Original research Persistent reduction of retinal microvascular vessel density in patients with moderate and severe COVID-­19 disease Sandra Banderas García ‍ ‍,1 David Aragón,1 Brahim Azarfane,1 Fernando Trejo,1 Xavier Garrell-­Salat,1 Adrián Sánchez-­Montalvá,2 Susana Otero-­Romero,3 Jose Garcia-­Arumi,1,4 Miguel Angel Zapata1 To cite: Banderas García S, ABSTRACT Aragón D, Azarfane B, et al. Objective This study aims to analyse the possible Key messages Persistent reduction of retinal recovery or worsening in retinal microvasculature after 8 microvascular vessel density months in a previously studied COVID-­19 cohort. What is already known about this subject? in patients with moderate and ► The long-­term consequences of COVID-­19 on retinal severe COVID-­19 disease. Methods and analysis A cross-­sectional case– control study and a prospective longitudinal cohort vascularisation are still unknown. Reduced retinal BMJ Open Ophthalmology vessel density in severe COVID-­ 19 cases is de- 2022;7:e000867. doi:10.1136/ study. Participants were the subjects of our previous study who re-­enrolled for a new examination including a scribed in the literature. bmjophth-2021-000867 fundus photograph (retinography), an optical coherence What are the new findings? ► Additional supplemental tomography (OCT) scan and an OCT angiography. ► We demonstrated persistent reduction in the central material is published online COVID-­19 diagnosed patients were divided into only. To view, please visit the vascular area over time in patients with moderate three groups: group 1: mild disease, asymptomatic/ and severe COVID-­19. journal online (http://​dx.​doi.​ org/​10.​1136/​bmjophth-​2021-​ paucisymptomatic subjects who received outpatient care; 000867). group 2: moderate disease and group 3: severe disease, How might these results change the focus of both of which required hospital admission because of research or clinical practice? pneumonia. Statistical analyses were performed using ► A close follow-­up of patients with severe disease Received 30 July 2021 SPSS software (V.23.0). Cross-­sectional intergroup should be performed, in order to detect possible Accepted 3 January 2022 differences were analysed by means of analysis of long-­term systemic and ophthalmological sequelae. variance for normally distributed variables and the Kruskal-­ Wallis test for non-­normally distributed ones. In reference to the prospective part of the study (intragroup differences, COVID-­ 19 is believed to bind the baseline with 8-­month comparison), a paired t-­test was angiotensin-­ converting enzyme 2 (ACE2) © Author(s) (or their used for normally distributed data and Wilcoxon signed receptor located on virtually all endothelia employer(s)) 2022. Re-­use permitted under CC BY-­NC. No ranks sum for non-­normally distributed data. with high affinity.4 It activates the renin/ commercial re-­use. See rights Results The fovea-­centered superficial and deep angiotensin system leading to a downregu- and permissions. Published by vascular densities were significantly diminished in severe lation in ACE2 expression and increasing BMJ. cases compared with mild cases (p=0.004; p=0.003, angiotensin II. Angiotensin II is known to be 1 Department of Ophthalmology, respectively, for superficial and deep) and to controls (p=0.014; p=0.010), also in moderate cases to mild group a potent vasoconstrictor and enhancer of the Vall d'Hebron University Hospital, Barcelona, Catalunya, (p=0.004; p=0.003) and to controls (p=0.012; p=0.024). In expression of tissue factor and plasminogen Spain the longitudinal study, no significant statistical differences activator inhibitor leading to thrombosis. 2 Infectious Diseases were found between baseline and 8-­month follow-­up It also provokes complement activation Department, Universitat vessel density values. via formation of reactive oxygen species Autònoma de Barcelona, Conclusion We demonstrated persistent reduction in the and downregulation of C1 inhibitor, which Hospital Universitari Vall d'Hebron, Barcelona, Spain central vascular area over time in patients with moderate increases vascular permeability and cytokine 3 Preventive Medicine and and severe COVID-­19. release, enhancing its thrombotic effect.5–7 Epidemiology Department, SARS-­CoV-­2 infection induces a generalised Hospital Universitari Vall microvascular thrombotic disorder and d'Hebron, Barcelona, Spain 4 INTRODUCTION markedly elevated D-­dimer, which has been IMO, Institute of Ocular Microsurgery, Barcelona, Spain Microvascular alterations in patients with associated with a worse prognosis of the COVID-­19 have recently been described in disease.8 several organs1 including the skin, lungs and Retinal vascular changes in the form of Correspondence to Dr Sandra Banderas García; ​ retina.2 The SARS-­CoV-­2 was associated with decrease in radial peripapillary capillary sandrabanderasgarcia@​gmail.​ organ dysfunction and generalised coagulop- plexus perfusion density,9 reduction in vessel com athy.3 density (VD) of foveal area and larger foveal Banderas García S, et al. BMJ Open Ophth 2022;7:e000867. doi:10.1136/bmjophth-2021-000867 1 Open access avascular area2 10 have been reported in patients with disease and group 3: severe disease, both of which COVID-­19 using optical coherence tomography angiog- required hospital admission because of pneumonia. raphy (OCT-­A). Funduscopy exams have shown other Group 3 subjects developed acute respiratory distress features that are positively correlated with disease severity syndrome with intensive care unit stay and interleukin 6 such as retinal venous diameter, presence of haemor- (IL-­6) values in excess of 40 pg/mL at some point during rhages and cotton wool spots.11 hospitalisation. The control group consisted of healthy In a previous study in which we assessed the state participants with negative serology for SARS-­CoV-­2 and of retinal vascularisation via OCT-­A in a cohort of 69 any viral symptom in the 4 weeks prior to enrolment. patients with COVID-­19, we found lower central retina Exclusion criteria were consistent with the previous VD of the superficial plexus, in relation to the severity of study: known prior ophthalmological diseases, myopia the infection.10 OCT-­A provides a safe and non-­invasive greater than −6 dioptres, diabetic retinopathy and active diagnostic way of studying retinal vascular plexuses: vascular diseases with retinopathy (multiple myeloma, superficial capillary plexus (SCP), deep capillary plexus hyperglobulinemia and malignant hypertension). Only (DCP) and choriocapillaris.12 OCT-­A triton uses a central one eye per patient is included in the analyses (the left wavelength of 1.050 nm with 100.000 A-­scans per second one) in order to minimise the risk of bias. to detect motion contrast and visualise blood flow in vivo. All case and control subjects were asked to reconsent to The OCT-­A software uses a commercially available algo- participate in the extension phase of the study 8 months rithm, OCTRA, which has been shown to provide highly later (between 17 November 2020 and 29 January 2021). sensitive detection of low blood flow in microvasculature, Data from the baseline examination was compared with used in DRI OCT triton equipment.13 the data obtained 8 months later. The control group was The long-­term consequences of COVID-­19 on retinal reassessed to confirm that they continued to meet the vascularisation are still unknown, as most of the avail- inclusion criteria, that is, no viral symptoms 4 weeks prior able studies are cross-­ sectional.2 10 Here, we provided to the new assessment and negative serology. A flow chart an 8-­month follow-­up analysis of our previously studied depicts drop-­outs (figure 1). cohort. Patient and public involvement statement MATERIALS AND METHODS Neither patients nor members of the public were involved Inclusion and selection of patients in the design, conduct, reporting or dissemination plans We present a cross-­sectional case–control study (8-­month of our research. follow-­up) and a prospective longitudinal cohort study (baseline and 8-­month follow-­up) performed at the Vall Examinations performed d’Hebron University Hospital (Barcelona, Spain). We A repetition of the same ocular examination from base- included the participants of our previous study (June– line was performed on all the subjects 8 months later: July 2020).10 Cases were defined as patients between 18 a 50° colour photograph of the retina posterior pole and 55 years of age with laboratory-­ confirmed SARS-­ (retinography), a macula-­centred high-­definition OCT CoV-­2 infection in the 3 months previous to enrolment (scan radial 12 mm) and an automatically fovea-­centred in the study (June–July 2020). They were divided into: OCT-­A (4.5×4.5 mm) were performed. All three imaging group 1: mild disease, asymptomatic/paucisymptomatic techniques were obtained with the DRI OCT Triton Swept subjects who received outpatient care; group 2: moderate Source equipment (Topcon Corporation, Tokyo, Japan). Figure 1 Flow chart of patient selection. Baseline accounts for 3-­month analysis. The same participants of each group were approached and asked to take part in the follow-­up analysis at 8 months. Subjects were once again assessed for inclusion and exclusion criteria. 2 Banderas García S, et al. BMJ Open Ophth 2022;7:e000867. doi:10.1136/bmjophth-2021-000867 Open access Figure 2 Example of the angiography by optical coherence tomography (OCT-­A). The first images on the left correspond to superficial plexus, the second to deep plexus, the third to automated vessel density (VD) analysis of the superficial plexus and the fourth (if present) to the analysis of the VD in deep plexus. (A) Images correspond to the baseline data of a control group patient. (B) Images correspond to 8-­month data of the same control group subject. (C) Images were obtained from a group three patient at baseline. (D) Images correspond to the same subject as in (C) but after 8 months. Note how in a macroscopic scale the tendency to a larger FAZ area and lower vascular density in the central area can be observed. No macroscopic abnormalities in the morphology of FAZ are noted: no notches, avascular areas or vessel irregularities. FAZ, foveal avascular zone. Retinography images and radial OCT sections were manual delineation of randomly numbered images analysed by two different ophthalmologists (SBG and with ImageJ Software (National Institutes of Health) by DA). A third ophthalmologist carried out a double evalu- three different blinded observers (SBG, DAR and MAZ). ation of the images (MAZ). A search for any abnormality The surface area was measured in square pixels and was such as haemorrhages, cotton wool spots or OCT layer converted to square millimetres using the formula given hyperreflectivities was carried out. by the software (scale setting): delineated area (mm2)=de- In contrast to our previous study, we gathered data lineated area (pixel2)/total surface area (pixel2) * from automatic segmentation from both superficial total surface area (mm2). No editing of the image was capillary plexus and deep capillary plexus from the performed to increase contrast or change colours prior macular region. Superficial layer VD was automatically to manual delineation. segmented: from the internal limiting membrane (ILM) at a depth of 2.6 µm to the inner plexiform layer (IPL)/ Statistical analysis inner nuclear layer (INL) at a depth of 15.6 µm. IPL/ Statistical analyses were performed using SPSS (V.23.0) INL at 15.6 µm to INL/outer plexiform layer at 70.2 µm software. Normal distribution of the variables was automatically encompasses the deep layer VD. The speci- checked using the Kolmogorov-­Smirnov test. In the first fied software calculated VD, and data were collected from part of the study, cross-­sectional intergroup differences superior, inferior, nasal, temporal and central areas. All were analysed (8-­ month time). Quantitative variables captures with a signal strength index of less than 40, incor- that followed a normal distribution were studied with an rect segmentation or motion artefacts were excluded. analysis of variance (ANOVA) using Bonferroni correc- Figure 2 shows an example of the images obtained. tion. A Kruskal-­Wallis test was performed to compare The area of the foveal avascular zone (FAZ) in the quantitative variables not normally distributed. Qualita- superficial capillary plexus was analysed by means of tive variables were examined using the χ2 test. Banderas García S, et al. BMJ Open Ophth 2022;7:e000867. doi:10.1136/bmjophth-2021-000867 3 Open access Table 1 Demographic features of study population Group 2: moderate Group 3: severe Significance Control group Group 1: mild disease disease disease among all groups Months from PCR   – – 8.20 (0.62) 8.74 (0.68) (mean, SD) Gender (n, %) Male 8 (42.10) 6 (33.33) 9 (42.85) 9 (52.94) 0.711* Female 11 (57.89) 12 (66.66) 12 (57.14) 8 (47.05) Age (years, SD)   38, 79 (8, 15) 40, 61 (11, 04) 43, 29 (8, 49) 46, 00 (5, 75) 0.078+ Race (n) Caucasian 19 17 15 9 0.008* Latin 0 1 6 7 Other 0 0 0 1 Diabetes (n) No 19 18 20 15 0.236* Yes 0 0 1 2 Smoker (n) No 16 15 21 15 0.292* Yes 3 3 0 2 Significant values shown in bold. *Χ2 test. †ANOVA test. ANOVA, analysis of variance. In reference to the prospective part of the study (intra- haemorrhage suggestive of a Valsalva manoeuvre (size group differences), a paired t-­test was used for normally of approximately 100 µm, round and well-­circumscribed distributed data in order to compare baseline and under ILM in the premacular area). 8-­month changes in the VD mean for superficial plexus as well as for the FAZ mean area. A Wilcoxon signed ranks OCT-A and VD sum test was used for not normally distributed VD means. Cross-sectional results Significance was fixed at p values below 0.05. At macroscopic scale, no abnormalities such as neovas- cularisation, ischaemic areas, vascular dilatation or RESULTS abnormal organisation of vessels were detected after Study population careful consideration of OCT-­A images by three observers We included a total of 75 patients from our previous (SBG, DA and MAZ) either in control or in case groups. study9 who reagreed to participate with no exclusion The FAZ tended to have a larger area in severe cases, criteria (figure 1). Control group was formed by 19 with a statistically significant difference in means among subjects. Among patients with laboratory-­ confirmed groups (p<0.001, (online supplemental figure 2) and SARS-­CoV-­2 infection, group 1 included 18 participants, table 2): control group versus group 2 (p<0.001); control group 2 included 21 and group 3 included 17. Mean group versus group 3 (p=0.001); group 1 versus group 2 duration from PCR-­confirmed diagnosis to ophthalmo- (p<0.001); and group 1 versus group 3 (p<0.001). The logical examination time was 8.74 months (SD 0.68) for mean superficial and deep VDs in control group for the group 2 and 8.20 months (SD 0.62) for group 3. As in central area were 16.14 and 15.49 (SD 3.40; 4.61), in the previous study, there were no statistically significant group 1, they were 17.41 and 15.78 (SD 4.07; 4.06), in differences between groups in terms of age, gender, group 2, they were 13.41 and 12.43 (SD 4.79; 5.38) and smoking or diabetes record (table 1). The mean age was in group 3, they were 12.96 and 11.38 (SD 3.56; 3.93) lower in the control group and group 1 than in groups (online supplemental figures 2 and 3) and table 2). 2 and 3, with no statistical significance. In reference to These data were non-­normally distributed. After analysis ethnicity, there was statistical difference: a predominance with Kruskal-­Wallis, a statistically significant difference of Caucasian origin in all groups with a statistically signif- of central VD means among groups was found: p=0.002 icant greater presence of Latin origin in groups 2 and 3 for both global comparison of all groups for superficial (p=0.008). and deep. Superficial and deep central VD means in control group and in group 1 were higher than in group Fundoscopic and structural OCT findings 2 and group 3. In a Mann-­Whitney U test (the first p No new abnormalities in previously healthy colour value represents superficial and the second represents fundus images and structural OCT sections were found. deep), statistical differences were found in the control No abnormalities attributable to SARS-­CoV-­2 were found. group versus group 2 (p=0.012; p=0.024), and the control In line with our previous findings, there were two subjects group versus group 3 (p=0.014; p=0.010), group 1 versus who presented asymptomatic small pigment epithe- group 2 (p=0.004; p=0.003) and group 1 versus group 3 lial detachments and one group 2 patient with a small (p=0.005; p=0.003). 4 Banderas García S, et al. BMJ Open Ophth 2022;7:e000867. doi:10.1136/bmjophth-2021-000867 Open access Table 2 Vascular density values at the 8-­month follow-­up and other characteristics within groups Group 1: mild Group 2: moderate Group 3: severe Significance among Control group disease disease disease all groups Superficial central VD Mean 16.1432 17.416 13.411 12.968 0.002* SD 3.405 4.079 4.793 3.566 Superficial inferior VD Mean 50.368 50.125 52.980 50.464 0.761* SD 2.712 3.321 8.293 2.483 Superficial superior VD Mean 49.995 49.013 50.128 50.141 0.391† SD 2.406 2.053 2.376 2.666 Superficial nasal VD Mean 43.373 42.716 44.530 44.340 0.163† SD 2.302 3.496 2.694 2.485 Superficial Mean 47.711 46.015 49.925 48.271 <0.001† temporal VD SD 2.369 3.461 2.602 2.307 Image quality Mean 63.42 63.44 61.950 64.120 0.452† SD 3.220 5.044 5.162 3.100 FAZ area Mean 0.223 0.210 0.411 0.343 <0.001† SD 0.150 0.017 0.022 0.026 Deep central VD Mean 15.495 15.782 12.430 11.388 0.002* SD 4.613 4.068 5.382 3.933 Deep inferior VD Mean 51.507 53.223 53.324 52.645 0.7347* SD 5.327 3.974 3.809 3.218 Deep superior VD Mean 51.066 50.996 52.466 51.361 0.484† SD 3.606 3.167 3.451 3.175 Deep nasal VD Mean 49.497 46.551 50.707 49.419 0.010† SD 3.693 3.021 4.499 3.650 Deep temporal VD Mean 48.034 48.430 50.092 47.708 0.043* SD 2.898 2.893 3.027 2.769 Significant values shown in bold. *H Kruskal-­Wallis. †ANOVA test. ANOVA, analysis of variance; FAZ, foveal avascular zone; VD, vessel density. Normally distributed data of superficial temporal and group 2 at 8-­month follow-­up was higher in men than nasal deep VD were analysed with ANOVA, and a statis- women (p=0.039) also for group 3 at 8-­month follow-­up tical difference was noted (p=0.002; p=0.010). After (p=0.031); and deep inferior VD for group 2 at 8-­month Bonferroni correction for multiple measurements, a follow-­up was higher in women group than men group statistical difference was found only between group 1 and (p=0.018). group 2 (p<0.001, p=0.006) with no differences among other groups. Non-­ normally distributed data from Prospective results temporal deep VD was analysed using Kruskal-­ Wallis, A paired t-­test was performed for normally distributed with a significant statistical difference in means among data and a Wilcoxon signed rank test for non-­normally groups (p=0.043) which, after a Mann Whitney U test, distributed variables (table 3). Comparison of the VD was found to be between the control group and group 2 means of every area of the superficial plexus between (p=0.021) and between group 2 and group 3 (p=0.011). baseline and its value at the 8 month follow-­up was made. No other statistically significant results were found. FAZ areas are also included in the prospective analysis. Table 2 shows a summary of all comparisons. No statistically significant differences were found (online VD differences between male and female within every supplemental figure 2). group were also examined (online supplemental table) as general tendency to lower values in female data were DISCUSSION found. The following statistically significant different The main finding of this study is lower central retinal VD means were found: superficial central VD at baseline in superficial and deep plexuses in patients who have for group 1 was higher in women than men (p=0.034); suffered from moderate and severe COVID-­ 19 pneu- superficial inferior VD at baseline for group 2 was higher monia. This is the first study to provide evidence of the in men than women (p=0.009); deep central VD for persistence of these findings after 8-­month follow-­up. Banderas García S, et al. BMJ Open Ophth 2022;7:e000867. doi:10.1136/bmjophth-2021-000867 5 Open access Table 3 Prospective analysis SCP VD Global Group 1 Group 2 Group 3 Control group Central p=0.143* p=0.102* p=0.835* p=0.647* p=0.147* Inferior p=0.065* p=0.215* p=0.073* p=0.369* p=0.094* Superior p= 0.105† p=0.756* p=0.230 * p=0.616* p=0.313 * Nasal p=0.714* p=0.679* p=0.375* p=0.557* p=0.117* Temporal p=0.612 * p=0.619 * p=0.58 * p=0.811* p=0.108* FAZ area p=0.181† p=0.330† p=0.141† p=0.453† p=0.082† Significant values shown in bold. Superficial plexus VD mean comparison at baseline and at 8-­month examination. *Wilcoxon signed rank test. †Paired t-­test. FAZ, foveal avascular zone; SCP, superficial capillary plexus; VD, vessel density. SARS-­CoV-­2 virus is suspected to cause a type of micro- was found regarding sex and central VD in our previous vascular angiopathy14 consisting of acute endothelial study,10 we decided to examine possible differences in dysfunction mediated by proinflammatory cytokines VD data between male and female groups at baseline and such as tumour necrosis factor alpha or IL-­6 that lead to at follow-­up (online supplemental material). Guemes-­ an enhanced risk of microvascular thrombosis and even- Villahoz et al23 suggested that women could undergo tually multiorgan dysfunction. Direct injury as a result either slower recovery or progressive worsening of retinal of the infection via this process has been reported in VD and questioned whether this might have any relation- postmortem studies in organs such as brain,15 liver16 and ship to major persistence of neurological symptoms after kidney.17 This increase in micro thrombosis could lead the illness in the female population. In our study, VDs of to diminished vascular density in affected organs such as all areas in the female group, although generally lower the vascular plexuses of the retina. than in the male group, were mostly no different. We The severe and moderate infection groups in our study found no other reports in the literature with evidence showed signs of respiratory distress and increased IL-­6 supporting the possibility that women would be at in the latter group, which are believed to be signs of an higher risk of COVID-­19 vascular-­related events. Further underlying thrombotic microangiopathy.18 IL-­6 increase research in this direction is planned. is a sign of inflammation in SARS-­CoV-­2 disease linked The long-­term consequences of COVID-­19 in the retina to an unfavourable outcome of the disease and the best have not been examined. As far as we know, no clinical or predictor for the requirement of invasive mechanical fundoscopic macroscopic abnormalities were discovered ventilation.19 IL-­6 correlates with macrophage activation at follow-­up, such as we could expect as a consequence of in severe COVID-­19 cases20 and seems to have a role in the reduced VD. We found only one isolated case report microvascular angiopathy by inhibition of ADMTS-­ 13 of persistent cotton wool spots on fundus examination activity, raising levels of fibrinogen, plasminogen inhib- at 6 months.23 For the moment, none of our patients itor-­1 and C reactive protein.21 22 In our first study,10 we presented any clinical consequences of the microvas- demonstrated the existence of a reduction in the central cular abnormalities evidenced herein. Some studies have VD of the superficial plexus via OCT-­A and a tendency to tried to relate retinal fibre nerve layer and ganglion cell a larger FAZ area. Other groups have ratified our find- layer thicknesses with decreases in VD, but no structural ings with similar results.2 23 24 There may be three reasons correlations have been found.24 It is still not known for this: one, ACE2 expressed in the endothelium of the whether vasculature structures of the retinal plexuses are retina20 with direct effect; two, induced-­endothelitis; and more vulnerable to thrombotic events than other body three, metabolic changes by serum IL-­6 leading to distur- plexuses, as these abnormalities seem to remain clini- bances of the blood retina barrier and dysfunction of cally silent. We believe long-­term follow-­up is needed to vessels.25 assess future possible ischaemic/neovascular diseases of In the present study, reduction of vascular density the retina, especially in patients who have suffered from remains stable after 8 months (online supplemental severe COVID-­19. figure 2). This is the most important finding of the If persistent microvascular changes are found in the present study, that is, the persistence of significant lower retina beyond 8 months, it may be presumed that other VDs in severe and moderate disease. A comparison organs and their plexuses may have sustained long-­ with the control group to evaluate the impact of ageing term damage. Current studies focus their attention was undertaken, which proves the consistency of the on ‘persistent COVID-­ 19’26 and reveal frequencies of persistently reduced central SCP, DCP and enlarged FAZ up to 72% of concentration ability difficulties, 68% of area, and this correlates with the results of other groups insomnia, 64% of memory impairment, 75% of fatigue, at 6 months.23 As a tendency to a negative correlation 55% of tachycardia and 63% of nonproductive cough, 6 Banderas García S, et al. BMJ Open Ophth 2022;7:e000867. doi:10.1136/bmjophth-2021-000867 Open access among other symptoms. There is currently a contro- In conclusion, this is the first study to find persistently versy regarding the relationship between psychological reduced VDs in the retina 8 months after COVID-­ 19 and functional impairment, systemic consequences and infection. We demonstrate persistent tendency towards a initial severity. Some studies, such as that of Huang et al,27 larger FAZ area and reduction in the central area of the did find an association between acute COVID-­19 disease superficial and deep plexuses, with correlation to disease severity, persistence of symptoms such as fatigue, weak- severity. Therefore, we believe close follow-­up of patients ness, anxiety, impaired pulmonary diffusion capacities with severe disease should be performed, in order to and chest imaging abnormalities. Further studies are detect possible long-­term systemic and ophthalmological needed to assess whether recovery symptoms in critical disease sequelae. illness patients are comparable with those typically seen in acute respiratory distress syndrome survivors,28 or Acknowledgements We would like to acknowledge the contribution of the Surgery Department of the Autonomous University of Barcelona (UAB – whether persistent poor health after COVID-­19 is finally Departament de Cirurgia i Ciències Morfològiques). not associated with respiratory complications.29 How Contributors MAZ and SBG made a substantial contribution to the present work in perfusion reduction during acute SARS-­CoV-­2 infection the design of the study, acquisition, analysis and interpretation of data, manuscript leads to long-­term sequelae of the virus in several organs drafting and approval of the version of the manuscript to be published. AS-­M and how this correlates to later symptoms is currently and SO-­R contributed to acquisition of data, conception and design of the study, being examined in research papers: pancreatic islets,30 critical revision for important intellectual content and approval of the version to be published. DA, BA, XG-­S and FT participated in the data acquisition, critical heart,31 brain32 or kidney.26 revision for important intellectual content and approval of the present version of If this were the case, VD of the retina measured by the manuscript. JG-­A’s contribution to the present work included conception and OCT-­A would be of use as a marker of widespread organ design of the study, critical revision for important intellectual content and approval injury in post-­COVID-­19 patients. Hekimsoy et al33 have of the present document. SBG is the guarantor and accepts full responsibility for the work, conduct of the study, had access to the data, and controlled the decision already established a negative correlation between retinal to publish. All the authors hereby mentioned meet all four ICMJE criteria for perfusion and idiopathic pulmonary arterial hyperten- authorship and are accountable for all aspects of the work regarding its accuracy sion, a pathology that has been studied in recent times and integrity. in the context of COVID-­1934 and that results in a higher Funding The authors have not declared a specific grant for this research from any fatality rate when SARS-­CoV-­2 infection occurs. Further funding agency in the public, commercial or not-­for-­profit sectors. studies to correlate systemic signs/symptoms are neces- Competing interests None declared. sary to elucidate this question. Interestingly, perfusion Patient consent for publication Consent obtained directly from patient(s) studies with tomography angiography in the lungs and Ethics approval The study was approved by the Vall d’Hebron University Hospital kidneys of COVID-­19 patients showed perfusion deficits ethical committee under the number PR (AG) 337/2020, and all enrolled subjects suggestive of microvascular occlusion with no detectable signed an informed consent statement. Participants gave informed consent to participate in the study before taking part. major dysfunction, similar to our findings.35 We acknowledge a series of limitations in our study. Provenance and peer review Not commissioned; externally peer reviewed. First, the small sample size of the case cohort, as a conse- Data availability statement Data are available on reasonable request. All data quence of the prospective nature of the study, with relevant to the study are included in the article or uploaded as supplementary information. All data relevant to the study are included in the article or uploaded reduced number of patients willing to attend a second as supplementary information, but additional data are available on reasonable ophthalmic examination, lower surveillance rates in request. moderate and severe disease groups, and the study of Supplemental material This content has been supplied by the author(s). It has only one eye per patient. Second, the absence of comor- not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been bidities in our population might need to be taken into peer-­reviewed. Any opinions or recommendations discussed are solely those account when assessing comparability to our studies. of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content Cotton wool spots and retinal haemorrhages are reported includes any translated material, BMJ does not warrant the accuracy and reliability in patients with COVID-­ 19 who commonly also have of the translations (including but not limited to local regulations, clinical guidelines, other pre-­ existing comorbid conditions.36 We believe terminology, drug names and drug dosages), and is not responsible for any error that the lack of comorbidities might be beneficial for and/or omissions arising from translation and adaptation or otherwise. measuring the real impact of the SARS-­CoV-­2 infection Open access This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-­NC 4.0) license, which on VD, since the presence of multiple pathologies might permits others to distribute, remix, adapt, build upon this work non-­commercially, affect its values. Third, no retinal periphery studies have and license their derivative works on different terms, provided the original work is been performed, and we lack any information regarding properly cited, appropriate credit is given, any changes made indicated, and the functional translation of our changes, for example, via use is non-­commercial. See: http://creativecommons.org/licenses/by-nc/4.0/. electroretinography or microperimetry, which would be ORCID iD of use to state with confidence that there are no clinical Sandra Banderas García http://orcid.org/0000-0002-3641-3806 consequences. Fourth, a longer follow-­up than 8 months should be carried out. Prospective studies on larger cohorts will be needed for follow-­up of retinal microvas- REFERENCES cular abnormalities in post-­COVID-­19 patients, especially 1 Magro C, Mulvey JJ, Berlin D, et al. Complement associated microvascular injury and thrombosis in the pathogenesis of in moderate and severe cases, to detect progression, severe COVID-­19 infection: a report of five cases. Transl Res reversibility or vascular consequences. 2020;220:1–13. Banderas García S, et al. BMJ Open Ophth 2022;7:e000867. doi:10.1136/bmjophth-2021-000867 7 Open access 2 González-­Zamora J, Bilbao-­Malavé V, Gándara E, et al. Retinal 20 McGonagle D, Sharif K, O'Regan A, et al. The role of cytokines microvascular impairment in COVID-­19 bilateral pneumonia including interleukin-­6 in COVID-­19 induced pneumonia and assessed by optical coherence tomography angiography. macrophage activation syndrome-­like disease. Autoimmun Rev Biomedicines 2021;9:247. 2020;19:102537. 3 Tang N, Li D, Wang X, et al. Abnormal coagulation parameters are 21 Libby P, Simon DI. Inflammation and thrombosis: the clot thickens. associated with poor prognosis in patients with novel coronavirus Circulation 2001;103:1718–20. pneumonia. J Thromb Haemost 2020;18:844–7. 22 Lee DD, Schwarz MA. Cell-­Cell communication breakdown and 4 Sparks MA, Crowley SD, Gurley SB, et al. Classical renin-­angiotensin endothelial dysfunction. Crit Care Clin 2020;36:189–200. system in kidney physiology. Compr Physiol 2014;4:1201–28. 23 Guemes-­Villahoz N, Burgos-­Blasco B, Vidal-­Villegas B, et al. 5 Shagdarsuren E, Wellner M, Braesen J-­H, et al. Complement Reduced retinal vessel density in COVID-­19 patients and elevated activation in angiotensin II-­induced organ damage. Circ Res D-­dimer levels during the acute phase of the infection. Med Clin 2005;97:716–24. 2021;156:541–6. 6 Forrester SJ, Booz GW, Sigmund CD, et al. Angiotensin II signal 24 Cennamo G, Reibaldi M, Montorio D, et al. Optical coherence transduction: an update on mechanisms of physiology and tomography angiography features in Post-­COVID-­19 pneumonia pathophysiology. Physiol Rev 2018;98:1627–738. patients: a pilot study. Am J Ophthalmol 2021;227:182–90. 7 Bekassy Z, Lopatko I, Bader M. Crosstalk between the renin– 25 Teuwen L-­A, Geldhof V, Pasut A, et al. COVID-­19: the vasculature angiotensin, complement and kallikrein–kinin systems in unleashed. Nat Rev Immunol 2020;20:389–91. inflammation. Nat Rev Immunol 2021:1–18. 26 Rando HM, Bennett TD, Byrd JB, et al. Challenges in defining 8 Zhou F, Yu T, Du R, et al. Clinical course and risk factors for mortality long COVID: striking differences across literature, electronic health of adult inpatients with COVID-­19 in Wuhan, China: a retrospective records, and patient-­reported information. medRxiv 2021. doi:10.110 cohort study. Lancet 2020;395:1054–62. 1/2021.03.20.21253896. [Epub ahead of print: 26 Mar 2021]. 9 Savastano A, Crincoli E, Savastano MC, et al. Peripapillary retinal 27 Huang C, Huang L, Wang Y, et al. 6-­Month consequences of vascular involvement in early Post-­COVID-­19 patients. J Clin Med COVID-­19 in patients discharged from Hospital: a cohort study. 2020;9:2895. Lancet 2021;397:220–32. 10 Zapata Miguel Ángel, Banderas García S, Sánchez-­Moltalvá A, 28 Parker AJ, Humbir A, Tiwary P, et al. Recovery after critical illness in et al. Retinal microvascular abnormalities in patients after COVID-­19 COVID-­19 ICU survivors. Br J Anaesth 2021;126:e217–9. depending on disease severity. Br J Ophthalmol 2020;317953. 29 Townsend L, Dowds J, O'Brien K, et al. Persistent poor health after doi:10.1136/bjophthalmol-2020-317953. [Epub ahead of print: 16 COVID-­19 is not associated with respiratory complications or initial Dec 2020]. disease severity. Ann Am Thorac Soc 2021;18:997–1003. 11 Invernizzi A, Torre A, Parrulli S, et al. Retinal findings in patients with 30 Hayden MR. An immediate and long-­term complication of COVID-­19: results from the SERPICO-­19 study. EClinicalMedicine COVID-­19 may be type 2 diabetes mellitus: the central role of β-cell 2020;27:100550. dysfunction, apoptosis and exploration of possible mechanisms. 12 Spaide RF, Fujimoto JG, Waheed NK, et al. Optical coherence Cells 2020;9:2475. tomography angiography. Prog Retin Eye Res 2018;64:1–55. 31 Liu J, Deswal A, Khalid U. COVID-­19 myocarditis and long-­term 13 Moussa M, Leila M, Khalid H. Imaging choroidal neovascular heart failure sequelae. Curr Opin Cardiol 2021;36:234–40. membrane using en face swept-­source optical coherence 32 Guedj E, Campion JY, Dudouet P, et al. 18F-­FDG brain PET tomography angiography. Clin Ophthalmol 2017;11:1859–69. hypometabolism in patients with long COVID. Eur J Nucl Med Mol 14 Iba T, Connors JM, Levy JH. The coagulopathy, endotheliopathy, Imaging 2021;48:2823–33. and vasculitis of COVID-­19. Inflamm Res 2020;69:1181–9. 33 Hekimsoy V, Kılınç Hekimsoy H, Şekeroğlu MA, et al. The relationship 15 Younger DS. Postmortem neuropathology in COVID-­19. Brain Pathol between retinal microcirculation and right heart catheterization and 2021;31:385–6. echocardiography findings in patients with idiopathic pulmonary 16 Zhao CL, Rapkiewicz A, Maghsoodi-­Deerwester M, et al. arterial hypertension. Microcirculation 2021;28:e12704. Pathological findings in the postmortem liver of patients with 34 Sulica R, Cefali F, Motschwiller C, et al. COVID-­19 in pulmonary coronavirus disease 2019 (COVID-­19). Hum Pathol 2021;109:59–68. artery hypertension (PAH) patients: observations from a large PAH 17 Varga Z, Flammer AJ, Steiger P, et al. Endothelial cell infection and center in New York City. Diagnostics 2021;11:128. endotheliitis in COVID-­19. Lancet 2020;395:1417–8. 35 Idilman IS, Telli Dizman G, Ardali Duzgun S, et al. Lung and kidney 18 Dolhnikoff M, Duarte-­Neto AN, de Almeida Monteiro RA, et al. perfusion deficits diagnosed by dual-­energy computed tomography Pathological evidence of pulmonary thrombotic phenomena in in patients with COVID-­19-­related systemic microangiopathy. Eur severe COVID-­19. J Thromb Haemost 2020;18:1517–9. Radiol 2021;31:1090–9. 19 Aziz M, Fatima R, Assaly R. Elevated interleukin-­6 and severe 36 Sim R, Cheung G, Ting D. Retinal microvascular signs in COVID-­19. COVID-­19: a meta-­analysis. J Med Virol 2020;92:2283–5. Br J Ophthalmol 2021. 8 Banderas García S, et al. BMJ Open Ophth 2022;7:e000867. doi:10.1136/bmjophth-2021-000867