Abstract
OBJECTIVES: Cardiovascular and renal diseases are closely related. Brain natriuretic peptide (BNP) and urinary albumin are established predictors for cardiac and renal morbidities, respectively. To date, no reports have investigated the combined predictive value of BNP and urinary albumin for long-term cardiovascular-renal events in patients with chronic kidney disease (CKD). The aim of this study was to investigate this theme. METHODS: Four hundred eighty-three patients with CKD were enrolled into this study and followed-up for 10 years. The endpoint was cardiovascular-renal events. RESULTS: During the median follow-up period of 109 months, 221 patients developed cardiovascular-renal events. Log-transformed BNP and urinary albumin were identified as independent predictors for cardiovascular-renal events, with a hazard ratio of 2.59 (95% confidence interval [CI], 1.81-3.72) and 2.27 (95% CI, 1.82-2.84) for BNP and urinary albumin, respectively. For the combined variables, the group with high BNP and urinary albumin had a markedly higher risk (12.41-times; 95% CI 5.23-29.42) of cardiovascular-renal events compared with that of the group with low BNP and urinary albumin. Adding both variables to a predictive model with basic risk factors improved the C-index (0.767, 0.728 to 0.814, p=0.009), net reclassification improvement (0.497, p<0.0001), and integrated discrimination improvement (0.071, p<0.0001) more than each of them alone. CONCLUSIONS: This is the first report to demonstrate that the combination of BNP and urinary albumin can stratify and improve the predictability of long-term cardiovascular-renal events in CKD patients.