Markers for cardiovascular disease risk prediction in patients with coronary artery disease

Abstract

In patients with stable coronary artery disease (CAD), conventional risk factors provide limited incremental predictive value for cardiovascular events. This thesis sought to investigate whether a panel of biomarkers from different pathways – alone or combined with conventional risk factors – would exhibit incremental value in the prediction of major adverse cardiovascular events (MACE). In addition, this research investigated the predictive value of visit-to-visit blood pressure variability (BPV) for MACE in patients with CAD.

In the first study, baseline serum adiponectin, adipocyte fatty acid-binding protein (A-FABP), lipocalin-2, fibroblast growth factor (FGF)- 19 and 21, retinol-binding protein 4, and plasminogen activator inhibitor-1 were measured in the discovery cohort of 1166 CAD patients. After a median follow-up of 35 months, 170 patients developed new-onset MACE. The model with age, type 2 diabetes mellitus (T2DM), and hypertension had an area under the curve (AUC) of 0.68 for predicting MACE. The AUC further increased to 0.75 when a combination of lipocalin-2, A-FABP, and FGF-19 was added to yield an age-biomarker-clinical risk factor model. Lipocalin-2, A-FABP, and FGF-19 levels remained independent predictors of MACE after adjustment for age and clinical risk factors. In the validation cohort of 1262 CAD patients with T2DM, the age-biomarkers-clinical risk factor model was confirmed to provide good discrimination and calibration over the conventional risk factors alone for prediction of MACE. In the second study, baseline plasma high-sensitivity troponin I (hs-TnI) and B-type natriuretic peptide (BNP) were measured in 2275 patients with stable CAD. After a median follow-up of 51 months, 402 patients experienced new-onset MACE. Elevated hs-TnI and BNP levels were independently associated with MACE after adjustment for variables of a risk factor model of age, sex, T2DM, and hypertension. The risk factor model had an AUC of 0.64 for MACE prediction. The AUC significantly increased to 0.68 by the addition of hs-TnI to the risk factor model. Among patients without T2DM, the addition of each biomarker yielded greater predictive accuracy than in T2DM patients, with the AUC further increased to 0.75 when a combination of hs-TnI and BNP was added to the model. In the last study, visit-to-visit BPV was calculated in 1140 patients with stable CAD. During the follow-up period, 192 patients experienced a new MACE. Higher systolic BPV and diastolic BPV were independently associated with MACE after adjustment for variables of a risk factor model (age, sex, T2DM, hypertension, antihypertensive agents, number of BP measurements) and mean BP. The risk factor model had an AUC of 0.70 for prediction of MACE. Adding systolic/ diastolic BPV to the risk factor model with mean BP significantly improved model performance, with the AUC further increased to 0.73/0.72. The association of BPV in CAD patients without T2DM with subsequent risk for MACE was stronger than in those with T2DM. Taken together, these findings support the potential utility of the multi-biomarker approach as well as the visit-to-visit BPV for risk stratification in patients with stable CAD.