The periventricular/periaqueductal gray (PAG/PVG) is critical for pain perception and is associated with the emotional feelings caused by pain. However, the electrophysiological characteristics of the PAG/PVG have been little investigated in humans with chronic pain. The present study analyzed the oscillatory characteristics of local field potentials (LFPs) in the PAG/PVG of eighteen neuropathic pain patients. Power spectrum analysis and neural state analysis were applied to the PAG/PVG LFPs. Neural state analysis is based on a dynamic neural state identification approach and discriminates the LFPs into different neural states, including a single neural state based on one oscillation and a combinational neural state based on two paired oscillations. The durations and occurrence rates were used to quantify the dynamic features of the neural state. The results show that the combined neural state forms three local networks based on neural oscillations that are responsible for the perceptive, sensory, and affective components of pain. The first network is formed by the interaction of the delta oscillation with other oscillations and is responsible for the coding of pain perception. The second network is responsible for the coding of sensory pain information, uses high gamma as the main node, and is widely connected with other neural oscillations. The third network is responsible for the coding of affective pain information, and beta oscillations play an important role in it. This study suggested that the combination of two neural oscillations in the PAG/PVG is essential for encoding perceptive, sensory, and affective measures of pain.
Dynamic neural state, Local field potential, Local network, Neural oscillation, Pain components, Periventricular/periaqueductal gray