Integrated Oculomics and Lipidomics Reveal Microvascular Metabolic Signatures Associated with Cardiovascular Health in a Healthy Cohort

Inamullah, Ernesto Elias Vidal Rosas, Imran Razzak, Shoaib JameelarXiv 2025 · arXiv · OpenReview

Abstract

Cardiovascular disease (CVD) remains the leading global cause of mortality, yet current risk stratification methods often fail to detect early, subclinical changes. Previous studies have generally not integrated retinal microvasculature characteristics with comprehensive serum lipidomic profiles as potential indicators of CVD risk. In this study, an innovative imaging omics framework was introduced, combining retinal microvascular traits derived through deep learning based image processing with serum lipidomic data to highlight asymptomatic biomarkers of cardiovascular risk beyond the conventional lipid panel. This represents the first large scale, covariate adjusted and stratified correlation analysis conducted in a healthy population, which is essential for identifying early indicators of disease. Retinal phenotypes were quantified using automated image analysis tools, while serum lipid profiling was performed by Ultra High Performance Liquid Chromatography Electrospray ionization High resolution mass spectrometry (UHPLC ESI HRMS). Strong, age- and sex-independent correlations were established, particularly between average artery width, vessel density, and lipid subclasses such as triacylglycerols (TAGs), diacylglycerols (DAGs), and ceramides (Cers). These associations suggest a converging mechanism of microvascular remodeling under metabolic stress. By linking detailed vascular structural phenotypes to specific lipid species, this study fills a critical gap in the understanding of early CVD pathogenesis. This integration not only offers a novel perspective on microvascular metabolic associations but also presents a significant opportunity for the identification of robust, non-invasive biomarkers. Ultimately, these findings may support improved early detection, targeted prevention, and personalized approaches in cardiovascular healthcare.

Introduction

Cardiovascular disease (CVD) is the leading cause of death globally, yet current diagnostic tools often detect vascular damage only after it has become clinically significant. Standard clinical markers, such as blood pressure and basic cholesterol panels (LDL/HDL), frequently fail to capture the subtle, subclinical changes occurring years before a major cardiac event.

Recent research suggests that the retinal microvasculature can serve as a non-invasive window into the systemic vascular state. However, the integration of retinal structural phenotypes (oculomics) with high-resolution molecular profiles of blood lipids (lipidomics) has rarely been performed at scale, especially in healthy populations. This study addresses this gap by identifying asymptomatic biomarkers in a healthy cohort to enable earlier, personalized cardiovascular prevention.

Methods

The researchers developed an "imaging-omics" framework to bridge computer vision and molecular biology using data from the Human Phenotype Project (HPP), involving over 7,000 healthy adults.

  1. Oculomics: Retinal microvascular traits were extracted from 7,068 fundus images using AutoMorph, an open-source deep learning pipeline. Key features included vessel width, density, and tortuosity.
  2. Lipidomics: Serum samples from 6,321 participants were analyzed using Ultra-High Performance Liquid Chromatography-Mass Spectrometry (UHPLC-ESI-HRMS) to quantify 187 distinct lipid species, including triacylglycerols (TAGs), diacylglycerols (DAGs), and ceramides (Cers).
  3. Statistical Analysis: Partial correlation analysis was used to adjust for age and sex, with Benjamini-Hochberg (FDR) corrections applied to identify robust associations between retinal traits and lipid profiles.

Results

The Primacy of Arterial Traits

A key finding of the study is that artery-based features are significantly stronger indicators of metabolic health than venous features. Artery Average Width emerged as a central hub of lipid sensitivity, showing over 40 statistically significant correlations with various lipid species. As shown in Figure 1, artery width significantly declines with age (r = -0.22), whereas vein width (Figure 2) remains remarkably stable (r = 0.01).

Artery Average Width vs Age and Sex. Artery width shows a significant negative correlation with age (r = -0.22).
Vein Average Width vs Age and Sex. Unlike arteries, vein width shows almost no correlation with age (r = 0.01).

Microvascular–Metabolic Signatures

The study identified a consistent inverse relationship between pro-atherogenic lipids and retinal structure. Specifically, higher levels of long-chain TAGs and DAGs were associated with narrower retinal arteries and reduced vessel density—a process known as rarefaction. For instance, Artery Average Width showed strong negative correlations with species like TAG 55:8 (r = -0.10) and DAG 34:2 (r = -0.10).

Demographic Influences

The researchers observed that retinal artery narrowing with age is more pronounced in females than in males, potentially reflecting hormonal influences on vascular tone during the menopausal transition. Interestingly, vein average width showed positive correlations with certain ceramides (e.g., Cer d18:0/c16:0, r = 0.15), which may represent a compensatory physiological response to perfusion deficits in the arterial network.

Discussion and Future Work

By linking detailed vascular structural phenotypes to specific lipid species, this study provides a granular view of early CVD pathogenesis. These "eye-blood" signatures offer a much more detailed assessment of risk than standard lipid panels, highlighting the role of specific subclasses like ceramides in vascular stiffness.

However, several limitations remain. The cross-sectional design prevents the establishment of definitive causality between lipid changes and retinal remodeling. Additionally, the analysis did not account for potential confounders such as diet, physical activity, or medication use. Future longitudinal studies are essential to determine if these microvascular-metabolic signatures can accurately predict the long-term development of cardiovascular disease. Moving forward, employing non-linear machine learning models may further uncover complex, network-level interactions between the eye and the systemic metabolome.

Summary updated

Abstract

Cardiovascular disease (CVD) remains the leading global cause of mortality, yet current risk stratification methods often fail to detect early, subclinical changes. Previous studies have generally not integrated retinal microvasculature characteristics with comprehensive serum lipidomic profiles as potential indicators of CVD risk. In this study, an innovative imaging omics framework was introduced, combining retinal microvascular traits derived through deep learning based image processing with serum lipidomic data to highlight asymptomatic biomarkers of cardiovascular risk beyond the conventional lipid panel. This represents the first large scale, covariate adjusted and stratified correlation analysis conducted in a healthy population, which is essential for identifying early indicators of disease. Retinal phenotypes were quantified using automated image analysis tools, while serum lipid profiling was performed by Ultra High Performance Liquid Chromatography Electrospray ionization High resolution mass spectrometry (UHPLC ESI HRMS). Strong, age- and sex-independent correlations were established, particularly between average artery width, vessel density, and lipid subclasses such as triacylglycerols (TAGs), diacylglycerols (DAGs), and ceramides (Cers). These associations suggest a converging mechanism of microvascular remodeling under metabolic stress. By linking detailed vascular structural phenotypes to specific lipid species, this study fills a critical gap in the understanding of early CVD pathogenesis. This integration not only offers a novel perspective on microvascular metabolic associations but also presents a significant opportunity for the identification of robust, non-invasive biomarkers. Ultimately, these findings may support improved early detection, targeted prevention, and personalized approaches in cardiovascular healthcare.
BibTeX
@misc{inamullah2025integrated,
  author = {Inamullah and Ernesto Elias Vidal Rosas and Imran Razzak and Shoaib Jameel},
  title = {{Integrated Oculomics and Lipidomics Reveal Microvascular Metabolic Signatures Associated with Cardiovascular Health in a Healthy Cohort}},
  year = {2025},
  eprint = {2507.12663},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2507.12663},
}
Figure