We explored the prognostic value of three circulating candidate biomarkers—midregional-proadrenomedullin (MR-proADM), soluble tumor necrosis factor receptor 1 (sTNFR1), and N-terminal prohormone brain natriuretic peptide (NT-proBNP)—for change in renal function in patients with type 2 diabetes.
Outcomes were defined as renal function loss (RFL), ≥40% decline of estimated glomerular filtration rate (eGFR) from baseline, and rapid renal function decline (RRFD), absolute annual eGFR slope <–5 mL/min/year. We used a proportional hazard model for RFL and a logistic model for RRFD. Adjustments were performed for established risk factors (age, sex, diabetes duration, HbA1c, blood pressure, baseline eGFR, and urinary albumin-to-creatinine ratio [uACR]). C-statistics were used to assess the incremental predictive value of the biomarkers to these risk factors.
Among 1,135 participants (mean eGFR 76 mL/min, median uACR 2.6 mg/mmol, and median GFR slope –1.6 mL/min/year), RFL occurred in 397, RRFD developed in 233, and 292 died during follow-up. Each biomarker predicted RFL and RRFD. When combined, MR-proADM, sTNFR1, and NT-proBNP predicted RFL independently from the established risk factors (adjusted hazard ratio 1.59 [95% CI 1.34–1.89], P < 0.0001; 1.33 [1.14–1.55], P = 0.0003; and 1.22 [1.07–1.40], P = 0.004, respectively) and RRFD (adjusted odds ratio 1.56 [95% CI 1.7–2.09], P = 0.003; 1.72 [1.33–2.22], P < 0.0001; and 1.28 [1.03–1.59], P = 0.02, respectively). The combination of the three biomarkers yielded the highest discrimination (difference in C-statistic = 0.054, P < 0.0001; 0.067, P < 0.0001 for RFL; and 0.027, P < 0.0001 for RRFD).
In addition to established risk factors, MR-proADM, sTNFR1, and NT-proBNP improve risk prediction of loss of renal function in patients with type 2 diabetes.