Back

Aspartate/alanine aminotransferase ratio and development of chronic kidney disease in non-diabetic Japanese men and women

Okawa, Y.; Mitsuhashi, T.; Suzuki, E.

2024-05-20 epidemiology
10.1101/2024.05.20.24307032 medRxiv
Show abstract

BackgroundThe relationship between serum aspartate/alanine aminotransferase ratio (AST/ALT) and subsequent development of chronic kidney disease (CKD) in non-diabetic Asian adults has not yet been fully investigated in longitudinal studies. MethodsThe study included all middle-aged and older non-diabetic Japanese citizens who received health check-ups in Zentsuji, Kagawa, Japan (1998-2023). AST/ALT was classified into three categories: <1.0 (reference), 1.0-<1.5, and [&ge;]1.5. CKD was defined as an estimated glomerular filtration rate of <60 mL/min/1.73 m2. The Weibull accelerated failure time model was used to examine the association between AST/ALT categories and subsequent CKD onset because the proportional hazards assumption was violated. ResultsOf 6309 men and 9192 women, 2966 men and 4395 women remained in the final cohort. After a mean follow-up of 7.50 years for men and 8.34 years for women, 33.7% of men and 34.0% of women developed CKD. Women had higher AST/ALT than men. In women, a dose-response relationship was observed, with a 9% shorter survival time to CKD onset for AST/ALT [&ge;]1.5 compared with AST/ALT <1.0. In contrast, men had a shorter survival time to CKD onset by point estimates, but the 95% confidence intervals crossed 1 in all models. ConclusionsIn this study comparing the risks of CKD development in non-diabetic men and women by AST/ALT levels, a dose-response relationship was only observed in women. Differences in the distribution of AST/ALT by sex may have affected the results. Therefore, in non-diabetic Japanese women, AST/ALT may be used as an indicator of future CKD development.

Matching journals

The top 1 journal accounts for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.