Effects of maximal speed locomotor training on spatiotemporal gait changes in individuals with chronic stroke: A secondary analysis of a randomized controlled trial.

This was a secondary analysis of an RCT and was not initially powered for these spatiotemporal outcomes. Only non-paretic step length showed a between-group difference, so the result could be a false positive. There was no group that received no training, so changes in both groups cannot be separated from spontaneous recovery. Spatiotemporal outcomes were obtained at 192 of 220 planned timepoints, with only 18 of 27 HIIT participants having 12-week data. Baseline gait data were obtained for 26 of 27 HIIT participants and 27 of 28 MAT participants. Participants had to be able to walk 10 meters without continuous assistance and have stable cardiovascular status, so results may not apply to more impaired or medically complex patients. The people collecting gait data were not blinded, although the walkway measurements were automated. Non-paretic step length is an imperfect proxy for paretic propulsion because it can also be affected by non-paretic hip flexion during swing. The study did not measure kinetic or kinematic data, so the underlying forces and joint movements were not assessed. Some participants had typical baseline values, which may have hidden improvements, especially for step length symmetry.

No competing interests are declared. The supplied publication types indicate NIH extramural research support.

Do not conclude that maximal-speed training improves paretic propulsion or walking efficiency. Non-paretic step length was only an imperfect proxy, no kinetic measures were collected, and there was no no-training control group.