Background: A substantial number of clinical studies have demonstrated the functional recovery induced by the use of brain-computer interface (BCI) technology in patients after stroke. The objective of this review is to evaluate the effect sizes of clinical studies investigating the use of BCIs in restoring upper extremity function after stroke and the potentiating effect of transcranial direct current stimulation (tDCS) on BCI training for motor recovery. Methods: The databases (PubMed, Medline, EMBASE, CINAHL, CENTRAL, PsycINFO, and PEDro) were systematically searched for eligible single-group or clinical controlled studies regarding the effects of BCIs in hemiparetic upper extremity recovery after stroke. Single-group studies were qualitatively described, but only controlled-trial studies were included in the meta-analysis. The PEDro scale was used to assess the methodological quality of the controlled studies. A meta-analysis of upper extremity function was performed by pooling the standardized mean difference (SMD). Subgroup meta-analyses regarding the use of external devices in combination with the application of BCIs were also carried out. We summarized the neural mechanism of the use of BCIs on stroke. Results: A total of 1015 records were screened. Eighteen single-group studies and 15 controlled studies were included. The studies showed that BCIs seem to be safe for patients with stroke. The single-group studies consistently showed a trend that suggested BCIs were effective in improving upper extremity function. The meta-analysis (of 12 studies) showed a medium effect size favoring BCIs for improving upper extremity function after intervention (SMD = 0.42; 95% CI = 0.18-0.66; I 2 = 48%; P < 0.001; fixed-effects model), while the long-term effect (five studies) was not significant (SMD = 0.12; 95% CI = − 0.28-0.52; I 2 = 0%; P = 0.540; fixed-effects model). A subgroup meta-analysis indicated that using functional electrical stimulation as the external device in BCI training was more effective than using other devices (P = 0.010). Using movement attempts as the trigger task in BCI training appears to be more effective than using motor imagery (P = 0.070). The use of tDCS (two studies) could not further facilitate the effects of BCI training to restore upper extremity motor function (SMD = − 0.30; 95% CI = − 0.96-0.36; I 2 = 0%; P = 0.370; fixed-effects model).

No abstract

Noninvasive brain stimulation (NIBS) techniques have been used to facilitate the recovery from prolonged unconsciousness as a result of brain injury. The aim of this study is to systematically assess the effects of NIBS in patients with a disorder of consciousness (DOC). We searched four databases for any randomized controlled trials on the effect of NIBS in patients with a DOC, which used the JFK Coma Recovery Scale-Revised (CRS-R) as the primary outcome measure. A random-effects meta-analysis was conducted to pool effect sizes. Fourteen studies with 273 participants were included in this review, of which 12 studies with sufficient data were included in the meta-analysis. Our meta-analysis showed a significant effect on increasing CRS-R scores in favor of real stimulation as compared to sham (Hedges’ g = 0.522; 95% confidence interval [CI], 0.318–0.726; P < 0.0001, I2 = 0.00%). Subgroup analysis demonstrated that only anodal transcranial direct current stimulation (tDCS) of the left dorsolateral prefrontal cortex (DLPFC) significantly enhances the CRS-R scores in patients with a DOC, as compared to sham (Hedges’ g = 0.703; 95% CI, 0.419–0.986; P < 0.001), and this effect was predominant in patients in a minimally conscious state (MCS) (Hedges’ g = 0.815; 95% CI, 0.429–1.200; P < 0.001). Anodal tDCS of the left DLPFC appears to be an effective approach for patients with MCS.