Differences in cortical response to acupressure and electroacupuncture stimuli

Abstract

Background: FMRI studies focus on sub-cortical effects of acupuncture stimuli. The purpose of this study was to assess changes in primary somatosensory (S1) activity over the course of different types of acupuncture stimulation. We used whole head magnetoencephalography (MEG) to map S1 brain response during 15 minutes of electroacupuncture (EA) and acupressure (AP). We further assessed how brain response changed during the course of stimulation.

Results: Evoked brain response to EA differed from AP in its temporal dynamics by showing clear contralateral M20/M30 peaks while the latter demonstrated temporal dispersion. Both EA and AP demonstrated significantly decreased response amplitudes following five minutes of stimulation. However, the latency of these decreases were earlier in EA (~30 ms post-stimulus) than AP (> 100 ms). Time-frequency responses demonstrated early onset, event related synchronization (ERS), within the gamma band at ~70-130 ms and the theta band at ~50-200 ms post-stimulus. A prolonged event related desynchronization (ERD) of alpha and beta power occurred at ~100-300 ms post-stimulus. There was decreased beta ERD at ~100-300 ms over the course of EA, but not AP.

Conclusion: Both EA and AP demonstrated conditioning of SI response. In conjunction with their subcortical effects on endogenous pain regulation, these therapies show potential for affecting S1 processing and possibly altering maladaptive neuroplasticity. Thus, further investigation in neuropathic populations is needed.

Figure 1

Evoked S1 responses to Electroacupuncture (EA) and Acupressure (AP): (a) For all subjects, the primary sources localized to the contralateral SI for EA (white dots) and AP (blue dots). Sources are shown as closest points on the cortex reconstructed from subjects' individual MRIs and morphed to an average inflated brain surface (sulci/gyri are dark/light gray respectively). (b) Average time-courses for all trials of EA and AP (black, dashed lines) demonstrated analogous M20, M30, M50/M60 and M120 (those of AP being slightly delayed). Five minute sub-averages (blue, red and green lines) show attenuation of peak-to-peak amplitude over the course of the run. (c) Peaks demonstrating significant decreases (p < 0.01) following 5 minutes of stimulation are marked by a red "*" For EA this was the M30 peaks and for AP the 130 ms peak. Significant differences occurring after > 10 minutes stimulation were seen for the M30 and M50 with EA and with the M50 and M120 with AP.

Figure 2

S1 Oscillatory Activity: (a) Post-stimulus time-frequency representations showed an early ~20-70 ms low-gamma (γ) ERS. This was followed by a prolonged ERD in alpha (α) and beta (β) power at ~100-300 ms post-stimulus. There was a simultaneous theta (θ) ERS at ~50-200 ms. (b) Beta ERD at 100-300 ms was calculated for each minute of stimulation and demonstrated a trend for decrease over the course of EA stimulation. Similar calculations were made for other frequency bands but did not show significance at the same p < 0.01 level.