Investigation of volumetric alterations in thalamic subnuclei in progressive and stable mild cognitive impairment using magnetic resonance imaging

Objective: This study aimed to evaluate the volumes of thalamic subnuclei known to function in large-scale cognitive networks between progressive mild cognitive impairment (pMCI) and stable MCI (sMCI) groups progressing to dementia. Materials and Methods: Magnetic resonance imaging and clinical data of 31 pMCI (Age: 68.66±6.86; education: 15.71±2.46; gender: 15 female) and 31 sMCI (Age: 70.18±7.24; education: 16.23±2.68; gender: 13 female) patients with no statistically significant differences in age, gender, education, and follow-up interval (mean 21 months) from the Alzheimer's Disease Neuroimaging Initiative database were used. FreeSurfer software was used for individual thalamus segmentation and volume calculation. Thalamic nuclei were divided into anterior, medial, posterior, lateral, ventral, and intralaminar nucleus groups. The volumes of each nucleus were normalised using intracranial volume. Normalised volumes were compared between groups using independent samples t-test, and false discovery rate (FDR) correction was applied. Correlation analysis was performed using florbetapir (AV45) PET scores to evaluate the relationship between amyloid burden in the brain and volumetric decrease. Results: Statistically significant volumetric decreases were detected in the bilateral anterior (Right: t=2.432 pFDR=0.048; Left: t=2.327 pFDR=0.048) and lateral (t=2.372 pFDR=0.048) nucleus groups in pMCI compared to sMCI. A negative correlation was found between PET scores and bilateral anterior nucleus groups (Right: r=-0.353 p=0.026, left: r=-0.350 p=0.026). Conclusion: In our study, the anterior thalamic nucleus group, closely associated with memory, showed reduced volume in MCI patients who progressed to dementia, and this reduction correlated with amyloid burden in the brain. Based on the findings of our study, the anterior thalamic nucleus group may provide supportive value to other MRI biomarkers in predicting conversion to dementia.