By: Prof. Dr. Seyed Saeid Zamanieh Shahri, MD and Prof. Dr. Sonia Sayyedalhosseini, MD

2) Complementary Indicators

In recent literature, much attention has been given to indicators beyond LDL, including:

• Non-HDL as an indicator of all ApoB-containing lipoproteins

• Apoprotein B (ApoB) to estimate the number of atherogenic particles

• Lipoprotein(a) as an independent risk factor

• Assessment of LDL and HDL particle size and number using NMR techniques

These indicators allow for a more precise depiction of the “quality” of the lipid profile.

3) Evaluation of Secondary Causes

To differentiate between primary and secondary types, additional tests are used to assess thyroid function, blood glucose, kidney and liver indices, and, if necessary, adrenal hormones.

4) Imaging Tools in Outcome Assessment

Although these methods are not for direct detection of blood lipids, they are used to estimate the “atherosclerotic burden,” such as:

• Coronary calcium score

• Carotid intima-media thickness (Carotid IMT)

• Advanced plaque imaging using CT or MRI

These methods are used in cardiovascular studies to link dyslipidemia with structural vascular changes.

Latest developments in the field of technology and technology in disease:

1) Lipidomics and NMR of lipoproteins

Quantitative NMR technologies and LC-MS/MS-based lipidomics enable the measurement of hundreds of lipoprotein and lipid features in a single sample. Studies from 2024–2025 show that combining NMR data (for particle size and number) with lipidomics profiles can provide a much more precise map of lipid metabolism. These data are used to discover new subgroups of dyslipidemia as well as to understand the relationship between molecular patterns and the risk of cardiovascular diseases.

2) Genomics and broad FH screening

Recent research is moving towards using extensive gene sequencing (whole-gene/panel sequencing) to diagnose different types of FH. Studies from 2025 have shown that in a significant portion of individuals with the clinical FH phenotype, no specific mutation is found in the three classic genes (LDLR, APOB, PCSK9), and adding other genes such as LDLRAP1 and genes associated with FH-like disorders can increase diagnostic yield.

Additionally, polygenic risk scores, which assess the simultaneous effect of Hundreds of genetic variants are considered and are becoming a research tool for estimating individual susceptibility to dyslipidemia.

3) Artificial Intelligence and Machine Learning in Lipidology Recent reviews have examined the role of artificial intelligence (AI), machine learning (ML), and deep learning (DL) in the field of lipids. Some key areas include: • Predictive models for cardiovascular disease risk that integrate not only lipid data but also lipidomic, proteomic, imaging, and clinical information simultaneously into the algorithm.

• Algorithms that predict LDL levels based on total cholesterol, HDL, and triglyceride data, which have been compared in 2025 studies with classical LDL calculation methods (formulas). • ML models for predicting the occurrence of hyperlipidemia in specific groups, such as individuals with HIV after starting combination antiretroviral therapy. • The use of AI in the automated screening of familial hypercholesterolemia (FH) based on laboratory patterns and electronic medical record data.

4) New Directions in HDL Research

Recent studies on the HDL lipidome and proteome indicate that HDL is not a “uniform particle,” but rather a collection of subpopulations with diverse compositions and functions. Analysis using NMR, MS, and chromatographic methods has separated various HDL subgroups based on their lipid and protein content, and these subgroups have been associated with different patterns of cardiovascular risk.

Summary: Hyperlipidemia is a concept that refers to a persistent increase in blood lipids and lipoproteins beyond reference ranges and appears in various genetic and acquired forms. Its pathophysiology is based on complex pathways of lipoprotein synthesis and metabolism, LDL receptor function, proteins such as PCSK9, and the structure of particles like Lp(a). This disorder is usually asymptomatic, but in the long term, it is associated with atherosclerosis, cardiovascular diseases, and in cases of very high triglycerides, with acute pancreatitis. Diagnosis is based on the lipid profile, supplementary markers such as ApoB and Lp(a), and in some cases, genetic assessment and vascular imaging. In rcent years, remarkable advances in lipoprotein NMR, lipidomics, genomics, and especially artificial intelligence and machine learning have provided a new perspective for the scientific study of blood lipids and have made it possible to achieve much more precise profiling of lipid status and cardiovascular risk at an individual level.End