| Literature DB >> 30768927 |
Ljiljana Jovanovic1, Vesna Subota1, Milena Stavric1, Bojana Subotic2, Boris Dzudovic2, Natasa Novicic2, Jovan Matijasevic3, Milica Miric3, Sonja Salinger4, Natasa Markovic-Nikolic5, Maja Nikolic6, Vladimir Miloradovic6, Ljiljana Kos7, Tamara Kovacevic-Preradovic7, Jelena Marinkovic8, Nikola Kocev8, Slobodan Obradovic9.
Abstract
Factors associated with provoked PE may influence a biomarker's predictive value for the primary outcome. The aim of this study was to investigate the value of BNP, cTnI, CRP and D-Dimer measurements taken soon after hospital admission for the prediction of 30-day PE-caused death in patients with spontaneous versus provoked PE.Data were extracted from a pool of 726 consecutive PE patients enrolled in the multicenter Serbian PE registry. Blood concentrations of BNP, cTnI, CRP and D-dimer were measured during the first 24 h of hospitalization. BNP blood level had strong predictive value for the primary outcome in spontaneous PE (c-statistics 0.943, 95% CI 0.882-1.000, p = .001) and a slightly lower predictive outcome in provoked PE (c-statistics 0.824, 95% CI 0.745-0.902, p < .001). NRI and IDI showed that none of the markers, when added to BNP, could improve Cox regression prediction models for 30-day PE-related mortality in either the spontaneous or provoked PE group. Blood levels of BNP measured during the first 24 h of hospital admission had an excellent predictive value for 30-day PE-related mortality in spontaneous PE and slightly lower predictive value in provoked PE, whereas CRP, cTnI and D-Dimer did not contribute significantly to the predictive value of BNP in either group.Entities:
Keywords: Biomarkers; Brain natriuretic peptide; Mortality; Pulmonary embolism
Mesh:
Substances:
Year: 2019 PMID: 30768927 DOI: 10.1016/j.cca.2019.02.008
Source DB: PubMed Journal: Clin Chim Acta ISSN: 0009-8981 Impact factor: 3.786