Stroke patients have shown expressive asymmetry between the paretic and non-paretic lower limb (NPLL). The aim of this study was to examine the effects of unilateral leg load used as a restraint for the NPLL during gait training on weight bearing asymmetry after stroke.

Thirty-eight stroke patients of both genders and with up to one year of first stroke that caused gait deficits (gait speed<0.8m/s) participated in the study. Participants were randomized into: treadmill training with load (equivalent to 5% of body weight) on the NPLL (experimental group) and treadmill training without load (control group). Interventions lasted 30min per day, for 2 weeks (9 sessions). Kinetic data were collected with a force plate, during standing position (static) and during gait (dynamic). Limb loading was evaluated by the mean of vertical component of ground reaction force (GRF), for paretic and non-paretic limbs. Static and dynamic symmetry ratio (paretic GRF/non-paretic GRF) was calculated for both groups at baseline, post-training and after 40-day follow-up.

At baseline, groups were homogeneous with regard to age, time since stroke, neurological status (according to National Institute of Health Stroke Scale) and walking ability (according to Functional Ambulatory Category). Repeated-measures analysis of variance showed no effects over time in static and dynamic symmetry ratio (static: F=0.985; P=0.364; dynamic: F=2.005; P=0.153). In addition, ANOVA showed no changes between experimental and control groups in static and dynamic symmetry ratio (static: F=0.031; P=0.862; dynamic: F=0.063; P=0.803).

Although load addition might favor the use of paretic limb by restraining NPLL, weight bearing asymmetry has not been changed over time in any group. Once this protocol is innovative, future studies should investigate other parameters such as training duration and amount of load, to improve weight bearing asymmetry after stroke.