Effects of the herbal prescription Kami Guibi-tang on brain function in amnestic mild cognitive impairment: a task-based and resting-state fMRI study

Amnestic mild cognitive impairment represents a prodromal stage of Alzheimer’s disease characterized by episodic memory deficits and subtle alterations in brain function. Although the traditional herbal formula Kami Guibi-tang (KGT) has been reported to exert cognitive benefits, its neural effects have not been evaluated using functional neuroimaging. In this randomized, double-blind, placebo-controlled trial, 84 individuals with amnestic mild cognitive impairment received either KGT or placebo for 24 weeks. A total of 73 participants (36 in the KGT group and 37 in the placebo group) completed the intervention and were included in the final analyses. All participants underwent functional magnetic resonance imaging during a face-name association task and a working memory task, as well as resting-state scans. Task-based imaging data were analyzed using voxel-wise and region-of-interest approaches, and associations between changes in task-related activation and cognitive performance were also examined. Resting-state functional connectivity was analyzed to examine regional connectivity changes. At baseline, the placebo group showed greater activation in memory- and attention-related regions, including the posterior cingulate cortex, frontal areas, and thalamus. These voxel-wise differences were no longer observed after the intervention. Region-of-interest analyses using linear mixed-effects models revealed significant group × time interactions in several frontal, cingulate, and thalamic regions, with activation declining in the placebo group but remaining relatively preserved or showing modest increases in the KGT group. Resting-state analyses further demonstrated increased connectivity between the left precuneus and right posterior cingulate cortex following KGT administration. These findings suggest that KGT may modulate brain activity and connectivity in regions involved in episodic and working memory. While preliminary, these results provide neuroimaging-based evidence for its potential effects on memory-related neural processes in the early stages of cognitive decline.