Abstract
PURPOSE: To test the hypothesis that inflammation measured by white blood cell count (WBC) and C-reactive protein (CRP) is associated positively with incident heart failure (HF). METHODS: Using the Atherosclerosis Risk in Communities (ARIC) Study, we conducted separate Cox proportional hazards regression analyses for WBC (measured 1987-1989) and CRP (measured 1996-1998) in relation to subsequent heart failure occurrence. A total of 14,485 and 9,978 individuals were included in the WBC and CRP analyses, respectively. RESULTS: There were 1647 participants that developed HF during follow-up after WBC assessment and 613 developed HF after CRP assessment. After adjustment for demographic variables and traditional HF risk factors, the hazard ratio (95% confidence interval) for incident HF across quintiles of WBC was 1.0, 1.10 (0.9-1.34), 1.27 (1.05-1.53), 1.44 (1.19-1.74), and 1.62 (1.34-1.96), p trend < .001; hazard ratio across quintiles of CRP was 1.0, 1.03 (0.68-1.55), 0.99 (0.66-1.51), 1.40 (0.94-2.09), and 1.70 (1.14-2.53), p trend .002. Granulocytes appeared to drive the relation between WBCs and heart failure (hazard ratios across quintiles: 1.0, 0.93 [0.76-1.15], 1.26 [1.04-1.53], 1.67 [1.39-2.01], and 2.19 [1.83-2.61], p trend < .0001), whereas lymphocytes or monocytes were not related. CONCLUSIONS: Greater levels of WBC (especially granulocytes) and CRP are associated with increased risk of heart failure in middle-aged adults, independent of traditional risk factors.