Association between coronary artery calcium scores and retinal ischemic perivascular lesions

TL;DR

This study finds a link between high coronary artery calcium scores and retinal ischemic lesions, suggesting retinal changes may indicate cardiovascular disease risk.

Contribution

The study establishes a novel association between retinal ischemic perivascular lesions and elevated coronary artery calcium scores as a marker of cardiovascular risk.

Findings

High-risk patients (CAC ≥ 300) had significantly more retinal ischemic perivascular lesions than low-risk patients.

High-risk patients were more likely to have elevated RIPL counts (≥ 2) compared to low-risk patients.

Known cardiovascular risk factors partially mediated the relationship between CAC scores and RIPL counts.

Abstract

Coronary artery calcium (CAC) scores measure atherosclerosis and are associated with increased cardiovascular disease (CVD) risk. Retinal ischemic perivascular lesions (RIPLs) represent focal retinal infarcts and have been linked to systemic CVD. Given their association with retinal microcirculation dysfunction, RIPLs might serve as early indicators of systemic vascular dysfunction. This study aims to assess the relationship between CAC scores and RIPLs.

Retrospective, single-institution, cross-sectional study. Patients consenting for research who had both high-quality, bilateral, macular optical coherence tomography (OCT) imaging and a CAC score of either 0 (low risk) or ≥ 300 (high risk) within one year after their macular OCT imaging. RIPLs were identified as regions of focal inner nuclear layer (INL) thinning, outer plexiform layer (OPL) inward deviation, and outer nuclear layer (ONL) expansion on OCT B-scans in the absence of underlying or overlying retinal pathology. The difference in mean total RIPLs per patient between the high- and low-risk groups was assessed. The percentages of individuals in each group with high RIPL count (≥ 2) were compared as well. The effects of potential confounders (known cardiovascular risk factors) on the relationship between RIPL counts and CAC-score-based risk group were examined with multivariate analysis.

90 patients (low-risk: 66; high-risk: 24) were included. High-risk patients were older (mean ± std of 72.1y ± 7.7 vs. 62.1y ± 10.5; p < 0.001), more likely male (46% vs. 17%; p = 0.004), and were more likely to have hypertension (79% vs. 30%; p < 0.001), type 2 diabetes mellitus (38% vs. 9%; p = 0.003), smoking history (62% vs. 22%; p = 0.021), atrial fibrillation (17% vs. 0%; p = 0.004), and lower total cholesterol (155.2 mg/dL ± 36.3 vs. 201.0 mg/dL ± 41.2; p < 0.001). Increased raw RIPL counts (5.29 ± 7.54 vs. 2.36 ± 3.52; p = 0.008) and increased probability of elevated RIPL presence (75% with ≥ 2 RIPLs vs. 48%; p = 0.025) were observed in high-risk patients.

CAC score ≥ 300 is associated with an increased number of RIPLs and an increased probability of elevated RIPL presence (≥ 2 RIPLs), potentially mediated by known CVD risk factors. RIPLs might identify those with elevated CVD risk, and RIPLs might suggest presence of established CVD risk factors.

Figures