High-sensitivity troponin I (hsTnI) and N-terminal pro-brain natriuretic peptide (NT-pro-BNP) are predictors of coronary artery disease. Recently, routine hematological parameters emerged as mortality predictors. We examined the predictive value of hematological parameters (from the Utrecht Patient Oriented Database) and hsTnI and NT-pro-BNP for mortality in a coronary angiography population (Utrecht Coronary Biobank n = 1913). Using Cox regression, receiver operating characteristics, integrated discrimination improvement (IDI), and continuous net reclassification improvement (cNRI) analysis, we compared the predictive properties of hematological parameters with hsTnI and NT-pro-BNP for mortality. During a median follow-up duration of 1.8 years, 77 deaths occurred. A panel of 7 hematological parameters (leukocyte count, reticulocyte mean corpuscular hemoglobin concentration, red blood cell [RBC] green (FL1) fluorescence, %neutrophils, %large [>120 fL] RBCs, %monocytes, and coefficient of variation of neutrophil complexity) was highly predictive. Added to clinical characteristics, hematological parameters (area under the curve [AUC]: 0.855, P < .001; IDI: 0.04, P = .02; cNRI: 0.41, P < .001) were better predictors than hsTnI (AUC: 0.818) or NT-pro-BNP (AUC: 0.834) alone or combined (AUC: 0.834). Hematological parameters may provide mortality risk information following coronary angiography and may be superior to hsTnI and/or NT-pro-BNP.
Keywords: biomarkers; coronary artery disease; mortality.