In the present study, we used SD-OCT to compare inner-, outer-, and full- retinal thickness and volume of the recovered COVID-19 patients with healthy control subjects. We found unremarkable incremental patterns in macular retinal thickness and volume in patients with a history of COVID-19. Although we failed to reach statistical significance, it seems that there is a trend toward retinal thickening in almost all sectors and different retinal layers of patients with COVID-19.
Recently, Marinho and colleagues reported hyperreflective bands in inner retinal layers in patients recovered from COVID-19 (4). Beside SD-OCT findings, they have also reported fine Cotton-Wool Spots (CWS) and microhemorrhages along the retinal arcades in four patients, suggesting an inflammatory or ischemic process. However, they did not report the detailed quantitative measures of the retinal thickness parameters. Moreover, they had no control group and their sample size was limited. In a later editorial on their correspondence, Vavvas et al. raised some concerns on the interpretation of the OCT and fundus findings in COVID-19 patients (5). They suggested that as CWS were very subtle and could be due to some comorbid conditions, the results provided by Marinho et al. should be interpreted with caution. In this study, we reported the detailed SD-OCT quantitative data of the retinal thickness in patients with a recent history of COVID-19 and compared our data with a healthy control group. Interestingly, we observed unremarkable changes in retinal thickness which may be in line with the theory suggested by Vavvas et al (5).
SARS-CoV-2 replication will be initiated after binding to epithelial cells in the nasopharynx and the nasal cavity. The virus propagates and migrates down the respiratory tract and triggers an innate immune response. ACE2 has been found as the main receptor for SARS-CoV-2 (16). ACE2 is an enzyme placed in cell membranes of type II alveolar cells of the lung, enterocytes of the small intestine, arterial and venous endothelial cells, and arterial smooth muscle cells in most organs (17). ACE2 counterbalances the activity of the ACE by reducing the amount of angiotensin-II and increasing angiotensin (1–7) (18). ACE2 has been found in human, rodent, and porcine retina. Moreover, ACE has been reported to present in the choroid and different cell types of the retina, including Müller cells, ganglion cells, retinal vascular endothelial cells, and photoreceptor cells (3). Hence, retina and choroid are well anticipated to be potential targets of SARS-CoV-2 infection (19).
In addition, subtle increasing in the thickness of retinal layers may be related to the minimal extra-cellular or intra-cellular fluid accumulation. Vinores et al. performed an experimental study on murine coronavirus infection in susceptible mice (BALB/c) and found that blood-retinal barrier breakdown occurs in the early phase of infection, which coincide with the onset of inflammation and indicates that blood-retinal barrier breakdown is primarily due to inflammation rather than to retinal cell destruction (20). On the other hand, apoptosis plays an important role in many ocular pathologies, such as glaucoma, retinitis pigmentosa, and diabetic retinopathy (21). Wang et al. showed that murine coronavirus induced a biphasic retinal disease in adult mice (22). The early phase was associated with inflammation, including retinal vasculitis and viral replication and the late phase was associated with retinal degeneration. Based on these pieces of evidence, we hypothesize that retinal thickening in COVID-19 patients may happen in two phases: an initial inflammatory phase with increased vascular permeability followed by a late apoptotic phase with infiltration of the retina by inflammatory cells. Although there is no direct evidence which associate the virus with retinal changes, as the virus has been found in retinal autopsy samples and
Our study had some limitations. First, we could not obtain OCT images during the acute phase of the disease. Second, our study had a relatively small sample size which might be the reason for insignificance findings. Third, since the control individuals were selected retrospectively from a large sample of an eligible cohort, selection bias might have occurred during the identification of the control arm. However, the mean thickness measures in our control group showed negligible difference with the previously published study in Iran which might be due to different measuring devices and sample ethnicity (23). A larger-scale study during the acute phase of the disease, followed by repeated exams at fixed intervals, would provide valuable information about the impact of COVID-19 on the retina.