Abstract
OBJECTIVE: To investigate the local brain functional changes after heat-sensitive moxibustion at the left ST35 (Dubi) acupoint in patients with knee osteoarthritis (KOA) based on resting-state functional magnetic resonance imaging (rs-fMRI), and to explore the possible neuroregulatory mechanisms of heat-sensitive moxibustion for pain relief using the fractional amplitude of low-frequency fluctuation (fALFF) analysis. METHODS: A total of 30 KOA patients who were found to be insensitive to the heat of moxibustion in the non-heat-sensitive moxibustion (NHSM) group, and enrolled another 30 KOA patients with moxibustion sensation in the heat-sensitive moxibustion (HSM) group. Both groups received moxibustion at the left ST35 acupoint for 10 min (once daily for 10 consecutive days) at a distance of about 3 cm from the skin. Before the first treatment and after the tenth treatment, we assessed knee pain using visual analog scale (VAS) and performed rs-fMRI scans on the patients. The fALFF data of both groups were processed using the SPM 12 module of MATLAB software. RESULTS: Compared with pre-moxibustion, the fALFF value of the HSM group in the frontal lobe, white matter, and left temporal lobe was significantly higher, while the occipital lobe and the right hemisphere was significantly lower. The region with the highest increase was the left temporal lobe, followed by white matter, and the region with the strongest decrease was the occipital lobe, followed by the frontal lobe and the right hemisphere. In the NHSM group, the fALFF value in the left occipital lobe, left medial frontal gyrus, left middle frontal gyrus, right superior frontal gyrus, right superior temporal gyrus, and right cerebellar posterior lobe was significantly lower, with the strongest decrease in the right cerebellar posterior lobe, followed by the right superior temporal gyrus. Compared with the NHSM group after treatment, the fALFF value of the HSM group in the external nucleus, white matter, right hemisphere, left cerebellum, and left hemisphere was significantly higher, and the frontal lobe, occipital lobe, and precentral gyrus was significantly lower. Additionally, a positive correlation was found between the fALFF changes of the left temporal lobe and the VAS score changes for each patient (pre- vs. post-treatment) in the HSM group (r = 0.764, p < 0.01), whereas a negative correlation was observed for the occipital lobe (r = -0.595, p < 0.01). CONCLUSION: This study reveals that the superior pain relief from heat-sensitive moxibustion is underpinned by a sensation-specific, bidirectional modulation of the brain's pain-processing network. Unlike the generalized suppression observed in the NHSM group, the heat-sensitive state is characterized by a concerted increase in temporal lobe activity and decrease in occipital lobe activity, both changes being strongly predictive of individual clinical improvement. These results offer compelling neuroimaging evidence that the subjective heat-sensitive sensation reflects a more efficient and integrated brain state for analgesia. CLINICAL TRIAL REGISTRATION: https://www.chictr.org.cn/, ChiCTR2000033075.