Objective: To review the pathogenesis of triglycerides (TG), triglyceride-rich lipoproteins (TRL) and their remnants as well as their role in the development of atherosclerosis and clinical relevance for assessing residual ASCVD risk.
Materials and methods: A literature search was conducted using keywords relevant to the topic and their combinations. A targeted PubMed search prioritized English-language guidelines, consensus statements, randomized clinical trials, cohort studies, and meta-analyses published between 2019 and 2025, with older landmark publications included where necessary. Reference lists of key articles were also manually screened to identify additional relevant studies.
Results: The literature shows that elevated TG concentrations are associated with a higher risk of ASCVD, regardless of LDL-C. Given that no clear signal indicates which features of TRL give rise to risk of ASCVD, plasma TG levels remain a reasonable surrogate marker for risk assessment. Evidence suggests that the atherogenicity of TRL is driven primarily by the cholesterol carried within TRL and their remnants rather than TG themselves. TRL remnants may be at least as atherogenic as, and potentially more atherogenic than, low-density lipoproteins (LDL) because they can be taken up by intimal macrophages without prior oxidative or structural modification, persist longer within the intima, are larger in size, and carry more cholesterol per particle. They also more effectively promote foam-cell formation and contribute to low-grade inflammation. Given the conflicting results of large-scale randomized clinical trials, it is not yet possible to conclude that lowering TG concentrations alone reduces the ASCVD risk.
Conclusions: Elevated plasma TG concentrations are associated with an increased risk of ASCVD, independent of LDL-C levels. Given the ongoing need to identify the most reliable metric for risk stratification, TG remain a rational surrogate marker for estimating residual ASCVD risk related to TRL and their remnants.

This work is licensed under a Creative Commons Attribution 4.0 International License.