Abstract
Objective
The Responsive Neurostimulation System (RNS) is a US Food and Drug Administration-approved closed-loop brain neurostimulation device for patients with drug-resistant epilepsy. Given the cognitive/neurobehavioral impact of epilepsy and the putative neuroplasticity with neuromodulation, we aimed to understand the effect of RNS on cognition.
Methods
We systematically searched four databases, PubMed/Medline, EMBASE, Web of Science, and Cochrane, from inception until April 2025. Primary outcomes were neuropsychological tests in global intellectual functioning and across five cognitive domains: language, memory, executive functioning, visuospatial functioning, and attention/processing speed. We also included measures evaluating depression and anxiety symptoms and quality of life in epilepsy. A random-effects meta-analysis was performed when at least two measures were available across studies, calculating the standardized mean difference (SMD) and 95% confidence interval (CI) between baseline and postoperative testing at or over 12 months.
Results
Of 2227 articles identified, we included 10 studies in our qualitative systematic review and five nonduplicated studies (373 patients) in our meta-analysis. Intellectual functioning, cognitive functioning, depression, anxiety, and quality of life either improved or remained stable across the included studies. Meta-analysis revealed nonsignificant changes in intellectual functioning (SMD = −.03, 95% CI = −.49 to .42) and two of the five cognitive domains: language (Boston Naming Test: SMD = .12, 95% CI = −.10 to .33; Verbal Comprehension Index: SMD = −.16, 95% CI = −.58 to .26) and visuospatial functioning (Perceptual Reasoning Index: SMD = .07, 95% CI = −.36 to .49). The remaining domains could not be evaluated given the limited number of studies. For mood, meta-analysis revealed a nonsignificant decrease in depression score (Beck Depression Inventory-II: SMD = −.14, 95% CI = −.33 to .06). GRADE (Grading of Recommendations Assessment, Development, and Evaluation) certainty of evidence was moderate for intellectual and visuospatial functioning and high for language and depression measures.
Significance
Our systematic review and meta-analysis suggest that RNS does not impact neuropsychological and neurobehavioral outcomes at a group level, at least for the measures examined. We review putative mechanisms by which RNS may impact cognition and offer suggestions for future research in this domain.
JUN