Changes in urinary albumin as a surrogate for kidney disease progression in people with type 2 diabetes.

ABSTRACT

BACKGROUND: It remains unclear whether urinary albumin changes can predict subsequent kidney disease progression in people with diabetes. METHODS: This retrospective cohort study included 4570 Japanese adults with type 2 diabetes (T2D). The exposure was changes in urinary albumin-to-creatinine ratio (UACR) over 3 years, categorized into three categories ≤ - 30%, minor change, or ≥ 30%. During the exposure period, eGFR decline was also examined and categorized into two categories < 30% or ≥ 30% decline. The primary outcome was the composite of eGFR halving or initiation of kidney replacement therapy (KRT). The secondary outcome was the initiation of KRT. RESULTS: In the spline model, the hazard ratio for the primary outcome increased linearly on the log2 scale of UACR changes. When classified into six groups based on the categories of UACR changes and eGFR decline, people with a ≤ - 30% UACR change and < 30% eGFR decline had a 38% lower incidence of the outcome compared to those with a minor UACR change and < 30% eGFR decline. Meanwhile, the risk in those with a ≤ - 30% UACR change and ≥ 30% eGFR decline was 2.89 times. People with a ≥ 30% UACR change had the higher risk, regardless of whether a ≥ 30% eGFR decline occurred. Similar results were obtained in the secondary outcome. CONCLUSIONS: UACR changes can be a useful surrogate for kidney disease progression in people with T2D. However, when setting a decrease in UACR as the surrogate, it may be necessary to simultaneously evaluate kidney function decline.