FR (mean +/- SD: pre 30.3 +/- 5.0 cm, post 33.5 +/- 5.0 cm, p<.01). TUG (mean +/- SD; pre 5.13 +/- .68 sec, post 4.86 +/- .53 sec, p<.05) and LTA (the perturbed side, mean +/- SD; pre 134 +/- 30 msec, post 116 +/- 22 msec, p<.01, the non-perturbed side, mean SD:pre 132 +/- 35 msec, post 114 +/- 27 msec, p <.05). On the other hand, the ordinary training subjects were improved only in TUG (mean +/- SD; pre 4.93 +/- . 52 sec, post 4.60 +/- .39 sec. p<.01). Additionally, the improvements of the LTA on the perturbed training subjects were retained one month after the training period (the perturbed side, mean +/- SD: pre 134 +/- 30 msec, after 1-month 117 +/- 19 msec, p<.01, the non-perturbed side, mean +/- SD: pre 132 +/- 35 msec, after 1-month 115 +/- 24 msec, p<.05). The improvements of LTA may represent the improvements of the ability of the stepping reaction which is one of the important factors to avoid the falling. Our treadmill training with perturbation improves the balance performance and may decrease the risk of falling in the community dwelling elderly subjects.