BDNF Val66Met but not transcranial direct current stimulation affects motor learning after stroke.

BACKGROUND: tDCS is a non-invasive neuromodulation technique that has been reported to improve motor skill learning after stroke. However, the contribution of tDCS to motor skill learning has only been investigated in a small number of studies. In addition, it is unclear if tDCS effects are mediated by activity-dependent BDNF release and dependent on timing of tDCS relative to training. OBJECTIVE: Investigate the role of activity-dependent BDNF release and timing of tDCS relative to training in motor skill learning. METHODS: Double-blind, between-subjects randomized controlled trial of circuit tracing task improvement (ΔMotor skill) in 80 chronic stroke patients who underwent tDCS and were genotyped for BDNF Val66Met. Patients received either short-lasting tDCS (20 min) during training (short-lasting online group), long-lasting tDCS (10 min-25 min break - 10 min) one day before training (long-lasting offline group), short-lasting tDCS one day before training (short-lasting offline group), or sham tDCS. ΔMotor skill was defined as the skill difference on the circuit tracing task between day one and day nine of the study. RESULTS: Having at least one BDNF Met allele was found to diminish ΔMotor skill (ßBDNF,Met = -0.217 95%HDI = [-0.431 -0.0116]), indicating activity-dependent BDNF release is important for motor skill learning after stroke. However, none of the tDCS protocols affected ΔMotor skill (ßShort-lasting,online = 0.0908 95%HDI = [-0.227 0.403]; ßLong-lasting,offline = 0.0242 95%HDI = [-0.292 0.349]; ßShort-lasting,offline = -0.108 95%HDI = [-0.433 0.210]). CONCLUSION: BDNF Val66Met is a determinant of motor skill learning after stroke and could be important for prognostic models. tDCS does not modulate motor skill learning in our study and might be less effective than previously assumed.