Abstract
Brain oscillatory activity and its role as a biomarker in mental health and human behavioral mechanisms stands as an unexplored tool to explain the placebo effect. This systematic review and meta-analysis characterize, for the first time, the neural mechanistic biomarkers of placebo effect in neurorehabilitation. A systematic literature search based on PRISMA guidelines was conducted in PubMed, Embase, and Cochrane Library including randomized controlled trials (RCTs), and cross-over trials up to August 2023. The review included studies reporting brain oscillations in resting state as outcome in placebo-controlled neurorehabilitation trials. A qualitative, semi-quantitative, quantitative analysis was performed using minimal statistical information from each study and calculating effect sizes of absolute power differences between placebo and active intervention groups using Hedges’ g statistic method. The analysis included 63 studies: 5 (180 healthy subjects) and 58 (1758 patients with neurological disorders). In healthy population, placebo interventions showed an increase in alpha power when comparing versus no intervention (g = 0.45, 95% CI [0.09; 0.8]). In subjects with neurological disorders sham intervention showed increase in alpha frontal (g = 0.08, 95% CI [0.07; 0.08] ), alpha central (g = 0.55, 95% CI [0.47; 0.65]), alpha parietal (g = 0.28, 95% CI [0.18; 0.44]), beta central (g = 1.31, 95% CI [1.06; 1.63]), and theta central (g = 0.58, 95% CI [0.46; 0.72]). Interestingly, these effects became non-significant when comparing with the active interventions. In both populations, the enhancement of alpha oscillations in fronto-central regions emerged as the primary neural biomarker of the placebo effect. Furthermore, increase in beta and theta bands within the same brain regions were also noted as potential biomarkers, particularly in non-healthy subjects. Intriguingly, these effects were reversed when compared to active rehabilitation interventions, suggesting that active rehabilitation techniques also encompass the placebo effect. Our novel results show that placebo neural effects depend on the brain oscillatory baseline (healthy vs. patient populations), and they seem to unmask a more homeostatic salutogenic rhythm, such as alpha oscillations in healthy subjects and theta and beta oscillations in patient populations.