Quantitative assessment of metabolic memory and its prediction of renal function decline in patients with type 2...

This study quantitatively assesses metabolic memory by modeling the relationship between hyperglycemic exposure and renal function decline in patients with type 2 diabetes (T2D).

This retrospective longitudinal study included 381 Japanese patients with T2D. Hyperglycemic exposure was presented by calculating the area under the curve (AUC) for HbA1c ≥ 6 % (AUC_{HbA1c ≥ 6 %}) during the observation period. A non-linear mixed-effects model was constructed to predict changes in estimated glomerular filtration rate (eGFR) based on AUC_{HbA1c ≥ 6 %}.

The relationship between AUC_{HbA1c ≥ 6 %} and eGFR changes was shown by a sigmoidal curve, with sex, age, diabetic retinopathy, dyslipidemia, and hypertension incorporated as covariates. The predictive utility of the model was validated using goodness-of-fit plot, visual predictive check, and bootstrap methods.

We developed an AUC_{HbA1c ≥ 6 %}-based model to predict renal function decline in patients with T2D, showing that AUC_{HbA1c ≥ 6 %} may serve as a quantitative indicator of metabolic memory.