Safety and Adverse Events following Non-invasive Electrical Brain Stimulation in Stroke: A Systematic Review

Abstract

Background: Noninvasive electrical stimulation (ES) could have therapeutic potential in stroke recovery. However, there is no comprehensive evaluation of adverse events. This study systematically searched the literature to document frequency and prevalence of adverse events. A secondary aim was to explore associations between adverse events and ES parameters or participant characteristics.

Methods: Databases were searched for studies evaluating ES in adults with stroke. All included studies were required to report on adverse events. Extracted data were: (1) study design; (2) adverse events; (3) participant characteristics; (4) ES parameters.

Results: Seventy-five studies were included. Adverse events were minor in nature. The most frequently reported adverse events were tingling (37.3% of papers), burning (18.7%), headaches (14.7%) and fatigue (14.7%). Cathodal stimulation was associated with greater frequency of itching (p = .02), intensities of 1–2 mA with increased tingling (p = .04) and discomfort (p = .03), and current density <0.4mA/cm2 with greater discomfort (p = .03). Tingling was the most prevalent adverse event (18.1% of participants), with prevalence data not differing between active and control conditions (all p ≥ 0.37). Individual participants were more likely to report adverse events with increasing current density (r = 0.99, p = .001). Two severe adverse events were noted (a seizure and percutaneous endoscopic gastrostomy placement).

Conclusion: ES appears safe in people with stroke as reported adverse events were predominantly minor in nature. An adverse events questionnaire is proposed to enable a more comprehensive and nuanced analysis of the frequency and prevalence of adverse events.

KEYWORDS:

• Stroke
• Transcranial Electrical Stimulation
• Safety
• Adverse Event
• Noninvasive Brain Stimulation

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

BH supported by a National Health and Medical Research Councilfellowship (1125054). TRS supported by a National Health and Medical Research Council fellowship (1141735)..