Background and aims:The prevalence of cognitive impairment is increasing, especially in South Korea and South America. It is hypothesized that the alanine transaminase-to-aspartate aminotransferase ratio (ALT/AST), a marker of liver function, may influence cognitive function. However, as of now, there is no consensus on the clinical evidence supporting the link between ALT/AST ratio and cognitive impairment.
Methods: Data from two cycles (2011-2014) of the National Health and Nutrition Examination Survey (NHANES) were utilized to investigate the correlation between serum ALP levels and cognitive impairment. Cognitive function was assessed through three tests: the Consortium to Establish a Registry for Alzheimer's Disease (CERAD), Animal Fluency (AF), and Digit Symbol Substitution Test (DSST) scores. The inflection point for cognitive impairment was defined as CERAD < 5, AF < 14, and DSST < 34. Weighted multiple logistic regression models were applied to explore the relationship between ALT/AST ratio and cognitive impairment. Generalized additive models (GAM) and a smooth curve fit (penalty spline method) were employed to examine potential nonlinear associations. Subgroup analyses and interaction tests were also carried out.
Results: The study enrolled a total of 2765 adults aged 60 years and older. Our findings revealed a significant correlation between ALT/AST levels and a reduced likelihood of cognitive impairment in the CERAD test results [fully adjusted odds ratio (OR) = 0.53, 95% confidence interval (CI): 0.34-0.83]. Specifically, in the highest quartile (Q4) of CERAD scores, this association remained significant [fully adjusted Q4: OR = 0.67, 95% CI: 0.50-0.88], with a trend significance (p = 0.012). A notable association was observed between ALT/AST levels and a decreased probability of cognitive impairment in the DSST test outcomes [fully adjusted OR = 0.46, 95% CI: 0.28-0.77]. Particularly, in Q3 [fully adjusted OR = 0.69, 95% CI: 0.50-0.94] and Q4 [fully adjusted OR = 0.60, 95% CI: 0.44-0.82], displaying a significant trend across quartiles (p = 0.004). The generalized additive models (GAM) and smooth curve analyses revealed a nonlinear relationship between serum ALT/AST levels and cognitive impairment. Additionally, inflection points for ALT/AST were determined as 1.29 (CERAD test), 0.63 (AF test), and 0.87 (DSST test), respectively.
Significant interactions were noted between ALT/AST and cognitive impairment concerning the CERAD test, taking into account race, education, and alcohol (P < 0.05). Moreover, significant interactions were observed between ALT/AST and cognitive impairment in connection with the CERAD test, considering age, race, alcohol, and hypertension (P < 0.05). Lastly, significant interactions were identified between ALT/AST and cognitive impairment associated with the CERAD test, accounting for alcohol, vigorous work activity, high cholesterol, and hypertension (P < 0.05).
Conclusions: In conclusion, our study unveils a non-linear relationship and a threshold effect between serum ALT/AST levels and cognitive impairment. Validating these results necessitates large-scale prospective clinical trials utilizing rigorous methodologies and comprehensive datasets.