Uric acid index is a risk for mild cognitive impairment in type 2 diabetes

Abstract

Background: A new uric acid (UA) index has recently been proposed, while serum uric acid (SUA), fasting triglyceride, and fasting blood glucose levels in the index are shown to affect cognitive function. This study aims to investigate the clinical value of the UA index for assessing mild cognitive impairment (MCI) in type 2 diabetes (T2D) patients.

Methods: This was an observational cross-sectional study with 616 participants. A generalized additive model was used to determine a linear or curvilinear relationship between cognitive performance and the UA index. Logistic regression and random forest models were both developed. A receiver operating characteristic curve (ROC) was delineated and the area under the curve (AUC) was calculated.

Results: MCI was diagnosed in 313 participants (50.81%). Compared with the T2D-normal cognitive function group, MCI subjects had higher UA indexes, lower cognitive scores, and lower education levels (p < 0.001). Generalized additive models showed the UA index and the Montreal Cognitive Assessment (MoCA) score to be decreased linearly (p < 0.001). The UA index AUC was 0.751 (95% CI = 0.713-0.789, p < 0.001). The optimal cut-off point for the identification of MCI based on the UA index was 11.26 (sensitivity: 62.3%, specificity: 75.9%). Results for females in the cohort yielded an AUC change of + 2.5%, the less-educated population (AUC change of + 4.7%), and the hypertensive population (AUC change of + 1.1%). The AUCs were 0.791 (95% CI = 0.720-0.863) for the random forest model and 0.804 (95% CI = 0.770-0.837) for the logistic regression model, and no statistical significance was found (p = 0.758).

Conclusion: This study showed that the increased UA index was independently associated with MCI in patients with T2D, especially among female, less-educated, and hypertensive patients. It could be a potential indicator of MCI in T2D patients.

Keywords: Logistic regression; Mild cognitive impairment; Random forest; Type 2 diabetes; Uric acid index.