Unravelling the Connection Between Reduced Mobility and Mild Cognitive Impairment in Chronic Pain Patients Through Cluster Analysis.

Cognitive impairment has been closely associated with systemic metabolic disorders, such as oxygen and energy metabolism disorders. We used deep learning to estimate the risk of mild cognitive impairment (MCI) in outpatients with pain (pain group) and those without pain (non-pain control group) from general blood test data reflecting systemic metabolic disorders. Univariate analysis was performed on blood test data from both groups to calculate the estimated Mini-Mental State Examination (MMSE) scores. Additionally, principal component analysis was conducted on data obtained from eight assessment batteries, comprising 12 items commonly used by Japanese pain treatment institutions. Cluster analysis was also performed using Ward's method on seven components with cumulative proportions exceeding 90%. Our results showed that patients suffering from pain had significantly lower estimated MMSE scores than controls, despite no significant age differences. Patients suffering from pain had significantly higher white blood cell, triglyceride, glucose, and potassium values and significantly lower red blood cell, hemoglobin (Hgb), and uric acid values. Notably, Hgb values were significantly lower only in men, with no significant differences in women. Cluster analysis of the assessment battery data revealed five distinct clusters (Euclidean distance: 10.14%). The average MMSE score of the cluster with an extremely low value for the first component was 27.1, whereas the scores for the other clusters fell below the MCI cutoff. The first component primarily reflects mobility, suggesting that patients suffering from pain with reduced mobility are at increased risk of mild dementia. This finding indicates that impaired mobility due to pain can promote systemic metabolic disorders, subsequently increasing the risk of MCI.