A recent study has found that nonhospitalized patients with long COVID—particularly those who experienced persistent neurologic symptoms—exhibited significant reductions in retinal capillary perfusion.

Using optical coherence tomography angiography, researchers in the Journal of Imaging article observed a decrease in vessel length density in the deep capillary plexus, which suggested potential microvascular involvement in the pathogenesis of long COVID.

This prospective cohort study, conducted at Northwestern University by lead study author Clayton E. Lyons and colleagues, recruited 30 nonhospitalized patients with long COVID and neurologic symptoms that lasted at least 6 weeks (Neuro-PASC), and 44 healthy controls. All of the participants underwent comprehensive ophthalmologic evaluations, including optical coherence tomography angiography (OCT-A) imaging. Neuro-PASC patients had previously tested positive for SARS-CoV-2 through reverse transcription–polymerase chain reaction, antigen, or serology tests. Those with prior COVID-19–related hospitalization or preexisting retinal conditions were excluded.

The researchers measured retinal foveal avascular zone, vessel density (VD), and vessel length density (VLD) in both the superficial capillary plexus (SCP) and deep capillary plexus (DCP). They also controlled for age, sex, race, body mass index, and common vascular risk factors such as hypertension and diabetes.

Neuro-PASC patients exhibited a statistically significant decrease in DCP VLD compared with healthy controls (P = .005), whereas no statistically significant differences were observed in DCP VD or SCP metrics. Greater reductions in DCP VLD correlated with increased severity of neurocognitive symptoms, which suggested that microvascular dysfunction may contribute to neurologic manifestations of long COVID.

However, retinal structural integrity, including foveal and parafoveal thickness, was preserved, and unlike other vascular pathologies affecting the retina, SCP parameters remained unchanged.

“Our results indicate that the reduction in DCP VLD in Neuro-PASC patients may reflect an ischemic state in the deep layer secondary to decreased capillary perfusion,” the study authors wrote.

The reduced DCP VLD observed in Neuro-PASC patients aligned with hypotheses that systemic inflammation and endothelial dysfunction may contribute to long COVID pathophysiology. Given the retina's role as a window to systemic vascular health, the findings suggested that retinal OCT-A imaging may serve as a noninvasive biomarker for small vessel disease in long COVID.

The researchers noted that their findings might also support hypotheses of central nervous system damage that may occur with SARS-CoV-2 infections. "Our findings of reduced capillary perfusion may support [the] hypothesis [that] systemic inflammation due to persistent infection and/or antigenic stimulation in PASC patients [causes] persistently elevated cytokine and chemokine levels. These inflammatory mechanisms may cause endothelial dysfunction in the brain, lungs, heart, intestines, kidneys and liver,” the study authors continued.

The study's relatively small sample size limited the generalizability of its findings, as well as its power to identify correlations between OCT-A metrics and NIH Toolbox and PROMIS measures. The researchers found no correlations between COVID-19 vaccines and retinal microvasculature. While potential asymptomatic or prior infections in controls may have introduced variability into the study, “young, asymptomatic patients with a prior history of SARS-CoV-2 infections have been shown to be no different than healthy controls on OCT-A,” they emphasized.

The researchers underscored: “The correlation of OCT-A parameters with antigens, cytokines, and antibodies suspected to play a role in PASC may help elucidate the specific mechanism causing microvascular drop-out in these patients. Future studies characterizing the relationship between longitudinal changes in patient symptoms, OCT-A parameters, and plasma biomarkers of mitochondrial and metabolic dysfunction known to be altered in PASC are warranted and may establish OCT-A as an independent biomarker capable of assisting in the diagnosis, management, and prognostication of Neuro-PASC."

The researchers also suggested stratification by subtype and initial infection severity in future research to increase understanding of the pathophysiologic causes of each PASC subtype, and to develop targeted treatments for future patients.

A full list of disclosures can be found in the published study.