Can Electroencephalography (EEG) Identify the Different Dimensions of Pain in Fibromyalgia?

The study was a pilot study with a cross-sectional and exploratory design carried out from 2020 to 2022. A quantitative electroencephalographic analysis was performed with women with Fibromyalgia and pain-free individuals as a control group, matched for gender and age. The sample was not probabilistic and recruitment occurred in a random manner. All volunteers were recruited from reference services in the state of Pernambuco, referred from other centers as well as through announcements in digital media.

Procedures All procedures were conducted respecting the Declaration of Helsinki (1964) and approved by the local ethical committee. Before clinical and sociodemographic evaluation, all volunteers signed a written informed consent, including all information regarding the risks and benefits of their participation in the study. During the study, all individuals with fibromyalgia were instructed not to change their medication use as well as eating habits. Clinical assessments and qEEG data acquisition took place in one single visit to the laboratory lasting around two hours. After signing the written informed consent, all volunteers were taken to an isolated room to perform an EEG evaluation. Then, they underwent sociodemographic and clinical assessments (only individuals with fibromyalgia).

EEG data acquisition and processing For each volunteer, signals were recorded using digital EEG equipment for 120 seconds in an isolated room - without any communication with the external environment - with volunteers rested, seated in a comfortable chair, and with closed eyes. Signal recording was performed through 19 Ag/AgCl electrodes positioned on the scalp following the predetermined points of the international 10-20 system of electroencephalography and, always maintaining a maximum impedance of 10 kΩ. Additionally, a ground electrode was positioned on the lateral third of the right clavicle, while two reference electrodes were positioned on the region of the right and left mastoid processes. A sampling rate for recording the 500 Hz signal was captured by the NeuronSpectrum signal amplifier and recorded by the Neuron-Spectrum/NET omega software. Additionally, the high-pass (0.5 Hz), low-pass (35 Hz), and notch (60Hz; suitable for 220V mains) filters were applied during data acquisition and processing.

Then, the collected data were pre-processed using the EEGLab toolbox in MATLAB® version R2014a software for Windows. In addition, an Independent Component analysis was performed using the Independent Components Analysis (ICA) algorithm to separate the components related to biological artifacts. The rejection of these components was done through the Multiple Artefact Rejection Algorithm (MARA) considering a 50% cutoff point. For time-frequency analysis of the relative spectral power for each epoch, the fast Fourrier transform method was used. The dominant frequency in each patient was identified in the following points of the international 10-20 EEG system: F3, F4, Fz, F7, F8 (frontal area), and C3, C4, Cz (central area) during rest. Spectral power density assessment was also performed, for each frequency band, considering the following bands: delta (0,5 a ≤ 4 Hz); theta ( > 4 a ≤ 8 Hz); alpha (> 8 a ≤13 Hz) e beta (> 13 a ≤ 30 Hz). For relative spectral power distribution calculations, the absolute spectral power of each frequency band was divided by the total power of all bands present in the 0.5-35Hz.