Changes of cerebral functional connectivity induced by foot reflexology in a RCT

www.nature.com/scientificreports OPEN Changes of cerebral functional connectivity induced by foot reflexology in a RCT Emeline Descamps 1,2,3*, Mathilde Boussac 1,3*, Karel Joineau 1 & Pierre Payoux 1 Non-Pharmacological Interventions (NPIs) are increasingly being introduced into healthcare, but their mechanisms are unclear. In this study, 30 healthy participants received foot reflexology (FR) and sham massage, and went through a resting-state functional magnetic resonance imaging (rs-fMRI) to evaluate NPIs effect on brain. Rs-fMRI revealed an effect of both NPIs on functional connectivity with changes occurring in the default-mode network, the sensorimotor network and a Neural Network Correlates of Pain (NNCP—a newly discovered network showing great robustness). Even if no differences were found between FR and SM, this study allowed to report brain biomarkers of well-being as well as the safety of NPIs. In further research, it could be relevant to study it in patients to look for a true reflexology induced-effect dependent of patient reported outcomes. Overall, these findings enrich the understanding of the neural correlates of well-being experienced with NPIs and provided insight into the basis of the mechanisms of NPIs. Non-Pharmacological Interventions (NPIs) are non-invasive, targeted and evidence-based interventions that aim to prevent, care for, or cure individual’s health problems, as defined by the Non-Pharmacological Interventions Society (NPIS)1. Among NPIs, Foot Reflexology (FR) consists of physical stimulation of the epidermis of the feet through the application of controlled pressure movement to specific areas, called reflex zones. This concept was used to activate homeostasis2,3 and currently, FR is increasingly being introduced into healthcare to improve the physical and emotional well-being of individuals. Some studies have already reported significant positive outcomes including pain management, regardless of etiology4–11, stress or anxiety12–14 and improvement in general well-being or quality of life15–17. Despite these benefits, research based on robust methodology was sparse and the mechanisms underlying the therapeutic effects of FR also remained u ndetermined18. Hence, clarifying the indications, verifying the safety and identifying the mechanisms of FR is currently a key priority. For this purpose, functional magnetic resonance imaging (fMRI) is a powerful tool to observe experiencerelated cerebral changes as brain’s response to a stimulus. Indeed, resting state fMRI (rs-fMRI) on healthy adult volunteers allows researchers to test hypotheses about particular functional networks and the impact of specific activities (for example mindfulness meditation) on intrinsic brain c onnectivity19,20. A few previous studies reported the use of fMRI to detect brain activity in FR and have suggested a correlation between somatosensory cortex activity and the stimulation of specific reflex areas in the feet21,22. Nevertheless, to our knowledge, the examination of resting state networks associated with FR has never been done. Such research may provide information about the integrity, organization and changes of major functional systems of the brain23. Several networks of interest could be investigated in FR such as the Default Mode Network (DMN)24 associated with episodic memory and self-referential processing; the Executive Control Network (ECN); the Salience Network (SN)25; and the Sensorimotor Network (SMN) related to the sense of touch26,27. Findings from functional neuroimaging studies on these specific networks have great potential to contribute to the understanding of the mechanisms underlying the therapeutic effect of FR. We hypothesized that networks associated with attention and sensory processing would show specific changes related to FR and tactile stimulation. The aim of the present study was so to determine if two different forms of short tactile stimulation (FR and foot massage as a sham massage of FR) could change the functional connectivity of intrinsic connectivity networks, physiological parameters and well-being in healthy participants. 1 Inserm Unité ToNIC, UMR 1214, CHU PURPAN – Pavillon BAUDOT, Place du Dr Joseph Baylac, 31024 Toulouse CEDEX 3, France. 2CNRS, Toulouse, France. 3These authors contributed equally: Emeline Descamps and Mathilde Boussac. *email: ; Scientific Reports | (2023) 13:17139 | https://doi.org/10.1038/s41598-023-44325-x 1 Vol.:(0123456789) www.nature.com/scientificreports/ Results Demographic data, electrophysiological measures and well‑being assessment Fifteen females and fifteen males were included. The mean age was of 30.3 ± 5.7 years old. Table 1 presents all the demographic characteristics at baseline (t0). Every participant (n = 30) from both groups was included in all the analyses. Concerning electrophysiological measures and well-being assessment, there were no significant differences between groups at t0 for sex ratio, age, heart rate, respiratory rate, oxygen saturation nor subjective well-being (Table 1). Table 1 presents the evolution of heart rate, respiratory rate and oxygen saturation between t0 and t1 for both groups. There was a significant effect of time (intervention and control) for heart and respiratory rates and subjective well-being (p = 0.0009, p = 0.01 and p = 0.007, respectively) with no effect of groups (interaction: p > 0.05): heart rates decreased in both groups, while respiratory rates and subjective well-being increased in both groups. Imaging results Before conducting our analyses on the whole cross-sectional study, we wanted to make sure that the “washout time” (10 min) between t1 and t2 (after the first intervention and before the second one) was enough for connectivity measures to return to «baseline functional connectivity». Therefore, we looked for differences between t1 and t2 in ROI-to-ROI analyses with the CONN toolbox in each network. No significant changes were observed between t1 and t2 in all networks (DMN, SMN, SN, ECN) or NNCP, while significant connectivity changes were observed between t0 and t2 in DMN, ECN and NNCP. Altogether, these results mean that the potential connectivity change after the first intervention (FR or SM) was probably maintained at t2 and that the short “washout period” (imposed by the constant experimental run in the MRI scanner) was not sufficient to see the effect of the second intervention (either FR or SM) on brain connectivity changes. Hence, regrouping every subject from both groups according to the intervention to look for connectivity changes will not be possible since connectivity measures from the second resting state fMRI acquisition (t2) would not be neutral (not corresponding to the baseline connectivity). Subsequently, only analyses concerning the first block (t0 and t1) of the cross-sectional study were done to avoid evaluating a connectivity “contaminated” (t2–t3) by the preceding intervention (1st intervention, t1). (...truncated)