Ocular MRI Findings in Patients with Severe COVID-19: A Retrospective Multicenter Observational Study

Introduction

Coronavirus disease 2019 (COVID-19) is a pandemic infection caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), negatively affecting various organs such as the central nervous system (1,2).

COVID-19 has been reported to be associated with ophthalmologic abnormalities, such as conjunctivitis, chemosis, retinopathy, or optic neuritis (3–7). MRI orbital abnormalities have been reported but no MRI studies have been published about globe abnormalities to the best of our knowledge to date (8–10).

This case series reports a series of patients with severe COVID-19 presenting with abnormal MRI findings of the globe.

Materials and Methods

This retrospective observational multicenter study was initiated by the French Society of Neuroradiology. It included clinical and imaging data of consecutive patients presenting with severe COVID-19 who underwent brain MRI from 16 hospitals including 11 university hospitals and five general hospitals from March 4, 2020, to May 1, 2020. The study was approved by the ethical committee of Strasbourg University Hospital (CE-2020–37).

Clinical and ophthalmologic data were extracted from the patients’ electronic medical records. Virologic assessment was performed with quantitative real-time reverse transcription polymerase chain reaction tests for SARS-CoV-2 nucleic acid performed on nasopharyngeal or lower respiratory tract swabs and cerebrospinal fluid. Imaging studies were conducted with either 1.5-T or 3.0-T MRI.

MRI examinations were anonymized and analyzed by two neuroradiologists with 9 years (A.L.) and 30 years (F.H.) of experience in ophthalmologic imaging

Results

Study Population

A total of 129 patients (43 of 129 [33%] women and 86 of 129 [67%] men; mean age ± standard deviation, 63 years ± 14) were included for analysis from March 4th to May 1st, 2020 (Fig 1). Among them, nine of 129 (7%) patients (one of nine [11%] women and eight of nine [89%] men; mean age, 56 years ± 13) had abnormal MRI findings of the globe consisting of the presence of one or several nodules of the posterior pole of the globe. Two of nine (22%) patients had diabetes, six of nine (67%) had obesity, two of nine (22%) had hypertension, and none of them had asthma.

Figure 1:

Download as PowerPointOpen in Image Viewer

Eight of nine (89%) patients were hospitalized in the intensive care unit (ICU) with severe COVID-19. All ICU patients presented with acute respiratory distress syndrome, with a median Simplified Acute Physiology Score II of 45 (interquartile range, 18.5). All ICU patients were intubated on high-flow supplementary oxygen, curarized, and sedated. Seven of nine (78%) patients were placed in the prone position. One of nine (11%) was placed on extracorporeal membrane oxygenation. Three of nine (33%) patients underwent dialysis. Median duration of hospitalization in the ICU was 17 days (interquartile range, 7 days).

Six of nine (67%) patients underwent brain MRI due to delayed awakening despite discontinuation of sedation, two of nine (22%) due to persisting agitation and confusion after awakening, and one of nine (11%) for agitation and hallucinations.

MRI Findings

The nodules of the posterior pole of the globe were all hyperintense at fluid-attenuated inversion-recovery–weighted imaging, isointense at T1-weighted imaging, and showed no enhancement after gadolinium-based contrast agent injection. No abnormal susceptibility was seen on susceptibility-weighted images. All nine of nine (100%) patients had nodules in the macular region, eight of nine (89%) had bilateral nodules, and two of nine (22%) had nodules outside the macular region (Fig 2).

Figure 2a:

Download as PowerPointOpen in Image Viewer

Figure 2b:

Download as PowerPointOpen in Image Viewer

Two of nine (22%) patients had diffuse extensive white matter hyperintense lesions at fluid-attenuated inversion-recovery–weighted imaging, with restricted diffusion, without postcontrast enhancement and disseminated multiple microhemorrhages at susceptibility-weighted imaging. Two of nine (22%) had diffuse nonconfluent multifocal white matter hyperintense lesions at fluid-attenuated inversion-recovery–weighted imaging without restricted diffusion or postcontrast enhancement. One of nine (11%) had multiple infarcts in the anterior cerebral artery territory. One of nine (11%) had a frontal hematoma. One of nine (11%) had multiple microhemorrhages of the splenium of the corpus callosum. One of nine (11%) had leptomeningeal enhancement without parenchymal abnormalities. Brain MRI was unremarkable for one patient (11%). Five of nine (56%) patients had bilateral fluid filling of the mastoid cells. No patient had optic nerve, optic chiasm, or optic tract abnormalities.

Correlations between MRI and Ophthalmologic Findings

Three of nine (33%) patients underwent a fundoscopy showing no abnormality of the posterior pole. One of nine (11%) underwent an optical coherence tomography showing no abnormality. One of nine (11%) had an occlusion of a central retinal artery branch at fluorescein angiography. One of nine (11%) presented with keratitis of the left eye.

Discussion

This case series reports a series of patients with severe coronavirus disease 2019 (COVID-19) presenting with abnormal MRI findings of the globe. This study showed that 7% of patients with severe COVID-19 presented with one or several nodules of the posterior pole of the globe. This rate is in line with the prevalence of 5.5% of ocular manifestations among patients with COVID-19 reported in a recent meta-analysis (3). Patients affected by severe COVID-19 were reported to be more at risk to develop ocular manifestations (4,11,12).

Nodules were mostly bilateral and were located in the macular region in all cases, in association with extramacular nodules in 22% of the cases. These nodules were not visible in the three patients who underwent ophthalmologic examination. This might be due to a lack of sensitivity of the clinical examination, which was difficult to perform in patients with severe COVID-19, or due to the delay between the completion of the MRI examination and the ophthalmologic examination.

No retinal involvement could be detected at fundoscopy in a single retrospective series of patients with severe COVID-19 (13). However, viral RNA of SARS-CoV-2 has been reported to be detectable in the retina of deceased patients with COVID-19. Moreover, three recent reports (14–16) showed that retinal findings, such as hemorrhages, cotton wool spots, dilated veins, or tortuous vessels, could be possibly associated with COVID-19 in humans. Two studies (14,15) displayed findings from the optical coherence tomography, which is a more sensitive tool than is fundoscopy. They reported hyperreflective lesions at the level of ganglion cell and inner plexiform layers at optical coherence tomography, as well as the presence of cotton wool exudates that are markers of vascular disease severity. Interestingly, these abnormalities were more prominent at the papillomacular bundle in both eyes, which was also the preferential location for the nodules visible at MRI in our study. Only one of our patients with a macular nodule underwent optical coherence tomography, showing no abnormal findings.

Although the nature of these nodules remains unknown, several hypotheses might be considered because human and animal coronaviruses were reported to cause inflammation of varying ocular segments, causing retinitis, choroiditis, retinal detachment, or optic neuritis in the literature (17,18). A wide variety of mechanisms have been described, such as direct infiltration of the retina and retina pigment epithelium by the virus, vasculitis, or autoimmune process, the latter mechanism being more frequent to explain retinal damages (17). The onset of an ocular microangiopathic syndrome is highly prevalent after viral infections. SARS-CoV-2 is known to target the angiotensin-converting enzyme 2 (ACE2) receptor, which is largely expressed in the retina and the choroid. ACE2 is the primary enzyme of the vasoprotective axis of the renin–angiotensin system in the retina (19). Its downregulation might induce the development of retinal ischemia (15). Another possible hypothesis is a Valsalva retinopathy secondary to orbital proptosis, which can occur when the ocular venous system is affected by the increased central venous pressure, leading to an inadequate ocular venous drainage. This situation can occur in patients hospitalized in ICUs, especially those in the prone position or intubated, which was the case for almost all our patients with severe COVID-19 (20,21).

In clinical practice, our study suggests that a dedicated exploration of the globes with high-spatial-resolution three-dimensional T2-weighted imaging and postcontrast fat-suppressed T1-weighted MRI, as well as fundoscopy and optical coherence tomography, should be considered in all patients presenting with severe COVID-19 to help detect posterior pole nodules. Severe eye problems might largely go unnoticed because these patients are often treated in ICUs for much more severe life-threatening conditions. Our data support the need for a screening and follow-up of these patients to provide appropriate treatment and to improve the management of potentially severe ophthalmologic manifestations.

Our study had several limitations. First, it was a retrospective study with a limited number of patients and no control group. Second, we analyzed ocular abnormalities at brain MRI performed in patients with severe COVID-19 infections. Dedicated ocular MRI sequences, especially high-spatial-resolution MRI sequences, were unavailable, preventing a precise description and understanding of ocular abnormalities. Third, ophthalmologic data were missing for the majority of our patients, because no systematic ophthalmologic examination was performed in our patients with severe COVID-19 who were hospitalized in the ICU; thus, clinical-radiologic correlations were limited. Fourth, no ocular or conjunctival sampling for detection of SARS-CoV-2 was available; thus, it is impossible to assess the direct causality of the virus in our findings. We do not know whether a high viral load was linked to more severe ophthalmologic findings. Fifth, pathologic material was lacking. Finally, we lacked follow-up data to assess whether the MRI abnormalities were transient or not and their consequences on visual acuity.

In conclusion, this case series reports a series of patients with severe coronavirus disease 2019 presenting with abnormal MRI findings of the globe. Screening of these patients might be suitable to provide appropriate treatment and to improve the management of potentially severe ophthalmologic manifestations.

Disclosures of Conflicts of Interest: A.L. disclosed no relevant relationships. F.C. disclosed no relevant relationships. F.L. disclosed no relevant relationships. S.K. Activities related to the present article: disclosed no relevant relationships. Activities not related to the present article: is a board member of Bayer; received payment for lectures including service on speakers’ bureaus from Roche and Bayer. Other relationships: disclosed no relevant relationships. F.H. disclosed no relevant relationships.

Author Contributions

Author contributions: Guarantors of integrity of entire study, A.L., F.L., F.H.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; agrees to ensure any questions related to the work are appropriately resolved, all authors; literature research, all authors; clinical studies, all authors; experimental studies, A.L., F.C.; statistical analysis, A.L.; and manuscript editing, A.L., F.C., F.L., S.K.