Prognostic performance of ceramide-based risk scores for cardiovascular events and mortality in patients with coronary artery disease and heart failure: a systematic review and dose–response meta-analysis

Conventional lipid biomarkers do not fully capture cardiovascular risk. We, therefore, synthesized the evidence on ceramide-based risk scores and evaluated their dose–response associations with major adverse cardiovascular events (MACE) and mortality in predominantly coronary artery disease (CAD) populations, with limited data from heart failure (HF) cohorts. We systematically searched PubMed, Embase, Scopus, Web of Science, and ProQuest databases, from inception to April 6, 2026. Cohort studies evaluating Cardiovascular Event Risk Tests 1 and 2 (CERT1 and CERT2) in patients with CAD or HF were included. The primary outcomes were MACE or cardiovascular mortality. Random-effects meta-analyses were conducted for continuous (per 1-SD increase) and categorical (highest versus lowest quartiles) exposures. Dose–response meta-analyses employed linear, quadratic, and restricted cubic spline models. Subgroup analyses explored heterogeneity according to diabetes prevalence, follow-up duration, population type, and study quality. Small-study effects were explored using Egger’s and Begg’s tests, with trim-and-fill applied when indicated. Ten studies comprising 15 cohorts and 43,659 participants met the inclusion criteria, of which eight studies were quantitatively pooled. Per 1-SD increase, higher CERT1 was associated with cardiovascular mortality (HR: 1.25; 95% CI 1.07–1.46) and MACE (HR: 1.17; 95% CI 1.07–1.29). CERT2 was also associated with cardiovascular mortality (HR: 1.51; 95% CI 1.24–1.84) and MACE (HR: 1.19; 95% CI 1.07–1.32), with a numerically larger pooled estimate for cardiovascular mortality than that observed for CERT1. Dose–response analyses supported a positive linear association between CERT2 and MACE. For CERT2 and cardiovascular mortality, the three-knot restricted cubic spline model had the lowest AIC, suggesting possible curvature; however, the test for non-linearity was not statistically significant (P = 0.104). Higher ceramide-based risk scores were associated with greater risks of cardiovascular mortality and MACE in predominantly CAD-centered populations. These scores may capture prognostic information not fully reflected by conventional lipid biomarkers, but their incremental clinical utility remains uncertain and requires evaluation in studies reporting discrimination, calibration, and reclassification. Future studies should clarify score-specific thresholds, potential non-linear associations, and clinical relevance across more diverse populations.