Marlien Pieters1, Maylene Ferreira2, Moniek P M de Maat3, Cristian Ricci4. 1. Centre of Excellence for Nutrition, North-West University, Potchefstroom, South Africa; Medical Research Council Unit for Hypertension and Cardiovascular Disease, North-West University, Potchefstroom, South Africa. Electronic address: marlien.pieters@nwu.ac.za. 2. Centre of Excellence for Nutrition, North-West University, Potchefstroom, South Africa. 3. Department of Hematology, Erasmus University Medical Centre, Rotterdam, the Netherlands. 4. Centre of Excellence for Nutrition, North-West University, Potchefstroom, South Africa; Pediatric Epidemiology, Department of Pediatrics, University Medicine Leipzig, Leipzig, Germany.
Abstract
BACKGROUND: While fibrinogen is a known cardiovascular disease (CVD) risk marker, its quantitative input to mortality risk is a topic of debate. METHODS: We investigated the contribution of fibrinogen, among that of other biomarkers, to prevalent CVD and incident CVD mortality in 4487 participants of the US National Health and Nutrition Examination Survey (NHANES). Participants were observed for a median period of 14 years, resulting in more than 58,000 person-years. RESULTS: At baseline 551 participants had CVD and during follow up, 1339 all-cause deaths occurred, 321 (24%) of which were due to CVD. Hierarchical cluster analysis and principal component analysis (PCA) were performed to derive clusters of association between biomarkers. Next, structural equation modelling was performed to investigate the association of these clusters with baseline CVD and all-cause and CVD mortality during follow-up. Mediation analysis was used to determine which biomarkers played a mediatory role between prevalent CVD and future mortality. Fibrinogen clustered with C-reactive protein only and was associated with CVD at baseline (p < 0.0001) and with all-cause (p < 0.001) and CVD (p < 0.001) mortality at follow-up. Only fibrinogen (4.7%), followed by gamma-glutamyl transferase (GGT) (3.5%) and uric acid (2.3%) were identified as possible mediators between CVD status and all-cause mortality, with fibrinogen (3.2%) and GGT (3.1%) the only mediators between CVD status and CVD mortality. CONCLUSION: This data shows that fibrinogen is not only cross-sectionally associated with CVD, but also contributes to all-cause and CVD mortality at follow-up. It furthermore appears to mediate the association between prevalent CVD and both all-cause and CVD mortality.
BACKGROUND: While fibrinogen is a known cardiovascular disease (CVD) risk marker, its quantitative input to mortality risk is a topic of debate. METHODS: We investigated the contribution of fibrinogen, among that of other biomarkers, to prevalent CVD and incident CVD mortality in 4487 participants of the US National Health and Nutrition Examination Survey (NHANES). Participants were observed for a median period of 14 years, resulting in more than 58,000 person-years. RESULTS: At baseline 551 participants had CVD and during follow up, 1339 all-cause deaths occurred, 321 (24%) of which were due to CVD. Hierarchical cluster analysis and principal component analysis (PCA) were performed to derive clusters of association between biomarkers. Next, structural equation modelling was performed to investigate the association of these clusters with baseline CVD and all-cause and CVD mortality during follow-up. Mediation analysis was used to determine which biomarkers played a mediatory role between prevalent CVD and future mortality. Fibrinogen clustered with C-reactive protein only and was associated with CVD at baseline (p < 0.0001) and with all-cause (p < 0.001) and CVD (p < 0.001) mortality at follow-up. Only fibrinogen (4.7%), followed by gamma-glutamyl transferase (GGT) (3.5%) and uric acid (2.3%) were identified as possible mediators between CVD status and all-cause mortality, with fibrinogen (3.2%) and GGT (3.1%) the only mediators between CVD status and CVD mortality. CONCLUSION: This data shows that fibrinogen is not only cross-sectionally associated with CVD, but also contributes to all-cause and CVD mortality at follow-up. It furthermore appears to mediate the association between prevalent CVD and both all-cause and CVD mortality.