Thromboembolic complications of SARS-CoV-2 infection are a major problem, especially in severe cases. Pulmonary embolism is the most common thromboembolic complication related to SARS-CoV-2 infection and it is also one of the causes of mortality. In severe cases, the prevalence of pulmonary embolism is reported as 2.6%, the prevalence of stroke is reported to be up to 3% and venous thromboembolism is reported to be up to 20%.7,8 Herein, we investigated the impact of non-severe SARS-CoV-2 infection on retinal microcirculation by OCTA, 144.6±82.2 days after the first positive PCR test.
The study group (Group 1) and control group (Group 2) were age and sex matched and revealed no significant differences in BCVA and refractive errors. Corroborating these data, Costa et al.9 did not report any significant difference in the prevalence of myopia and hyperopia among infected and non-infected individuals, compared to prevalence studies in the general population.
Regarding the anterior segment findings, Costa et al.9 reported the presence of cataracts in two eyes with the diagnosis known before the infection; no other pathologies or signs of uveitis were observed. In the current study, cataract was the only anterior segment change detected in both groups; additionally, there was no significant difference in the incidence of cataracts between the groups.
In the current study, intraocular pressure measurements were similar in both groups. Costa et al.9 reported a modest but statistically significant higher IOP in critical COVID-19 patients compared to severe and moderate cases (approximately 14.00 mmHg vs 12.00 mmHg). However, a large proportion of patients with critical infection received systemic corticosteroid treatment (48% of the cases evaluated),9 potentially leading to such a low difference in IOP.
None of the eyes evaluated in the current study showed any pathologies in fundus evaluation in both groups. Since none of the participants in the study or control groups had any systemic diseases, this was foreseeable. Bypareddy et al10 also reported the lack of any fundus pathologies in non-severe COVID-19 patients, except a single streak of retinal hemorrhage in one eye. Costa et al9 reported the frequency of diabetic retinopathy as 52.7%. As mentioned previously, all of our cases had a history of non-severe COVID-19 infection and none of them had any chronic diseases such as diabetes or hypertension. Costa et al9 reported the presence of yellowish-white dots in the outer retinal layers in two patients, both of whom had critical disease.99 Zago Filho et al11 also previously reported yellowish-white lesions in macular area of a COVID-19 patient in both eyes; however, these lesions were located in posterior hyaloid surface, inner plexiform and ganglion cell layers.
We observed significantly lower macular vessel densities of both SCP and DCP in all quadrants, except for parafoveal region in Group 1 (recovered non-severe COVID-19 patients) compared to Group 2 (healthy controls). In parafoveal regions of SCP and DCP, VD was slightly lower in Group 1 compared to Group 2; this difference approached but did not reach statistical significance (p=0.07 for SCP and p=0.06 for DCP). None of the patients in the current study were hospitalized during the acute phase of the infection and many of them were asymptomatic (41.8%). Guemez- Villahos et al12 also reported that VD was reduced in COVID-19 patients compared to healthy controls. The same study also reported that COVID-19 patients with and without thromboembolic events related to infection did not show any significant difference in VD. Interestingly, Turker et al13 reported that VD was significantly lower in the parafoveal region of COVID-19 patients compared to controls, but the foveal region was not reported to be affected. Of note, the study included patients who were hospitalized during the acute phase of the infection, which suggests the presence of moderate or severe disease. Cennamo et al14 reported that VD was significantly lower in both foveal and parafoveal regions of SCP and DCP. This study also included patients who were hospitalized for COVID-19 pneumonia, which again might indicate a moderate to severe infection.
The parafoveal region was found to be either not affected or the least affected in the current study, while the foveal and perifoveal regions were found to be significantly affected. The parafoveal region is richest in VD of SCP and DCP, therefore it is likely to be relatively better preserved.15,16 The choriocapillaris flow area was significantly larger in COVID-19 patients in the current study, similar to other studies in the literature.13
In the current study, all patients were evaluated after a mean duration of 144.6±82.2 days (range, 30-270 days) of the acute phase of the infection and a positive PCR test. In statistical analysis, we did not observe any correlations between VD of SCP or DCP and the duration from SARS-CoV-2 infection to ocular examination. This suggests that the impact of COVID-19 on macular capillary perfusion might be permanent. To the best of our knowledge, this is the first study to investigate this correlation.
SARS-CoV-2 infection causes immune dysfunction and vascular endothelial injury, resulting in systemic microangiopathy. SARS-CoV-2 related inflammation was thought to be the primary cause of endothelial dysfunction; however, studies suggest that endothelial cells lack ACE2 expression, which is required for the virus to infect a cell directly.17,18 On the other hand, studies also indicate that the virus itself can directly infect endothelial cells in the lungs.19,20 More recent studies have shown that SARS-CoV-2 can directly infect mature vascular endothelial cells both in vivo and ex vivo.21 It was also reported that infection with SARS-CoV-2 can lead to the degradation of the glycocalyx, the protective layer of vascular endothelial cells, and causes increased levels of hyaluronan a major component of glycocalyx in the systemic circulation.22 Significant damage of the glycocalyx layer was reported to correlate with the severity of the disease and increased levels of hyaluronan was reported to be associated with endothelial barrier dysfunction.22
Irrespective of whether the etiology is related to infection-associated inflammation or a direct entry of the virus into the vascular endothelial cells, data from the current study suggests that COVID-19 can lead to a reduction in VD of retinal capillary plexus, even when the disease is mild or asymptomatic. Studies with longer follow up time are needed to confirm whether this reduction in VD is irreversible or not.