The effect of experimental muscle pain on the background electrical brain activity

Abstract

The purpose of this project was to investigate whether specific effects in the background activity of the brain associated with the experience of pain can be depicted by means of quantitative electroencephalography (EEG). Lasting pain was induced by intramuscular infusion of hypertonic saline. The infusion was titrated to maintain pain for a sufficient time to obtain enough data for meaningful analysis. In a first study on 12 subjects, using a single, blind, repeated measures design with randomization of the administration of isotonic (0.9%) and hypertonic (5%) saline, and with subjects unaware of the fact that one substance was isotonic saline, a statistically significant pain response could be attributed to the administration of hypertonic saline. In a second study on 19 subjects, again using a randomized repeated measures design, topographic EEG measures were examined with respect to experimentally induced pain and pain from memory. Prior to each of these experimental stages, baseline recordings were obtained to satisfy the requirement of the crossover design. In addition to the common frequency bands used in EEG, we also obtained data in the frequency range of 35–100 Hz. The short-term variability of the selected EEG measures and their suitability as a sample estimate were assessed by computing the coefficient of variation from all selected epochs of a given subject at baseline. When compared to baseline, spectral analyzed EEG measures during experimental pain demonstrated statistically significant increases in the beta and 35–100 Hz frequency ranges, most notably at the temporal recording sites. There was no statistically significant difference between the EEG measures for (1) experimental pain vs. pain from memory, and (2) the 2 baseline recordings. The great variability in the topographical aspect of the between-subject response was interpreted as being strongly suggestive of the contamination of EEG measures by phenomena attributed to the jaw, facial and scalp musculature. In fact, Pearson correlation coefficients, as high as 0.92, were found between measures in the frequency band of 35–100 Hz and the beta frequency range. The unexplained variance in the heightened beta cortical power density can be attributed to the vigilance scanning of pain processes. Due to the fact that the statistically significant effect of pain on the topographic EEG measures were not different from imagined pain, we concluded that these effects are non-specific for pain.