The heart knows best: baseline heart rate variability as guide to transcutaneous auricular vagus nerve stimulation in depression.

Major Depressive Disorder can be conceptualized as a chronic stress condition associated with autonomic dysregulation, including blunted heart rate reactivity, elevated basal cortisol levels and higher low-grade peripheral inflammation, pointing to an autonomic imbalance with relative sympathetic overactivation and parasympathetic attenuation. Transcutaneous vagus nerve stimulation (taVNS) offers a non-invasive method to stimulate the vagus nerve, as key component of the parasympathetic system, to restore autonomic balance. Here, we examined whether emotional (i.e., positive and negative emotions), cardiac (i.e., heart rate and heart rate variability (HRV)), and inflammatory (i.e., TNF- α, IL-6) reactivity to stress are differentially influenced by taVNS in participants with MDD and controls. Additionally, we performed a post-hoc analysis with participants stratified by baseline RMSSD, as a proxy for vagus nerve activity, to evaluate the utility of biological stratification beyond diagnostic criteria. To assess the effect of chronic stress we conducted a single-blinded, cross-over, randomized controlled trial with 110 participants (51 controls and 59 MDD participants). For the analysis stratified by RMSSD group, we grouped participants into low (n = 54) vs. high (n = 55) RMSSD groups, regardless of diagnosis. All participants were subjected to an acute stress paradigm, both with taVNS and sham stimulation on two separate days, in a counter-balanced order. There was no difference in any of the outcomes regarding the effect of taVNS in participants with MDD and controls. Analyses split by RMSSD group, however, showed that for those with low RMSSD, taVNS restored the blunted cardiac stress response and numerically decreased TNF-α levels. Unexpectedly, in participants with high RMSSD, the opposite pattern was observed: heart rate and TNF-α were significantly increased, and vagally mediated heart rate variability was significantly decreased under taVNS compared to sham stimulation. Analyses using RMSSD as continuous predictor yielded similar results. Our findings suggest that RMSSD-based stratification may be a useful tool for predicting outcome of (ta)VNS treatment. We encourage researchers with HRV data to re-evaluate their findings through RMSSD stratification.