Cortical mapping of painful electrical stimulation by quantitative electroencephalography: unraveling the time–frequency–channel domain

Abstract

Lisa Goudman,1–3 Jorne Laton,4 Raf Brouns,4,5 Guy Nagels,4–6 Eva Huysmans,2,3,7,8 Ronald Buyl,7,9 Kelly Ickmans,2,3,10 Jo Nijs,2,3,10 Maarten Moens,1,2,4,11 1Department of Neurosurgery, Universitair Ziekenhuis Brussel, 2Pain in Motion International Research Group, 3Department of Physiotherapy, Human Physiology and Anatomy, Faculty of Physical Education and Physiotherapy, 4Center for Neurosciences (C4N), Vrije Universiteit Brussel (VUB), 5Department of Neurology, Universitair Ziekenhuis Brussel, 6National MS Center, 7Department of Public Health (GEWE), Faculty of Medicine and Pharmacy, Vrije Universiteit Brussel, 8Interuniversity Centre for Health Economics Research (I-CHER), 9Department of Biostatistics and Medical Informatics, Faculty of Medicine and Pharmacy, Vrije Universiteit Brussel, 10Department of Physical Medicine and Physiotherapy, 11Department of Radiology, Universitair Ziekenhuis Brussel, Brussels, Belgium Abstract: The goal of this study was to capture the electroencephalographic signature of experimentally induced pain and pain-modulating mechanisms after painful peripheral electrical stimulation to determine one or a selected group of electrodes at a specific time point with a specific frequency range. In the first experiment, ten healthy participants were exposed to stimulation of the right median nerve while registering brain activity using 32-channel electroencephalography. Electrical stimulations were organized in four blocks of 20 stimuli with four intensities – 100%, 120%, 140%, and 160% – of the electrical pain threshold. In the second experiment, 15 healthy participants received electrical stimulation on the dominant median nerve before and during the application of a second painful stimulus. Raw data were converted into the time–frequency domain by applying a continuous wavelet transform. Separated domain information was extracted by calculating Parafac models. The results demonstrated that it is possible to capture a reproducible cortical neural response after painful electrical stimulation, more specifically at 250 milliseconds poststimulus, at the midline electrodes Cz and FCz with predominant δ-oscillations. The signature of the top-down nociceptive inhibitory mechanisms is δ-activity at 235 ms poststimulus at the prefrontal electrodes. This study presents a methodology to overcome the a priori determination of the regions of interest to analyze the brain response after painful electrical stimulation. Keywords: electroencephalography, Parafac model, painful electrical stimulation, conditioned pain modulation

Keywords

• Electroencephalography
• Parafac model
• Electrical painful stimulation