[Central mechanism of the short-needling method for treating knee osteoarthritis based on EEG signals]

To explore the central mechanism of the short-needling method for treating knee osteoarthritis (KOA) based on electroencephalogram (EEG) signals.

Thirty-one KOA patients were recruited as the KOA group, and 31 age-, sex-, and BMI-matched healthy subjects were recruited as the healthy group. The KOA group received "short-needling" therapy at Neixiyan (EX-LE5), Dubi (ST35), Yinlingquan (SP9), Zusanli (ST36), and Liangqiu (ST34) on the affected side, once daily, 6 consecutive days as a treatment course, with a 1-day break between treatment courses, for a total of 4 treatment courses. Western Ontario and McMaster Universities osteoarthritis index (WOMAC) score was assessed before and after treatment; the levels of inflammatory cytokines (interleukin-1β [IL-1β] and tumor necrosis factor-α [TNF-α]) in synovial fluid were measured; the relative power of resting-state EEG signals (Delta, Theta, Alpha, Beta, Gamma frequency bands) and EEG network topological indexes (global clustering coefficient, global efficiency, mean degree, and network density) were analyzed in KOA patients before and after treatment and in healthy subjects.

(1) Compared before treatment, the KOA group showed significantly reduced WOMAC scores in joint pain, stiffness, physical function, and total scores after treatment (P<0.001); IL-1β and TNF-α levels in synovial fluid were also significantly reduced (P<0.001). (2) In the Theta band, compared before treatment, the relative power in 10 brain regions was decreased significantly after treatment in the KOA group (P<0.05, P<0.01); except for the prefrontal cortex, the relative power in other regions of the KOA group remained higher than the healthy group before treatment (P<0.01). In the Beta band, compared before treatment, the relative power in all regions except the prefrontal cortex increased after treatment in the KOA group (P<0.05, P<0.01); before treatment, KOA patients had lower relative power in the left/right frontal lobes, left/right parietal, central parietal region, left/right temporal, and occipital lobes compared to the healthy group (P<0.01). (3) Compared with the healthy group, the KOA group showed abnormally enhanced functional connectivity in Theta and Gamma bands in the prefrontal cortex before treatment; after treatment, this abnormal connectivity was attenuated. (4) Compared before treatment, the global clustering coefficient, mean degree, and global efficiency in the Theta band networks of the prefrontal cortex decreased significantly after treatment (P<0.05), approaching levels of the healthy group (P>0.05); all four topological metrics before treatment in the KOA group were significantly higher than those in the healthy group (P<0.01). (5) The relative Theta power in the left frontal lobe was positively correlated with the difference in WOMAC score before and after treatment (P<0.05); the relative Gamma power in the right frontal lobe was negatively correlated with the difference in WOMAC score before and after treatment (P<0.05).

The short-needling method could effectively improve joint function in KOA patients, reduce slow-wave Delta activity, increase fast-wave Beta activity, and regulate abnormally enhanced brain functional connectivity networks. These EEG changes are correlated with WOMAC score.