Long-term effect of non-invasive brain stimulation on hemispatial neglect, functional outcomes, and mortality after stroke: ELETRON trial extend

Introduction: Our primary clinical trial indicated that anodal stimulation of the right posterior parietal region associated with specific and perceptual task training was superior to placebo in reducing stroke-induced hemispatial neglect (HN) immediately after the treatment protocol. However, our primary study did not investigate whether this benefit was maintained in the long term after stroke. Therefore, this study aimed to evaluate the long-term effects of the protocol applied in the ELETRON trial on outcomes associated with HN, functionality, and mortality. Methods: This was a pilot, multicenter, prospective, randomized, double-blind trial in patients with HN after stroke who underwent either active tDCS (anodal tDCS or C-tDCS) or sham tDCS in addition to specific and perceptual task training. The outcomes were evaluated on the last day of the session and after 1 year of stimulation (follow-up). Daily evolution rates were calculated as the difference between the values observed between the moments divided by the follow-up time for each individual. The primary outcome was the rate of HN evolution according to the BIT-C scale. The secondary outcome was CBS evolution rate. The exploratory outcomes were the evolution rate of functional disability and autonomy assessed by FIM and BI, quality of life assessed by EQ 5D, stroke severity using the NIHSS, and functional dependence assessed by mRS. Death was examined separately. A linear regression model with a time-adjusted identity link function was used to explain the evolution rates of each outcome of A-tDCS and C-tDCS as a function of sham tDCS. Survival models were adjusted to compare mortality groups. Results: The evolution rate of BIT-C was not different between A-tDCS (B = 3.18; CI: −4.84–11.19; p = 0.438) and C-tDCS (B = −0.95; CI: −8.97–7.07; p = 0.816) with sham. The secondary and exploratory outcomes showed the same pattern. In addition, there were no statistically significant differences in mortality over time between A-tDCS and S-tDCS (B = 0.322; 95 % CI 0.284–6.707; p = 0.689) and between C-tDCS and S-tDCS (B = −0.798; 95 % CI 0.063–3.195; p = 0.425). Conclusion: The benefits of A-tDCS and C-tDCS were maintained for all long-term post-stroke outcomes. Trial registration: RBR-78jvzx - Brazilian Registry of Clinical Trials (Rebec), registered on March 13, 2016.