Nomograms Based on Non-High-Density Lipoprotein to Predict Outcomes in Patients with Prior Coronary Artery Bypass Grafting with Acute Coronary Syndrome: A Single-Center Retrospective Study

Abstract

Non-high-density-lipoprotein cholesterol (non-HDL-C) is a secondary therapeutic target in cardiovascular diseases and is used for residual risk assessment in patients with coronary artery syndrome (ACS). This study was designed to determine the association between non-HDL-C in patients with prior coronary artery bypass graft (CABG) with ACS and clinical outcomes. We retrospectively analyzed 468 patients with prior CABG with ACS and categorized them into two groups based on the median non-HDL-C level. The primary endpoints were major adverse cardiovascular events (MACEs), including cardiovascular death and recurrent myocardial infarction. Kaplan-Meier curves, Cox proportional-hazard regressions, and restricted cubic splines were used to determine the association between non-HDL-C and MACEs. The discrimination and reclassification of the nomogram based on non-HDL-C were assessed using time-dependent receiver operating characteristic (ROC) curves and net reclassification improvement (NRI). During the average follow-up time of 744.5 days, non-HDL-C was independently associated with the occurrence of MACEs (hazard ratio [HR] = 5.01, 95% confidence interval [CI] = 1.65-15.24; p = 0.005) after adjusting for other lipid parameters. The spline curves indicated a linear relationship between non-HDL-C and MACEs (p-nonlinear: 0.863). The time-dependent areas under the ROC curves of prior-CABG-ACS nomograms containing non-HDL regarding MACEs in two consecutive years were 91.7 (95% CI: 85.5-97.9) and 91.5 (95% CI: 87.3-95.7), respectively. The NRI analysis indicated that the prior-CABG-ACS model improved the reclassification ability for 1- and 2-year MACEs (22.4% and 7%, p < 0.05, respectively). Non-HDL is independently associated with the risk of MACEs in patients with prior CABG with ACS. The prior-CABG-ACS nomogram based on non-HDL-C and five convenient variables generates valid and stable predictions of MACE occurrence.