A Multi-Marker Approach to Predict Incident CKD and Microalbuminuria

Traditional risk factors do not adequately identify individuals at risk for CKD. We related a multi-marker panel consisting of the following seven circulating biomarkers to the incidence of CKD and microalbuminuria (MA) in 2345 participants who attended the sixth Framingham Offspring Study examination (1995 to 1998): C-reactive protein, aldosterone, renin, B-type natriuretic peptide (BNP), plasminogen-activator inhibitor type 1, fibrinogen, and homocysteine. We defined CKD at follow-up (2005 to 2008) as estimated GFR (eGFR) <60 ml/min per 1.73 m(2); we defined MA as urine albumin-to-creatinine ratio >= 25 (women) or 17 (men) mg/g on spot urine samples. We identified a parsimonious set of markers related to outcomes adjusting for standard risk factors and baseline renal function, and we assessed their incremental predictive utility. During a mean 9.5-year follow-up, 213 participants developed CKD and 186 developed MA. In multivariable logistic regression models, the multi-marker panel was associated with incident CKD (P < 0.001) and MA (P = 0.003). Serum homocysteine and aldosterone both were significantly associated with CKD incidence, and log-transformed aldosterone, BNP, and homocysteine were significantly associated with incident MA. Biomarkers improved risk prediction as measured by improvements in the c-statistics for both CKD and MA and by a 7% increase in net risk reclassification. In conclusion, circulating homocysteine, aldosterone, and BNP provide incremental information regarding risk for incident CKD and MA beyond traditional risk factors.