Tai Chi Improves Brain Functional Connectivity and Plasma Lysophosphatidylcholines in Postmenopausal Women With Knee Osteoarthritis: An Exploratory Pilot Study

Figures

• 

  Figure 1. Effect of TC on pain (A), physical function, and stiffness parameters (B). Data are presented as boxplots and scores from the same subject is connected with a gray line. Statistical significance of differences was determined at an alpha level adjusted for multiple comparisons (i.e., Benjamin-Hochberg adjustment). WOMAC: Pain subscale from 7 (no pain) to 35 (extreme pain); Stiffness subscale from 2 (no stiffness) to 10 (extreme stiffness); Physical function subscale from 17 (no difficulties with activity of daily living) to 85 (extreme difficulties with activity of daily living). Overall WOMAC scale from 26 (best possible score) to 130 (worst possible score) due to knee pain. VAS (visual analog scale) from 0 (no knee pain) to 10 (unbearable pain). BPI (Brief Pain Inventory) from 0 (no pain) to 10 (unbearable pain).

• 

  Figure 2. A principal component analysis (PCA) plot of the plasma metabolites concentration in the subjects (n = 10) before and after TC intervention. Samples are colored based on the time-point (before or after intervention) and samples from the same subject are connected by a black line.

• 

  Figure 3. Boxplot shows the concentration of amino acids metabolites before and after TC intervention. Paired samples are linked by a line for each subject. All shown metabolites were significantly decreased after intervention (adjusted p < 0.05). ADMA, asymmetric dimethylarginine; Ala, alanine; c4-OH-Pro, cis-4-hydroxyproline; Lys, lysine; Met, methionine; SDMA, symmetric dimethylarginine.

• 

  Figure 4. Boxplot shows the concentration of lipids metabolites before and after TC intervention. Paired samples are linked by a line for each subject. All shown metabolites were significantly increased after intervention (p < 0.05).

• 

  Figure 5. Transverse sections of the brain depicting regions within the mPFC that showed a post- vs. pre-TC intervention increase in rs-fMRI connectivity in with the (A) left amygdala (orange-yellow) and (B) right amygdala (64) after 8-week TC intervention. The functional maps have been registered on the mean of T1-weighted structural images of all participants. An increase in rs-fMRI connectivity was observed between bilateral mPFC and the left amygdala [n = 7, p = 0.355, (A)] and right amygdala [n = 7, p = 0.282; (B)] seed regions with 8 weeks of TC intervention. The images represent within-subject pre-post comparisons. After correcting to maintain family-wise error rate at 0.05, none of the observed changes remained statistically significant.

• 

  Figure 6. Cross correlations between post- vs. pre-TC intervention changes in amygdala-mPFC rs-fMRI / DTI connectivity and the corresponding changes in pain/physical function. The left lower half of the figure depicts univariate scatter plots with Loess regression lines. The diagonal of the figure depicts probability distribution functions of each variable. The right upper half of the figure presents the Spearman's rho (ρ) of each univariate correlation, the corresponding unadjusted p-value (p) and the p-value adjusted for using the Benjamini-Hochberg procedure (p'). LAF, L/amygdala-mPFC functional connectivity; RAF, R/amygdala-mPFC functional connectivity; LAD, L/amygdala-mPFC DTI connectivity; RAD, R/amygdala-mPFC DTI connectivity; Physical, physical function.