The technology behind the finding is optical coherence tomography angiography, or OCTA, a dye-free imaging method that has transformed how clinicians visualize the living microcirculation. Unlike traditional fluorescein angiography, which requires an intravenous injection and captures only the superficial retinal layers, OCTA detects motion contrast from flowing red blood cells to build high-resolution, three-dimensional maps of distinct vascular layers. In a matter of seconds, it can quantify vessel density in the superficial capillary plexus and the deep capillary plexus, and measure the foveal avascular zone, the small central patch of the macula that normally contains no capillaries at all.

Across the seven included studies, the most robust and statistically consistent signal was a reduction in vessel density within the deep capillary plexus. Reported p-values for this layer ranged from below 0.001 to 0.045, while changes in the superficial capillary plexus were weaker or absent, with p-values between 0.051 and 0.230. In parallel, several independent cohorts reported an enlarged foveal avascular zone in patients compared with matched healthy controls, with significance levels between 0.01 and 0.05. Because the foveal avascular zone is a well-validated surrogate marker of capillary non-perfusion in diseases such as diabetic retinopathy, its enlargement points to early capillary dropout and incipient central macular ischemia.

Why should a liver condition leave fingerprints in the eye? The authors frame the answer around the concept of a liver-eye axis, rooted in the fact that the retinal and systemic microcirculations share embryological origins, anatomical architecture, and physiological regulatory mechanisms. Metabolic liver disease is no longer viewed as a purely hepatic problem; it is a multisystem disorder in which the leading cause of death is not liver failure but cardiovascular events. The liver, in this view, acts as an active endocrine and immunological organ rather than a passive depot for fat.

According to the pathophysiological model laid out in the review, the excessive intrahepatic accumulation of free fatty acids triggers lipotoxicity and the systemic release of pro-inflammatory cytokines such as TNF-alpha and IL-6, along with pro-coagulant factors and reactive oxygen species. This circulating milieu sustains a state of chronic low-grade inflammation which, together with peripheral insulin resistance, impairs the activity of endothelial nitric oxide synthase and reduces the bioavailability of nitric oxide. The resulting endothelial dysfunction undermines microvascular autoregulation, promotes pericyte apoptosis, and ultimately produces the capillary rarefaction that shows up on OCTA as reduced vessel density.

The selective vulnerability of the deep capillary plexus has a plausible anatomical explanation. The superficial plexus is fed directly by primary retinal arterioles and is supported by a robust smooth muscle cell layer, whereas the deep plexus functions as a hemodynamic watershed zone: a fine, intricate capillary network that lacks smooth muscle actin and depends entirely on local perfusion pressure gradients. When systemic endothelial dysfunction raises microvascular resistance, the deep plexus is the first to experience subtle hypoperfusion. The consistent deep-layer rarefaction observed across cohorts suggests that microvascular injury in metabolic liver disease begins in the most hemodynamically fragile corners of the vascular system.

The clinical implications could be considerable. MASLD is the most prevalent chronic liver condition worldwide and can progress silently from simple steatosis to nonalcoholic steatohepatitis, cirrhosis, and hepatocellular carcinoma, while its systemic vascular burden often goes undetected until end-organ disease manifests. Because the retina offers a non-invasive, easily accessible window into the body’s microcirculation, the review proposes that OCTA metrics—particularly deep capillary plexus vessel density and foveal avascular zone area—could serve as novel biomarkers for identifying patients at high risk of systemic vascular complications. The authors envision interdisciplinary collaboration between hepatologists and ophthalmologists, using OCTA screening to stratify vascular risk, monitor longitudinal disease progression, and evaluate the microvascular effects of emerging systemic therapies for MASLD.

Such enthusiasm, however, must be tempered by the limitations the authors themselves enumerate. The evidence base is small: only seven cross-sectional or retrospective studies with roughly 520 participants survived a screening process that began with 142 database records. The review protocol was not prospectively registered in PROSPERO, which the authors flag as a methodological shortcoming. Cross-sectional designs cannot establish temporal or causal relationships between liver disease progression and retinal microvascular decline, and because MASLD frequently coexists with hypertension, dyslipidemia, and subclinical dysglycemia, the observed retinal changes cannot be definitively attributed to hepatic steatosis alone, even though some primary studies adjusted for these factors.

Technical heterogeneity further complicates the picture. The included studies used different OCTA platforms—including Optovue AngioVue, Zeiss Cirrus HD-OCT, and Heidelberg Spectralis—with varying scan dimensions of 3 by 3 millimeters or 6 by 6 millimeters and different proprietary segmentation algorithms. These differences, along with projection artifacts and the substantial physiological interindividual variability of foveal avascular zone area, which is also influenced by ocular axial length and image signal quality, made formal mathematical meta-analysis potentially misleading. The authors therefore chose a semi-quantitative synthesis that summarizes the direction and consistency of findings rather than pooling effect sizes, and they caution that the apparent layer-specific involvement of the deep plexus remains a hypothesis rather than an established physiological selectivity in the absence of formal comparative analyses.

What the review ultimately delivers is a coherent observational pattern rather than a proven clinical tool: metabolic liver disease is significantly associated with early, subclinical retinal microvascular alterations, prominently characterized by deep capillary plexus rarefaction and foveal avascular zone enlargement, detectable before visually significant ocular symptoms develop. The authors call for large-scale, prospective, longitudinal studies using standardized OCTA protocols to validate whether retinal microvascular metrics can genuinely predict cardiovascular and systemic outcomes in this population. Until then, the findings stand as an intriguing piece of evidence that a fatty liver and a thinning web of deep retinal capillaries may be two visible ends of the same systemic vascular story.

Subject of Research: Retinal microvascular changes measured by OCTA in metabolic liver disease

Article Title: Retinal microvascular alterations assessed by OCTA in metabolic liver disease: insights into systemic microvascular dysfunction—a systematic review and semi-quantitative synthesis

Article References: Wang, X., & Meng, Z.-Q. (2026). Retinal microvascular alterations assessed by OCTA in metabolic liver disease: insights into systemic microvascular dysfunction—a systematic review and semi-quantitative synthesis. Frontiers in Medicine, 13, 1897327. https://doi.org/10.3389/fmed.2026.1897327

Image Credits: AI Generated

DOI: 10.3389/fmed.2026.1897327

Keywords: MASLD, NAFLD, OCTA, retinal microvasculature, deep capillary plexus, foveal avascular zone, endothelial dysfunction, liver-eye axis, systematic review, vessel density, cardiovascular risk, macular ischemia