The Current Status of OCT and OCTA Imaging for the Diagnosis of Long COVID

Abstract

(1) With persistent symptoms emerging as a possible global consequence of COVID-19, the need to understand, diagnose, and treat them is paramount. This systematic review aims to explore the potential of optical coherence tomography (OCT) and/or optical coherence tomography angiography (OCTA) in effectively diagnosing long COVID. (2) The database PubMed and, to reduce selection bias, the AI research assistant Elicit, were used to find relevant publications in the period between February 2021 and March 2024. Included publications on OCT and OCTA analysis of participants with acute COVID symptoms, those after recovery, and participants with long COVID symptoms were organized in a table. Studies with participants under the age of 18, case reports, and unrelated studies, such as pure slit-lamp examinations and subgroup analyses were excluded. (3) A total of 25 studies involving 1243 participants and 960 controls were reviewed, revealing several changes in the posterior eye. Long COVID participants displayed significant thinning in retinal layers in the OCT, including the macular retinal nerve fiber layer (mRNFL), ganglion cell layer (GCL), and inner plexiform layer (IPL). Divergent findings in recovered cohorts featured mRNFL reduction, GCL increase and decrease, and GCL-IPL decrease. Long COVID OCTA results revealed reduced vessel density (VD) in the superficial capillary plexus (SCP), intermediate capillary plexus (ICP), and deep capillary plexus (DCP). In recovered patients, SCP consistently showed a reduction, and DCP exhibited a decrease in five out of six publications. The foveal avascular zone (FAZ) was enlarged in five out of nine publications in recovered participants. (4) During various stages of COVID-19, retinal changes were observed, but a comparison between long COVID and recovered cohorts was aggravated by diverse inclusion and exclusion criteria as well as small sample sizes. Changes in long COVID were seen in most OCT examinations as thinning or partial thinning of certain retinal layers, while in OCTA a consistently reduced vessel density was revealed. The results suggest retinal alterations after COVID that are variable in OCT and more reliably visible in OCTA. Further research with larger samples is important for advancing long COVID diagnosis and management.

1. Introduction

After an infection with SARS-CoV-2 and a subsequent negative test result, indicating an alleged recovery, many people keep experiencing continuous after-effects of the illness. This is mostly referred to as long COVID, also known as post-acute sequelae syndrome (PASC), post-acute COVID syndrome (PACS), chronic COVID syndrome (CCS), or COVID-19 long-hauler [1]. Diagnosing long COVID can be challenging as there are no specific tests [2]. The diagnosis is based on the remainder of symptoms or the worsening of a pre-existing condition [1] longer than four weeks after acute SARS-CoV-2 infection and eliminating other possible causes [1,3]. The illness can affect multiple organ systems [4]. It can include symptoms such as fatigue [5,6,7,8], difficulties focusing, and shortness of breath [9]. While research is investigating various causes, it is believed that organ and tissue damage could be due to malfunctioning small blood vessels, damaged endothelium [10], and impaired blood flow [11,12]. These effects can be detected indirectly in the eye, allowing the drawing of conclusions about the whole body’s microvasculature. With optical coherence tomography (OCT), the thickness of the retinal layers can be measured using incident laser light, which is reflected by the different retinal tissue structures [13]. Changes in the tissue layers of the retina can be ascribed to changes such as a restricted blood supply or inflammatory processes [14]. Optical coherence tomography angiography (OCTA), an advancement of OCT, creates a three-dimensional visualization of retinal and choroidal blood vessels [15]. It allows a visualization of deeper blood plexuses than previously possible with traditional methods involving the use of dye injections [16,17,18]. An OCT-image at a single axial point of the eye is called an A-scan. Multiple A-scans in one plane together form a B-scan (Figure 1a). For the OCTA-image, a cross section of several consecutive B-scans is taken, static pixels, typically representing tissue, are subtracted, resulting in a consolidated image of all blood vessels with moving erythrocytes. Intensity projections can be calculated using layer segmentation and the volume dataset to create en face images (Figure 1b) [16]. Using OCT and OCTA as detective tools for long COVID may enable an early diagnosis, reduce patients’ anxiety, facilitate research, and improve treatment options.

2. Materials and Methods

This review was performed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and the protocol was registered with Open Science Framework (OSF) under the Registration DOI: https://doi.org/10.17605/OSF.IO/JW2TS. In the PubMed database (Search terms A1 and A2), which was filtered for clinical studies, and on Elicit.com (Search terms A3), two independent researchers manually identified 17 publications on OCT and OCTA in COVID-19 patients to minimize selection bias. A total of 12 complementary publications were added by cross-referencing. All 29 articles were screened to include thematically relevant publications on OCT and OCTA in COVID-19 patients in the acute phase, recovered participants, and participants with remaining symptoms in this review. To achieve greater homogeneity and better comparability, we excluded studies on patients under the age of 18, case reports, and non-topic-related articles. Unrelated topics comprised pure slit-lamp examinations and subgroup analyses. The final 25 articles included in this review were checked for plausibility by one reviewer, while relevant data were collected and extracted by the second reviewer. The sought outcome parameters were retinal layer and choroidal thicknesses in OCT and vessel densities in the different vascular plexuses as well as the FAZ area in OCT. For a more differentiated presentation, significant and insignificant OCT and OCTA results were separately tabulated (Table 1 and Table 2) and the number of participants, the time frame, and the devices used for examination of each publication included in the respective tables. All publications used OCTA and/or OCT devices as tools to assess the chorioretinal microvessels (Table 1 and Table 2). A long COVID cohort was examined by Kanra et al. [20] using the Spectralis OCT (Heidelberg Engineering GmbH, Heidelberg, Germany). In the scans, the macula was divided according to the Early Treatment Diabetic Retinopathy Study ETDRS grid (Figure A1). Dağ Şeker et al. [21] used SD-OCT (HeidelbergEngineering, Heidelberg, Germany) to measure retinal layer thicknesses according to the ETDRS grid (Figure A1). Schlick and Lucio et al. [22] and Szewczykowski and Mardin et al. [23] examined long COVID cohorts with the Spectralis II OCTA (Heidelberg Engineering GmbH, Germany). They explored the retinal microvasculature in the papillary and macular regions, the latter being segmented into the superficial vascular plexus (SVP), the intermediate capillary plexus (ICP), and the deep capillary plexus (DCP) (Figure 1). OCTA scans were taken of a 2.9 × 2.9 mm² area around the macula, while in the compared publications, the scan size varied between 3 × 3 mm² and 6 × 6 mm². To provide clarity, the exclusion criteria for the study and control groups of each publication were listed in an additional table, where missing information was marked with a slash (

Table A1. Exclusion criteria for study participants for every publication.

Publication	Group	Systemic Exclusion Criteria	Ocular Exclusion Criteria
Abrishami et al. [38]	COVID	diabetes mellitus, auto-immune disease, current pregnancy, breastfeeding, migraine	history of refractive or intraocular surgery, spherical refractive error > 5 D, cylindrical refractive error > 2 D, glaucoma, clinically apparent retinal disease, ocular media opacity preventing high-quality imaging or reduced OCTA scan quality
	Control	/	/
Abrishami et al. [41]	COVID	diabetes mellitus, glaucoma, migraine, breastfeeding, current pregnancy, clinically apparent retinal disease, auto-immune diseases, hospitalization, systemic corticosteroid treatment for COVID-19	refractive or intraocular surgery, spherical refractive error > 5 D, cylindrical refractive error > 2 D, ocular media opacity preventing high-quality imaging or reduced OCTA scan quality
	Control	/	/
Abrishami et al. [44]	COVID	history of diabetes mellitus, systemic hypertension, dementia	history of intraocular surgery, glaucoma, ocular hypertension, macular disease
	Control	history of diabetes mellitus, systemic hypertension, dementia	ocular or disc abnormalities, history of intraocular surgery, glaucoma, ocular hypertension, macular disease
Bayram et al. [25]	COVID	any systemic diseases, signs and symptoms of COVID-19 or otherwise abnormal laboratory tests, including high acute phase reactants as systemic inflammatory markers	spherical equivalent > ±3 D, or >26 mm or <21 mm axial length, any ocular diseases, previous ocular surgery, ocular trauma
	Control	/	/
González-Zamora and Bilbao-Malavé et al. [39]	COVID	diabetes mellitus	cataract, vitreous hemorrhages, glaucoma, high myopia, fovea plana, AMD
	Control	/	/
Bilbao-Malavé and González-Zamora et al. [27]	COVID	diabetes mellitus	cataract, vitreous hemorrhages, glaucoma, high myopia, fovea plana, AMD
	Control	/	/
Burgos-Blasco et al. [24]	COVID	still presenting symptoms, on quarantine, unable to attend the hospital, concomitant psychiatric or neurological diseases	glaucoma, congenital optic nerve head abnormalities, myopia/hyperopia > ±6 D, macular disease, retinal vascular disorders, uveitis, and history of previous ophthalmic procedures other than cataract surgery and capsulotomy
	Control	/	/
Cennamo et al. [33]	COVID	history of stroke, blood disorders, diabetes, uncontrolled hypertension, neurodegenerative disease	congenital eye disease, myopia/hyperopia > ±6 D, retinal vascular diseases, macular diseases, previous ocular surgery except uneventful cataract surgery, history of other ocular disorders, significant lens opacity
	Control	/	/
Dağ Şeker et al. [21]	COVID	systemic diseases	ocular disease, history of ophthalmic surgery, systemic or topical drug administration, >±3 D spherical equivalent of refractive errors
	Control	/	/
Dipu et al. [30]	COVID	/	myopia/hyperopia > ±6 D, ocular congenital anomaly, macular disease, media opacity precluding OCTA
	Control	systemic illness likely to influence the orbital vascular flow	/
Hazar et al. [26]	COVID	severe COVID-19 requiring intensive care, diabetes, hypertension, rheumatic disease	glaucoma, retinal disease or eye trauma, media opacities affecting the imaging quality
	Control	/	/
Jevnikar et al. [29]	COVID	diabetes, arterial hypertension, hyperlipidemia, coronary artery disease, history of stroke; concomitant infectious diseases: HIV, HSV, VZV, CMV; systemic treatment linked to retinal toxicity or smoking and other conditions that could have affected the retinal morphology	age-related macular degeneration and other retinal diseases, a history of glaucoma, myopia >−6 D
Jevnikar et al. [35]	COVID	diabetes, arterial hypertension, hyperlipidemia, coronary artery disease, history of stroke; concomitant infectious diseases: HIV, HSV, VZV, CMV; systemic treatment linked to retinal toxicity or smoking and other conditions that could have affected the retinal morphology	age-related macular degeneration and other retinal diseases, a history of glaucoma, myopia >−6 D
	Control	/	/
Kal et al. [37]	COVID	diabetes mellitus	myopia/hyperopia > ±3 D, retinal vascular disease, macular and optic nerve disease, previous ocular surgery (including cataract or glaucoma surgery), uveitis, ocular trauma, AMD, other retinal degenerations and media opacity affecting the OCTA’s scan or image quality
	Control	current or past COVID-19 symptoms, close contact with patients with COVID-19 within the 14 days before the examination	concomitant eye diseases
Kal et al. [36]	COVID	diabetes mellitus, stroke, myocardial infarction, autoimmune diseases	myopia/hyperopia > ±3 D, central and peripheral retinal disorders, optic nerve disorders, a history of intraocular surgery, uveitis, ocular injury, opaque media affecting the quality of the OCT scan
	Control	same as in COVID group, current or past COVID-19 symptoms, close contact with patients with COVID-19 within the 14 days before the examination	same as in COVID group, concomitant eye diseases
Kanra et al. [20]	COVID	neurologic pathology	ocular pathology, myopia/hyperopia > ±3 D
	Control	/	myopia/hyperopia > ±3 D
Mavi Yildiz et al. [31]	COVID	pregnant or breastfeeding	diabetic retinopathy, other choroidal/retinal pathologies, high myopia (an axial length ≥ 26.5 mm), uveitis, glaucoma, previous optic neuropathy, history of intraocular surgery or laser treatment (except for phacoemulsification)
	Control	interviewed potential signs and symptoms of COVID-19 and potentially exposed contacts within the 14 days before the examination	/
Savastano et al. [32]	COVID	ongoing chemotherapy, drug abuse	myopia ≥ 6 D, choroidal atrophy, previously diagnosed glaucoma, retinal occlusive diseases, choroidal neovascularization, central serous chorioretinopathy, infectious choroiditis
	Control	same as in COVID group	same as in COVID group
Savastano et al. [42]	COVID	/	choroidal atrophy, high myopia, exudative AMD, previous episode of central serous chorioretinopathy, glaucoma, acquired and hereditary optic neuropathy, hereditary retinal diseases, demyelinating disorders, neurodegenerative disorders, and keratoconus
	Control	/	same as in COVID group
Schlick and Lucio et al. [22]	COVID	systemic disorders with retinal affection	local disorders with retinal affection
	Control	/	/
Szewczykowski and Mardin et al. [23]	COVID	systemic disorders with retinal affection	local disorders with retinal affection
	Control	/	/
Szkodny et al. [34]	COVID	history of symptomatic SARS-CoV-2 infection without a positive PCR test result, severe general conditions, including acute respiratory distress syndrome (ARDS), myocarditis, cardiac arrhythmia, respiratory insufficiency, and kidney or multiple organ failure, unable to take part in the study	/
	Control	/	ocular surface problems
Turker et al. [28]	COVID	systemic disease that could affect retinal circulation, e.g., diabetes, hypertension, rheumatic diseases; systemic treatment with hydroxychloroquine or steroids	any ocular or systemic disease that could affect retinal circulation, intraocular pressure > 21 mmHg, axial length < 20 mm or > 24 mm, spherical refractive error > ±4 D, astigmatism > 1.5 D that may affect the OCTA results
	Control	/	/
Hohberger and Ganslmayer et al. [43]	COVID	/	no history of a previously known retinal or papillary disorder, no history of ocular laser therapy or surgery
	Control	/	no history of ocular disorders or had a history of laser therapy or ocular surgery
Koçkar et al. [40]	COVID	severe COVID-19, diabetes mellitus, hypertension, chronic obstructive lung diseases, such as asthma, connective tissue disorders, autoimmune diseases	ocular surgery history, spherical equivalent > ±3 D, corneal astigmatism > ±3 D, cataract, corneal diseases, glaucoma, retinal vascular obstruction, AMD
	Control	/	/

AMD = age-related macular degeneration, D = diopters.

). In the context of the existing literature, we discussed the outcome, assessed review strengths and weaknesses, and provided recommendations for future research.

3. Results

This review includes 25 clinical studies (Figure 2) involving 1243 participants and 960 controls from eight European and Asian countries (Germany, India, Iran, Italy, Poland, Slovenia, Spain, and Turkey). All studies were published between February 2021 and March 2024. The data collected using OCT and/or OCTA included studies with participants with acute SARS-CoV-2 (N = 4) and/or participants who recovered from COVID-19 (N = 20) or had long COVID (N = 4). They were compared with controls who either recovered from or never had COVID-19. The recovered COVID-19 group averaged about 48 participants, with an average of around 37 participants in their healthy control group. The control groups were largely age- and gender-matched COVID-19-recovered participants, healthy individuals without COVID-19 history, or were taken from previous data of healthy participants as a retrospective cohort. All studies, except one [24], measured visual acuity in their study groups. Bayram et al. [25] measured the BCVA in the study group of participants with acute COVID-19 but not in controls. Hazar et al. [26] included measurements for their control group that were not precisely defined. Five studies [20,21,23,27,28] did not provide further details on the outcomes, whereas the remaining 16 publications [22,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43] stated no impairment of visual acuity in neither their study cohorts nor controls.

3.1. OCT (Table 1)

3.1.1. Long COVID Participants

Kanra et al. [20] showed significant thinning in at least half of the macular retinal nerve fiber layer (mRNFL), ganglion cell layer (GCL), inner plexiform layer (IPL), and inner retinal layer, with no significant changes in peripapillary retinal nerve fiber layer (pRNFL) or central macular thickness compared to healthy controls [20]. Dağ Şeker et al. [21] showed thinning in parts of the macular after one month and in the mRNFL, GC-IPL, outer nuclear layer (ONL), and pRNFL after one month, as well as 12 months, in comparison with healthy controls. There were no significant differences in longitudinal comparison [21].

3.1.2. Recovered COVID-19 Participants

Five out of nine studies comparing recovered participants with healthy controls had significant outcomes. Burgos-Blasco et al. [24] reported partially decreased mRNFL thickness, while Dipu et al. [30], Mavi Yildiz et al. [31], and Szkodny et al. [34] had no significant results. Burgos-Blasco et al. [24] also displayed significant pRNFL thickening. Mavi Yildiz et al. [31] showed significant thickening in several quadrants, and Cennamo et al. [33] found significant thinning. Dipu et al. [30] and Savastano et al. [42] reported no significant changes in the mRNFL. Similarly, none were found in the pRNFL by Dipu et al. [30] and Abrishami et al. [44]. The GCL-IPL thickness was decreased in the study of Dipu et al. [30], and GCL thickness was likewise decreased in the study of González-Zamora and Bilbao-Malavé et al. [39]; however, the optic nerve (ON) RNFL showed thickening in the latter study [39]. Mavi Yildiz et al. [31] found a significant increase in ONL thickness and central foveal thickness (CFT) but no significant outcomes in the inner nuclear layer (INL), the outer plexiform layer (OPL), or the retinal pigment epithelium (RPE). Two studies showed no significant outcomes in the CFT, ganglion cell complex [32,33] and three studies in the optic nerve head [32,33,44]. Six months post-recovery, a significant further increase in foveal thickness and decreases in pRNFL, ON RNFL, and GCL compared to after two weeks was seen by Bilbao-Malavé and González-Zamora et al. [27].

3.1.3. Participants with an Acute COVID-19 Infection

Jevnikar et al. [29] observed decreased RNFL thickness in half of the measured area during active disease compared to one-year post-recovery, while Bayram et al. [25] found increased pRNFL, ONL, and choroid thickness but no significant change in mRNFL, GCL, IPL, INL, ONL, RPE, and outer retinal layer (ORL) compared to recovered and healthy controls [25]. Another study by Jevnikar et al. [35] comparing active COVID-19 patients with healthy controls showed an increase in parts of the mRNFL and GCL thickness in the active group with severe COVID-19. However, Koçkar et al. [40], comparing similar cohorts, revealed no significant changes.

Table 1. Overview of all included publications on OCT changes in chorioretinal parameters. The table includes the OCTA device used for examination, the number of participants in each cohort, the time of examination, and a summary of their outcomes. The results refer to the group of participants stated first in the headlines.

Publication	Device	Number of Participants	Time of Examination	Results: Significant Changes	Results: No Significant Changes
Long COVID Participants vs. Healthy Controls
Kanra et al. [20]	Spectralis, Heidelberg Engineering, Heidelberg, Germany	20 (34 eyes) 23 (39 eyes)	At least 4 weeks after completed treatment	mRNFL ↓ SI GCL ↓ SI, SO, IO, TO IPL ↓ SI, SO, TI, TO, NI	Remaining subregions pRNFL CMT
Dağ Şeker et al. [21]	SD-OCT Heidelberg Engineering, Heidelberg, Germany	27 (54 eyes) 27 (54 eyes)	31.18 ± 12.35 days after recovery	RT ↓ I, O mRNFL ↓ O GC-IPL ↓ I ONL ↓ C, I pRNFL ↓ C, NI	Remaining subregions INL OPL
			12 months later	mRNFL ↓O ONL ↓ I pRNFL ↓ C, IN	Remaining subregions GC-IPL INL OPL
					None in longitudinal comparison
Recovered COVID-19 participants vs. healthy controls
Dipu et al. [30]	Spectral domain RS-3000 LITE, NIDEK Inc.	35 (70 eyes) 12 (24 eyes)	4 to 6 weeks after hospital discharge	GC-IPL ↓	pRNFL mRNFL
Mavi Yildiz et al. [31]	Spectral domain, Spectralis (HRA + OCT) Heidelberg Engineering	63 (119 eyes) 59 (117 eyes)	2 to 8 weeks after positive real-time RT-PCR	pRNFL ↓ IT, ↑ S ONL ↑ CFT ↑	Remaining subregions mRNFL GCL IPL INL OPL RPE
Savastano et al. [42]	Zeiss Cirrus 5000-HD-OCT Angioplex, Carl Zeiss, Meditec, Inc.	80 ** 30 **	1 month from hospital discharge	None	mRNFL GCC CFT
Cennamo et al. [33]	Spectral domain-OCT AngioVue, Optovue Inc.	40 (40 eyes) 40 (40 eyes)	6 months after hospital discharge	ON RNFL ↓	CFT
Szkodny et al. [34]	Swept Source OCT-DRI OCT Triton, Topcon Inc.	78 (156 eyes) 49 (98 eyes)	1 to 4 months after recovery	None	mRNFL ON
Burgos-Blasco et al. [24]	Spectralis Spectralis, Heidelberg Engineering, Heidelberg, Germany	90 (90 eyes) 70 (70 eyes)	4 weeks	pRNFL ↑ mRNFL↓ SI, NI, NO GCL ↑ SO, NO, IO 88 eyes, 70 eyes: ON RNFL↑ C, SN, IN	Remaining subregions
González-Zamora and Bilbao-Malavé et al. [39]	DRI OCT Triton SS-OCT Angio, Topcon Medical Systems Inc.	25 (25 eyes) 25 (25 eyes)	2 weeks after hospital discharge	GCL ↓ foceal, central ON RNFL ↑	RT foveal, central RNFL foveal, central CT foveal, central
Bilbao-Malavé and González-Zamora et al. [27]	DRI OCT Triton SS-OCT Angio, Topcon Medical Systems Inc.	17 (17 eyes) 17 (17 eyes)	6 months after first examination	None	FT CCT ON RNFL
Abrishami et al. [44]	AngioVue system RTVue XR Avanti, Optovue, Fremont, CA, USA	30 (15 eyes) 60 (30 eyes)	At least 2 weeks asymptomatic	None	pRNFL ONH
Follow-up vs. recovered COVID-19 participants
Bilbao-Malavé and González-Zamora et al. [27]	DRI OCT Triton SS-OCT Angio, Topcon Medical Systems Inc.	17 (33 eyes) 17 (33 eyes)	6 months after first examination 2 weeks after hospital discharge	RNFL ↓ parafoveal GCL ↓ parafoveal ON RNFL ↓ FT ↑	CCT
Acute COVID-19 infection vs. healthy control
Jevnikar et al. [35]	SS-OCT, Topcon DRI OCT Triton; Topcon Corp., Tokyo, Japan	75 (75 eyes) 101 (101 eyes)	0	Severe (n = 59): mRNFL ↑ S, I GCL ↑ TO Mild (n = 16): none	Severe (n = 59): Remaining subregions RT Mild (n = 16): mRNFL GCL RT
Koçkar et al. [40]	RTVue-100 OCT, Optovue Inc, Fremont, CA	20 (40 eyes) 20 (40 eyes)	0	None	RNFL MT GCC
Acute COVID-19 infection vs. follow-up
Jevnikar et al. [29]	SS-OCT, Topcon DRI OCT Triton; Topcon Corp., Tokyo, Japan	30 (30 eyes) 30 (30 eyes)	0 1 year after first examination	mRNFL ↓ II, IO, NO, SO	Remaining subregions
Acute COVID-19 infection vs. follow-up & healthy control
Bayram et al. [25]	Spectral domain, Spectralis (HRA + OCT) Heidelberg Engineering Spectralis, Heidelberg, Germany	53 (106 eyes) 53 (106 eyes) 53 (106 eyes)	0 3 months after first examination	pRNFL ↑ OPL ↑ CT ↑	mRNFL GCL IPL INL ONL ORL RPE CMT

** Information of the number of eyes included is not evident from the respective publication. Abbreviations of examined parameters: ↑ = increased, ↓ = decreased, OCT = optical coherence tomography CCT = central choroid thickness, CFT = central foveal thickness, CMT = central macular thickness, CT = choroidal thickness, GCC = ganglion cell complex, GC-IPL = ganglion cell—inner plexiform layer, GCL = ganglion cell layer, INL, = inner nuclear layer, IPL = inner plexiform layer, mRNFL = macular retinal nerve fiber layer, MT = macular thickness, ON = optic nerve, ONH = optic nerve head, ONL = outer nuclear layer, OPL = outer plexiform layer, ORL = outer retinal layers, pRNFL = peripapillary retinal nerve fiber layer, RNFL = retinal nerve fiber layer, RT = retinal thickness, RPE = retinal pigment epithelium; Abbreviations of subdivisions of the parameters, including the ETDRS-scale (Figure A1a): C = central, I = inferior, II = inferior inner, IO = inferior outer, NO = nasal outer, O = outer, SI = superior inner, SO = superior outer, T = temporal, TI = temporal inner, TO = temporal outer and pRNFL subdivisions (Figure A1b): IT = inferotemporal, S = superior, C = central, IN = inferonasal.

Table 1. Overview of all included publications on OCT changes in chorioretinal parameters. The table includes the OCTA device used for examination, the number of participants in each cohort, the time of examination, and a summary of their outcomes. The results refer to the group of participants stated first in the headlines.

Publication	Device	Number of Participants	Time of Examination	Results: Significant Changes	Results: No Significant Changes
Long COVID Participants vs. Healthy Controls
Kanra et al. [20]	Spectralis, Heidelberg Engineering, Heidelberg, Germany	20 (34 eyes) 23 (39 eyes)	At least 4 weeks after completed treatment	mRNFL ↓ SI GCL ↓ SI, SO, IO, TO IPL ↓ SI, SO, TI, TO, NI	Remaining subregions pRNFL CMT
Dağ Şeker et al. [21]	SD-OCT Heidelberg Engineering, Heidelberg, Germany	27 (54 eyes) 27 (54 eyes)	31.18 ± 12.35 days after recovery	RT ↓ I, O mRNFL ↓ O GC-IPL ↓ I ONL ↓ C, I pRNFL ↓ C, NI	Remaining subregions INL OPL
			12 months later	mRNFL ↓O ONL ↓ I pRNFL ↓ C, IN	Remaining subregions GC-IPL INL OPL
					None in longitudinal comparison
Recovered COVID-19 participants vs. healthy controls
Dipu et al. [30]	Spectral domain RS-3000 LITE, NIDEK Inc.	35 (70 eyes) 12 (24 eyes)	4 to 6 weeks after hospital discharge	GC-IPL ↓	pRNFL mRNFL
Mavi Yildiz et al. [31]	Spectral domain, Spectralis (HRA + OCT) Heidelberg Engineering	63 (119 eyes) 59 (117 eyes)	2 to 8 weeks after positive real-time RT-PCR	pRNFL ↓ IT, ↑ S ONL ↑ CFT ↑	Remaining subregions mRNFL GCL IPL INL OPL RPE
Savastano et al. [42]	Zeiss Cirrus 5000-HD-OCT Angioplex, Carl Zeiss, Meditec, Inc.	80 ** 30 **	1 month from hospital discharge	None	mRNFL GCC CFT
Cennamo et al. [33]	Spectral domain-OCT AngioVue, Optovue Inc.	40 (40 eyes) 40 (40 eyes)	6 months after hospital discharge	ON RNFL ↓	CFT
Szkodny et al. [34]	Swept Source OCT-DRI OCT Triton, Topcon Inc.	78 (156 eyes) 49 (98 eyes)	1 to 4 months after recovery	None	mRNFL ON
Burgos-Blasco et al. [24]	Spectralis Spectralis, Heidelberg Engineering, Heidelberg, Germany	90 (90 eyes) 70 (70 eyes)	4 weeks	pRNFL ↑ mRNFL↓ SI, NI, NO GCL ↑ SO, NO, IO 88 eyes, 70 eyes: ON RNFL↑ C, SN, IN	Remaining subregions
González-Zamora and Bilbao-Malavé et al. [39]	DRI OCT Triton SS-OCT Angio, Topcon Medical Systems Inc.	25 (25 eyes) 25 (25 eyes)	2 weeks after hospital discharge	GCL ↓ foceal, central ON RNFL ↑	RT foveal, central RNFL foveal, central CT foveal, central
Bilbao-Malavé and González-Zamora et al. [27]	DRI OCT Triton SS-OCT Angio, Topcon Medical Systems Inc.	17 (17 eyes) 17 (17 eyes)	6 months after first examination	None	FT CCT ON RNFL
Abrishami et al. [44]	AngioVue system RTVue XR Avanti, Optovue, Fremont, CA, USA	30 (15 eyes) 60 (30 eyes)	At least 2 weeks asymptomatic	None	pRNFL ONH
Follow-up vs. recovered COVID-19 participants
Bilbao-Malavé and González-Zamora et al. [27]	DRI OCT Triton SS-OCT Angio, Topcon Medical Systems Inc.	17 (33 eyes) 17 (33 eyes)	6 months after first examination 2 weeks after hospital discharge	RNFL ↓ parafoveal GCL ↓ parafoveal ON RNFL ↓ FT ↑	CCT
Acute COVID-19 infection vs. healthy control
Jevnikar et al. [35]	SS-OCT, Topcon DRI OCT Triton; Topcon Corp., Tokyo, Japan	75 (75 eyes) 101 (101 eyes)	0	Severe (n = 59): mRNFL ↑ S, I GCL ↑ TO Mild (n = 16): none	Severe (n = 59): Remaining subregions RT Mild (n = 16): mRNFL GCL RT
Koçkar et al. [40]	RTVue-100 OCT, Optovue Inc, Fremont, CA	20 (40 eyes) 20 (40 eyes)	0	None	RNFL MT GCC
Acute COVID-19 infection vs. follow-up
Jevnikar et al. [29]	SS-OCT, Topcon DRI OCT Triton; Topcon Corp., Tokyo, Japan	30 (30 eyes) 30 (30 eyes)	0 1 year after first examination	mRNFL ↓ II, IO, NO, SO	Remaining subregions
Acute COVID-19 infection vs. follow-up & healthy control
Bayram et al. [25]	Spectral domain, Spectralis (HRA + OCT) Heidelberg Engineering Spectralis, Heidelberg, Germany	53 (106 eyes) 53 (106 eyes) 53 (106 eyes)	0 3 months after first examination	pRNFL ↑ OPL ↑ CT ↑	mRNFL GCL IPL INL ONL ORL RPE CMT

** Information of the number of eyes included is not evident from the respective publication. Abbreviations of examined parameters: ↑ = increased, ↓ = decreased, OCT = optical coherence tomography CCT = central choroid thickness, CFT = central foveal thickness, CMT = central macular thickness, CT = choroidal thickness, GCC = ganglion cell complex, GC-IPL = ganglion cell—inner plexiform layer, GCL = ganglion cell layer, INL, = inner nuclear layer, IPL = inner plexiform layer, mRNFL = macular retinal nerve fiber layer, MT = macular thickness, ON = optic nerve, ONH = optic nerve head, ONL = outer nuclear layer, OPL = outer plexiform layer, ORL = outer retinal layers, pRNFL = peripapillary retinal nerve fiber layer, RNFL = retinal nerve fiber layer, RT = retinal thickness, RPE = retinal pigment epithelium; Abbreviations of subdivisions of the parameters, including the ETDRS-scale (Figure A1a): C = central, I = inferior, II = inferior inner, IO = inferior outer, NO = nasal outer, O = outer, SI = superior inner, SO = superior outer, T = temporal, TI = temporal inner, TO = temporal outer and pRNFL subdivisions (Figure A1b): IT = inferotemporal, S = superior, C = central, IN = inferonasal.

3.2. OCTA (Table 2)

3.2.1. Long COVID

Schlick and Lucio et al. [22] and Szewczykowski and Mardin et al. [23] found a significant decrease in vessel density (VD) in the ICP and Szewczykowski and Mardin et al. [22] additionally found a decrease in all capillary plexuses of the retina in long COVID patients.

3.2.2. Recovered COVID-19 Participants

All but two [36,43] studies in which recovered patients were compared with healthy controls identified a significant reduction in VD in the superficial capillary plexus (SCP) [26,27,28,30,33,34,37,38,39], with eight out of nine examined groups also showing a significant decrease in the DCP [26,27,28,30,33,37,38,39]. The study by Bilbao-Malavé and González-Zamora et al. [28,42], however, showed no significant change. In follow-up observations, a significant decrease in the VD of the SCP and DCP was seen by half of the studies [28,37,41], with Abrishami et al. [41] showing the decrease in the DCP at 1 month, as well as 3 months after initial recovery. Five out of nine publications comparing recovered patients with healthy controls showed a significant increase in FAZ [27,30,37,38,39]. In the follow-up study, an increase in FAZ was seen in the superficial area by Bilbao-Malavé and González-Zamora et al. [27] and in the DCP by Kal et al. [37]. The remaining three follow-up studies revealed no significant changes in FAZ between the groups [28,32,41].

3.2.3. Acute COVID-19 Infection

Jevnikar et al. [29,35] found no significant changes in VD and FAZ parameters in acute patients compared to either their follow-up examination after 1 year or healthy controls.

Table 2. Overview of all included publications on OCTA changes in chorioretinal parameters. The table includes the OCTA device used for examination, the number of participants in each cohort, the time of examination, and a summary of their outcomes. The results refer to the group of participants stated first in the headlines.

Publication	Device	Number of Participants	Time of Examination	Results: Significant Changes	Results: No Significant Changes
Long COVID Participants vs. Healthy or Recovered * Controls
Szewczykowski and Mardin et al. [23]	Heidelberg Spectralis II, Heidelberg, Germany	48 (92 eyes) 6 (9 eyes)	200 ± 110 days (34-484 days) after confirmed SARS-CoV-2 infection	VD SVP ↓ VD ICP ↓ VD DCP ↓	none
Schlick and Lucio et al. [22]	Heidelberg Spectralis II, Heidelberg, Germany	173 ** 28 **	231± 111 days of post-COVID-19 symptom persistency	VD ICP ↓	VD SVP VD DCP
Recovered COVID-19 participants vs. healthy controls
Dipu et al. [30]	Spectral domain RS-3000 LITE, NIDEK Inc.	35 (70 eyes) 12 (24 eyes)	4 to 6 weeks	VD SCP ↓ VD DCP ↓ FAZ ↑	VD RPCP
Hazar et al. [26]	Optovue Angiovue, Optovue Inc.	50 ** 55 **	1 month after hospital discharge	VD SCP ↓ parafoveal I, S VD DCP ↓ parafoveal S	FAZ
Cennamo et al. [33]	Optovue Angiovue, Optovue Inc.	40 (40 eyes) 40 (40 eyes)	6 months after hospital discharge	VD SCP ↓ VD DCP ↓ VD RPCP ↓	FAZ
Szkodny et al. [34]	Swept Source OCT-DRI OCT Triton, Topcon Inc.	78 (156 eyes) 49 (98 eyes)	1 to 4 months after recovery	VD SCP ↓	FAZ
Abrishami et al. [38]	Optovue Angiovue, Optovue Inc.	31 (31 eyes) 23 (23 eyes)	≥2 months after recovery	VD SCP ↓ foveal, parafoveal I VD DCP ↓ foveal FAZ ↑	VD DCP parafoveal
Turker et al. [28]	Optovue Angiovue, Optovue Inc.	25 (50 eyes) 25 (50 eyes)	6 months after hospital discharge	VD SCP ↓ parafoveal VD DCP ↓ parafoveal S, I	VD SCP foveal VD DCP foveal FAZ
		25 (50 eyes 25 (50 eyes)	shortly after hospital discharge	VD SCP ↓ parafoveal VD DCP ↓ parafoveal	VD SCP foveal VD DCP foveal FAZ
González-Zamora and Bilbao-Malavé et al. [39]	DRI OCT Triton SS-OCT Angio, Topcon Medical Systems Inc.	25 (25 eyes) 25 (25 eyes)	2 weeks after hospital discharge	VD SCP ↓ foveal, parafoveal VD DCP ↓ foveal FAZ ↑ SCP	VD DCP parafoveal FAZ DCP, CC
Bilbao-Malavé and González-Zamora et al. [27]	DRI OCT Triton SS-OCT Angio, Topcon Medical Systems Inc.	17 (17 eyes) 17 (17 eyes)	6 months after first examination	VD SCP ↓ foveal FAZ ↑ superficial	VD SCP parafoveal VD DCP foveal, parafoveal
Kal et al. [36]	DRI-OCT Triton Topcon Inc., Tokyo, Japan	63 (120 eyes) 43 (83 eyes)	2 months after hospital discharge	none	VD RPCP VD ONH
Kal et al. [37]	Swept Source DRI-OCT Triton SS-OCT Angio, Topcon Inc., Tokyo, Japan	49 (75 eyes) 43 (83 eyes)	8 months after hospital discharge	VD SCP ↓ S, N, I, T VD DCP ↓ FAZ ↑ VD CC ↓ S, I, T	remaining subregions
Hohberger and Ganslmayer et al. [43]	Heidelberg Spectralis II, Heidelberg, Germany	33 (33 eyes) 28 (28 eyes)	34–281 days, 138.13 ± 70.67 days after positive SARS-CoV-2 PCR test	VD ICP ↓ VD peripapillary ↓	VD SVP VD DCP
Follow-up vs. recovered COVID-19 participants
Turker et al. [28]	Optovue Angiovue, Optovue Inc.	25 (50 eyes) 25 (50 eyes)	6 months vs. shortly after recovery	VD SCP ↓ parafoveal VD DCP ↓ parafoveal	VD SCP foveal VD DCP foveal FAZ
Bilbao-Malavé and González-Zamora et al. [27]	DRI OCT Triton SS-OCT Angio, Topcon Medical Systems Inc.	17 (33 eyes) 17 (33 eyes)	6 months vs. 2 weeks after hospital discharge	FAZ ↑ superficial	VD SCP parafoveal, foveal VD DCP parafoveal, foveal
Abrishami et al. [41]	AngioVue system RTVue XR Avanti, Optovue, Fremont, CA, USA	18 (36 eyes) 18 (36 eyes)	1 month vs. shortly after recovery	VD DCP↓ mean, parafoveal, perifoveal	VD SCP FAZ
			3 months vs. shortly after recovery	VD SCP ↓ inferior hemifield, superior region, inferior region VD DCP ↓ mean, parafoveal, perifoveal	VD SCP FAZ
Kal et al. [37]	Swept Source DRI-OCT Triton SS-OCT Angio, Topcon Inc., Tokyo, Japan	49 (75 eyes) 63 (120 eyes)	8 months vs. 2 months after hospital discharge	VD SCP ↓ F, S, N, I VD DCP ↓ F, S, N, I VD CC ↓ S, N, I, T FAZ ↑ DCP	remaining subregions
Savastano et al. [32]	Zeiss Cirrus 5000-HD-OCT Angioplex, Carl Zeiss, Meditec, Inc.	70 (70 eyes) 22 (22 eyes)	1 month after hospital discharge and 2 months from symptom onset	none	VD SCP VD DCP FAZ
Acute COVID-19 infection vs. healthy control
Jevnikar et al. [35]	SS-OCT, Topcon DRI OCT Triton; Topcon Corp., Tokyo, Japan	75 (75 eyes) 101 (101 eyes)	0	none	VD SCP VD DCP FAZ
Acute COVID-19 infection vs. follow-up
Jevnikar et al. [29]	Automated Retinal Image Analyser, Topcon Corp.	30 (30 eyes) 30 (30 eyes)	0 1 year after first examination	none	VD SCP parafoveal VD DCP parafoveal FAZ

* Information is not evident from the respective publication; ** Information of the number of eyes included is not evident from the respective publication; Abbreviations of examined parameters: ↑ = increased, ↓ = decreased, OCTA = optical coherence tomography angiography, CC = choriocapillaris, DCP = deep capillary plexus, FAZ = foveal avascular zone, ICP = intermediate capillary plexus, RPCP = radial peripapillary capillary plexus, SCP = superficial capillary plexus, SVP = superficial vascular plexus, VD = vessel density; RT-PCR = reverse transcription-polymerase chain reaction, RPC = radial peripapillary capillary; ONH = optic nerve head Quadrants; Abbreviations of subdivisions of the parameters I = inferior, N = nasal, S = superior, T = temporal.

Table 2. Overview of all included publications on OCTA changes in chorioretinal parameters. The table includes the OCTA device used for examination, the number of participants in each cohort, the time of examination, and a summary of their outcomes. The results refer to the group of participants stated first in the headlines.

Publication	Device	Number of Participants	Time of Examination	Results: Significant Changes	Results: No Significant Changes
Long COVID Participants vs. Healthy or Recovered * Controls
Szewczykowski and Mardin et al. [23]	Heidelberg Spectralis II, Heidelberg, Germany	48 (92 eyes) 6 (9 eyes)	200 ± 110 days (34-484 days) after confirmed SARS-CoV-2 infection	VD SVP ↓ VD ICP ↓ VD DCP ↓	none
Schlick and Lucio et al. [22]	Heidelberg Spectralis II, Heidelberg, Germany	173 ** 28 **	231± 111 days of post-COVID-19 symptom persistency	VD ICP ↓	VD SVP VD DCP
Recovered COVID-19 participants vs. healthy controls
Dipu et al. [30]	Spectral domain RS-3000 LITE, NIDEK Inc.	35 (70 eyes) 12 (24 eyes)	4 to 6 weeks	VD SCP ↓ VD DCP ↓ FAZ ↑	VD RPCP
Hazar et al. [26]	Optovue Angiovue, Optovue Inc.	50 ** 55 **	1 month after hospital discharge	VD SCP ↓ parafoveal I, S VD DCP ↓ parafoveal S	FAZ
Cennamo et al. [33]	Optovue Angiovue, Optovue Inc.	40 (40 eyes) 40 (40 eyes)	6 months after hospital discharge	VD SCP ↓ VD DCP ↓ VD RPCP ↓	FAZ
Szkodny et al. [34]	Swept Source OCT-DRI OCT Triton, Topcon Inc.	78 (156 eyes) 49 (98 eyes)	1 to 4 months after recovery	VD SCP ↓	FAZ
Abrishami et al. [38]	Optovue Angiovue, Optovue Inc.	31 (31 eyes) 23 (23 eyes)	≥2 months after recovery	VD SCP ↓ foveal, parafoveal I VD DCP ↓ foveal FAZ ↑	VD DCP parafoveal
Turker et al. [28]	Optovue Angiovue, Optovue Inc.	25 (50 eyes) 25 (50 eyes)	6 months after hospital discharge	VD SCP ↓ parafoveal VD DCP ↓ parafoveal S, I	VD SCP foveal VD DCP foveal FAZ
		25 (50 eyes 25 (50 eyes)	shortly after hospital discharge	VD SCP ↓ parafoveal VD DCP ↓ parafoveal	VD SCP foveal VD DCP foveal FAZ
González-Zamora and Bilbao-Malavé et al. [39]	DRI OCT Triton SS-OCT Angio, Topcon Medical Systems Inc.	25 (25 eyes) 25 (25 eyes)	2 weeks after hospital discharge	VD SCP ↓ foveal, parafoveal VD DCP ↓ foveal FAZ ↑ SCP	VD DCP parafoveal FAZ DCP, CC
Bilbao-Malavé and González-Zamora et al. [27]	DRI OCT Triton SS-OCT Angio, Topcon Medical Systems Inc.	17 (17 eyes) 17 (17 eyes)	6 months after first examination	VD SCP ↓ foveal FAZ ↑ superficial	VD SCP parafoveal VD DCP foveal, parafoveal
Kal et al. [36]	DRI-OCT Triton Topcon Inc., Tokyo, Japan	63 (120 eyes) 43 (83 eyes)	2 months after hospital discharge	none	VD RPCP VD ONH
Kal et al. [37]	Swept Source DRI-OCT Triton SS-OCT Angio, Topcon Inc., Tokyo, Japan	49 (75 eyes) 43 (83 eyes)	8 months after hospital discharge	VD SCP ↓ S, N, I, T VD DCP ↓ FAZ ↑ VD CC ↓ S, I, T	remaining subregions
Hohberger and Ganslmayer et al. [43]	Heidelberg Spectralis II, Heidelberg, Germany	33 (33 eyes) 28 (28 eyes)	34–281 days, 138.13 ± 70.67 days after positive SARS-CoV-2 PCR test	VD ICP ↓ VD peripapillary ↓	VD SVP VD DCP
Follow-up vs. recovered COVID-19 participants
Turker et al. [28]	Optovue Angiovue, Optovue Inc.	25 (50 eyes) 25 (50 eyes)	6 months vs. shortly after recovery	VD SCP ↓ parafoveal VD DCP ↓ parafoveal	VD SCP foveal VD DCP foveal FAZ
Bilbao-Malavé and González-Zamora et al. [27]	DRI OCT Triton SS-OCT Angio, Topcon Medical Systems Inc.	17 (33 eyes) 17 (33 eyes)	6 months vs. 2 weeks after hospital discharge	FAZ ↑ superficial	VD SCP parafoveal, foveal VD DCP parafoveal, foveal
Abrishami et al. [41]	AngioVue system RTVue XR Avanti, Optovue, Fremont, CA, USA	18 (36 eyes) 18 (36 eyes)	1 month vs. shortly after recovery	VD DCP↓ mean, parafoveal, perifoveal	VD SCP FAZ
			3 months vs. shortly after recovery	VD SCP ↓ inferior hemifield, superior region, inferior region VD DCP ↓ mean, parafoveal, perifoveal	VD SCP FAZ
Kal et al. [37]	Swept Source DRI-OCT Triton SS-OCT Angio, Topcon Inc., Tokyo, Japan	49 (75 eyes) 63 (120 eyes)	8 months vs. 2 months after hospital discharge	VD SCP ↓ F, S, N, I VD DCP ↓ F, S, N, I VD CC ↓ S, N, I, T FAZ ↑ DCP	remaining subregions
Savastano et al. [32]	Zeiss Cirrus 5000-HD-OCT Angioplex, Carl Zeiss, Meditec, Inc.	70 (70 eyes) 22 (22 eyes)	1 month after hospital discharge and 2 months from symptom onset	none	VD SCP VD DCP FAZ
Acute COVID-19 infection vs. healthy control
Jevnikar et al. [35]	SS-OCT, Topcon DRI OCT Triton; Topcon Corp., Tokyo, Japan	75 (75 eyes) 101 (101 eyes)	0	none	VD SCP VD DCP FAZ
Acute COVID-19 infection vs. follow-up
Jevnikar et al. [29]	Automated Retinal Image Analyser, Topcon Corp.	30 (30 eyes) 30 (30 eyes)	0 1 year after first examination	none	VD SCP parafoveal VD DCP parafoveal FAZ

* Information is not evident from the respective publication; ** Information of the number of eyes included is not evident from the respective publication; Abbreviations of examined parameters: ↑ = increased, ↓ = decreased, OCTA = optical coherence tomography angiography, CC = choriocapillaris, DCP = deep capillary plexus, FAZ = foveal avascular zone, ICP = intermediate capillary plexus, RPCP = radial peripapillary capillary plexus, SCP = superficial capillary plexus, SVP = superficial vascular plexus, VD = vessel density; RT-PCR = reverse transcription-polymerase chain reaction, RPC = radial peripapillary capillary; ONH = optic nerve head Quadrants; Abbreviations of subdivisions of the parameters I = inferior, N = nasal, S = superior, T = temporal.

4. Discussion

Long COVID prompts interest in potential diagnostic tools like OCT and OCTA, but varied outcomes from multiple studies reveal inconsistencies.

The thickness of the mRFNL was decreased in all cohort groups, with significant results [20,24,27,29] compared to their controls. The thickness of the IPL decreased in one recovered cohort [30] as well as in the long COVID cohort [20], while the thickness of the GCL had opposite results. It was significantly thinner in the long COVID cohort [20] and in participants at six months compared to those at two weeks after COVID-19 infection [27]. In the recovered cohort compared to healthy controls, Dipu et al. [30] found significant thinning of the GC-IPL, while Burgos-Blasco et al. [24] observed significant thickening in three outer quadrants of the GCL. Changes in the retinal layers are detectable in inflammatory processes and vary in thickness depending on the stage of inflammation [14]. Thickening of the ONL, as observed by Mavi-Yildiz et al. [31], and thinning of the RNFL, GCL, IPL, INL, and OPL can be attributed to previous precapillary arterial ischemia [45]. Thinning of the GC-IPL, which Kanra et al. [20] ascribe to retrograde transsynaptic loss of retinal ganglion cells, and thinning of the RNFL can also be observed in Alzheimer’s disease [46] and multiple sclerosis [47]. Diabetes mellitus type 1 and 2 as a catalyst for ocular complications [48], diseases of the optic nerve head such as glaucoma, which lead to thinning of the pRNFL [49], and arterial hypertension that damages the retinal microcirculation and the RNFL [50] may also confound research results. These pre-existing conditions should generally be excluded to obtain the most reliable and comparable results. Smoking also has a confounding impact on the microvessels of both chronic smokers [51] and individuals shortly after smoking a cigarette [52]. Except for Jevnikar et al. [29], who excluded people with a history of smoking from their study, and Hazar et al. [26], who reported the number of smokers in each cohort, smoking history was not included in any of the publications.

Microcirculation may also have been affected by whether oxygenation was required. Hospitalized patients most likely underwent therapy that, according to the guidelines, included the administration of steroids and other medications [53] that have a positive effect on reducing inflammation but can distort the OCT results.

In the OCTA outcome publications with significant results, vascular density was congruently reduced. It is unclear whether there is a higher gap between the decrease in VD in long COVID patients and recovered patients, as Schlick and Lucio et al. [22] and Szewczykowski and Mardin et al. [23] did not specify whether their control groups consisted exclusively of recovered COVID-19 participants or healthy people who assumably never have had COVID-19.

The enlargement of FAZ parameters [27,30,38] indicate microvessel dysfunction with impaired blood flow, potentially related to hypercoagulative reactions during inflammation [29]. Cennamo et al. [33] associated the decrease in recovered participants to eventual microvascular damage during SARS-CoV-2 entry and endothelial inflammation. Angiotensin-converting enzyme 2 (ACE2) receptors, expressed in various eye structures [33], play a role in metabolizing angiotensin II (ANG II), a vasoconstrictor [54]. ANG II is also considered a probable regulator of retinal blood flow [55]. Disruption of ACE2 may therefore impact retinal blood flow autoregulation and subsequently affect VD in OCTA [56]. Significantly reduced autoregulation was found in diabetes mellitus patients by Zhang et al. [57], who revealed an increase in the parameters in the FAZ and a decrease in the perfusion density of the macular SCP and DCP [57].

Furthermore, signs of ischemia and persistent inflammation, such as cotton wool spots, were found during acute SARS-CoV-2 infection [35,58,59] and post-SARS-CoV-2 infection in several publications [27,30,59,60,61]. Additionally, Koçkar et al.[40] found hyperreflective lesions in the multiple retinal layers of the macula and Marinho et al. [59] discovered them in the INL and GCL within approximately one month after COVID-19 infection in all of their 12 participants [59].

OCTA has the advantage of visualizing microvessels in the choroid and retina without being invasive, as was previously the case with fluorescein angiography [17] and indocyanine green angiography [18]. Because OCTA senses erythrocyte movement, it cannot detect leakage from blood vessels and is inferior to the other two techniques in this regard [62]. Using similar search terms, as of May 2023 there are no published studies on COVID-19 or long COVID using fluorescein or indocyanine green angiography. Variations in OCT- and OCTA-devices, software, and differently experienced analysts limit result comparability. A wide age range for the examined participants, from 18 to 85 years, may lead to variability in outcomes due to physiological aging processes. There was variance in the selection of participants with different exclusion criteria, which could potentially affect ocular manifestations (Table A1). Another limitation of this review is that it comprises all publications with COVID-19-related participants, regardless of the severity and the previous treatment of the disease. However, reporting bias was positively affected by the great interest in finding solutions in managing the disease during the pandemic. Additionally, with a focus on imaging biomarkers and their presence in participants with active, prolonged, or historic COVID-19, sensitivity analyses and certainty assessments for outcomes were not evaluated. Given the small number of included publications, summary statistics and effect estimates were not analyzed. In all studies that examined either OCT or OCTA, the most stated limitation was small sample size. Finally, potential bias may have been caused by an incomplete identification of all relevant publications in our database search.

5. Conclusions

Changes in retinal layers were observed at various COVID-19 infection stages, before, during, and after infection and across different participant groups. The significant morphological OCT atrophy in RNFL, GCL, and IPL of the long COVID cohort was limited in comparability with published findings of recovered cohorts due to diverse inclusion and exclusion criteria, varying severity of the disease and treatment, and small sample sizes. OCTA outcomes revealed that, when significant, VD was consistently reduced in both long COVID and recovered individuals. FAZ, which was not examined in the long COVID analysis, was significantly enlarged in almost half of the studies with participants after recovery. Future studies on long COVID should include measurements of the FAZ area. To enhance OCT and OCTA validity as a diagnostic tool for long COVID, further research on long COVID with a focus on longitudinal studies, larger and more diverse sample sizes, clear comorbidity specifications in research, as well as control groups and simultaneous blood pressure measurement is advisable. These efforts have the potential to advance our understanding and improve long COVID diagnosis and management. [35]