We want to thank Peng et. al for the very thoughtful and important comments on our previous publication. They postulated that the utilization of systemic glucocorticoids might exert an influence on endothelial health in individuals previously afflicted with SARS-CoV-2 infection, particularly those treated in the acute phase with high doses of corticosteroids such as dexamethasone. Given existing literature suggesting the potential impact of high-dose corticosteroids on endothelial function, this hypothesis warrants further investigation.

In our cohort study, 4 patients (9.8%) were found to still be taking systemic corticosteroids at the time of analysis. 2 patients (4.8%) had received hospital treatment involving high-flow oxygen, indicating a potential indication for dexamethasone administration (with one patient fulfilling both criteria). Unfortunately, the precise administration of systemic glucocorticoids during the hospitalization phase among these patients was not documented. Nevertheless, given that the majority of our cohort experienced a mild acute infection, it is reasonable to estimate that only a minority received systemic glucocorticoid treatment during acute SARS-CoV-2 infection.

To address this knowledge gap, we conducted a comparative analysis between patients who are either currently using low doses of steroids or had received systemic glucocorticoid treatment and those with no history of glucocorticoid usage.

A total of five patients (mean age 47.0 years ± 7.9, 60.0% female) were identified as having received systemic glucocorticoid treatment (GC). When compared to glucocorticoid-naïve post-COVID syndrome (PCS) patients, those who had received GC showed a significantly more severe acute SARS-CoV-2 infection. Additionally, GC patients displayed higher C19-YRS scores, which is an established measure of PCS symptom severity. Although not reaching statistical significance, there was a trend towards elevated levels of markers associated with acute and chronic infection, such as ferritin and IL-6, in GC patients. Notably, there was also a tendency towards increased levels of vascular endothelial growth factor (VEGF) in GC patients (Table 1).

To partially answer the question, we compared retinal vessel analysis (RVA) parameters in PCS patients with GC use versus those without. We found no significant differences in venular (vFID) or arteriolar dilation (aFID) between the two cohorts.

For static retinal analysis, the central retinal arteriolar equivalent (CRAE) tended to be lower in GC patients (179.1 ± 16.3 vs. 168.4 ± 18.2, p = 0.19), although this difference did not reach statistical significance. Similarly, the central retinal venular equivalent (CRVE) also showed a tendency to be lower in GC patients (213.9 ± 16.7 vs. 200.2 ± 9.3, p = 0.081), but again did not reach statistical significance. Given that both CRAE and CRVE were lower in GC patients, the arteriovenous ratio (AVR) was not significantly altered in patients using glucocorticoids (0.84 ± 0.07 vs. 0.84 ± 0.08, p = 0.97) (see Fig. 1).

Parameters of retinal microvasculature in PCS patients with and without systemic CS use. Boxplots show arteriolar dilatation (aFID) and venular dilation (vFID) (a and b) and SVA parameters CRAE, CRVE and AVR (c–e) in patients without CS use (blue) and with CS use (green). Mean values are shown as a rectangle and median values as a line. To compare groups Wilcoxon rank sum test was used for skewed data and Welch´s t-test for normally distributed data

As noted by Peng et al., potential confounders such as arterial hypertension, obesity, age, gender, and nicotine abuse may influence retinal microcirculation. After controlling for these confounders, our observations still did not show a significant association with systemic glucocorticoid use (see Table 2 a, b, and c).

.