Although some concerns have been raised regarding the reliability of choriocapillaris imaging with optical coherence tomography angiography (OCTA), until now, this is the only technique that allows high-resolution in vivo visualization of choriocapillaris.1,2 Several OCTA studies have shown that alteration of choriocapillaris may be one of the earliest changes of intraocular microcirculation that is associated with systemic cardiovascular risk factors, including diabetic retinopathy and hypertension.3,4 Dai has recently shown that the perfusion of the choriocapillaris decreases even before detectable changes in retinal microcirculation in diabetic patients;5 however, it is not known if the changes of choriocapillaris are related to other retinal vascular conditions. Changes in retinal microcirculation in fellow eyes of patients with unilateral retinal vein occlusions (RVO) were shown,6 but little is known about status of choriocapillaris in those eyes.
In a recent study, small resolved paracentral acute middle maculopathy (PAMM) lesions were reported to be highly prevalent in fellow eyes of patients with unilateral RVO.7 Since PAMM is considered one of the earliest and mildest form of retinal ischemia,8–11 we suggest that some level of association may be found between this phenomena and alteration of choriocapillaris in those eyes.
The aim of this study was therefore to evaluate the status of the choriocapillaris in fellow eyes of patients with unilateral RVO and elucidate its possible association with the presence of resolved PAMM lesions.
Patients and Methods
This was a single-center, case-control study that included patients with unilateral RVO and healthy individuals. Exclusion criteria were myopia more than 6.0 diopters, any diseases of the posterior eye segment, history of stroke or myocardial infarction, diabetes mellitus, decreased transparency of optical media or quality of OCTA image Q7 or lower, or the presence of retinal vein or arterial occlusion in both eyes of RVO patients. Healthy individuals with resolved PAMM lesions were also excluded. To reduce the effect of age-related changes of the choriocapillaris metric on the results of this study, only individuals younger than 65 years were included. The study followed the ethical standards stated in the Declaration of Helsinki and was approved by the Local Ethics Committee.
In this prospective study, we included consecutive patients with unilateral branch retinal vein occlusion (BRVO) or central retinal vein occlusion (CRVO) who had visited our clinic from February 2018 to December 2019. All healthy participants were referred from the cardiological department after a comprehensive cardiological examination.10
All patients received a comprehensive ophthalmic examination and OCT with RTVue-XR Avanti (Optovue, Fremont, CA) running software version 2017.1.0.150. All imaging procedures were performed after mydriasis had been achieved with topical administration of 1% tropicamide. OCT examination included 3-mm (304 2×B-scans each of 304 A-scans), and 6-mm (400 2×B-scans each of 400 A-scans) OCTA scans centered on the center of the fovea.
Values of superficial capillary plexus (SCP) and deep capillary plexus (DCP) vessel density for the whole image as well as foveal avascular zone (FAZ) area were extracted from 3-mm OCTA scans of healthy eyes of RVO patients and right eyes of healthy participants. Only default segmentation was used to evaluate OCTA parameters for both the SCP and DCP. The SCP slab was taken between the segmentation lines of the inner limiting membrane and the inner plexiform layer with 0 and −9 µm offset, respectively. The DCP slab was taken between the segmentation lines of the inner plexiform layer and the outer plexiform layer with −9 and 9 µm offset, respectively.
Additionally, the device-specific total choriocapillaris flow area was calculated for 3-mm OCTA scan in a standard choriocapillaris slab using the “flow area” tool with a radius of 2.25 mm covering the whole scan area. Using this tool, the device-specific flow area for the whole 3-mm image was obtained and used for further analysis.
Since the metric of choriocapillaris flow follows power-law distribution, the number of flow voids was calculated as described by Spaide.10 Briefly, customized choriocapillaris slabs were constructed with two Bruch's membrane segmentation lines with −31 and −41 µm offset.12–14 The posterior offset was chosen to avoid potential segmentation errors, which may cause false hypointense signals due to the inclusion of the retinal pigment epithelium layer into the slab. Further, for each eye 600 × 600 pixels image of the choriocapillaris customized slab was upload in ImageJ and subjected to auto-thresholding with Phansalkar method followed by binarization.15 For the resultant image, after excluding projections of large retinal vessels, the number of voids larger than 25,000 µm2 and their total area was calculated using “analyze particles” function. This relatively large void size was chosen since the analysis of larger voids allow better discrimination for choriocapillaris damage.14
Identification of resolved PAMM lesions was performed using 6-mm OCTA scans. For this, each volume scan was independently reviewed by two masked specialists (DSM and MAB) by scrolling all cross-sectional scans in AngioVue software. A resolved PAMM lesion on a cross-sectional scan was defined as the area of the inner nuclear layer thinning associated with outer plexiform layer disruption/elevation, as has been described previously.7,10 Additionally, for each eye, the slab constructed using two segmentation lines of the outer plexiform layer with −9 µm and 0 µm offset was reviewed by two masked specialists independently. On en face image, a resolved PAMM lesion was defined as a small sharply delineated black area. The presence of at least one resolved PAMM lesion, confirmed by two specialists upon cross-sectional and en face images simultaneously, was sufficient for an individual to be considered as having resolved PAMM lesions (Figure 1). The number of resolved PAMM lesions was calculated by two specialists based on en face images, but only lesions confirmed with cross-sectional scans were counted. The interrater correlation coefficient was calculated for the number of resolved PAMM lesions identified by two specialists.
Representative example of visualizing resolved paracentral acute maculopathy lesions using 10-µm en face slab between two segmentation lines of the outer plexiform layer with −9 µm and 0 µm offset. The dotted line represents the position of the cross-sectional scan. Arrowheads indicate resolved paracentral acute maculopathy lesions on the cross-sectional scan.
Statistical analysis was performed in MedCalc 18.4.1 (MedCalc Software, Ostend, Belgium). All data are presented as a mean ± standard deviation. Oneway analysis of variance was used to compare continuous variables between RVO patients and healthy individuals. The chi-square test was used to compare categorical variables. The Pearson correlation coefficient was calculated for the whole study population to evaluate relationships between the number of resolved PAMM lesions and choriocapillaris parameters (total flow area, number of flow voids, and total flow voids area). In a multiple regression model predicting the number of resolved PAMM lesions the age, gender, vessel density of the SCP, vessel density of the DCP, area of the FAZ, and the device-specific total choriocapillaris total flow area were included.
Thirty-two RVO patients (23 BRVO and nine CRVO patients) and 16 healthy participants were included in this study. There was no statistically significant difference in age, male-to-female ratio, and smoking status between RVO patients and controls (Table 1).
Demographic and Clinical Characteristics of Study Groups
The number of flow voids and the total area of flow voids were statistically significantly higher, whereas the device-specific choriocapillaris total flow area was lower in the fellow eyes of unilateral RVO patients compared to the controls (P < .01). Vessel density of the SCP and DCP were statistically significantly lower (P < .05) in fellow eyes of unilateral RVO patients compared to the controls (Table 2). There was no statistically significant difference in FAZ area between fellow eyes of unilateral RVO patients and eyes of healthy participants (P = .67). The interrater correlation coefficient for the number of PAMM lesions identified by two specialists was 0.96 (95% confidence interval, 0.97−0.98).
Optical Coherence Tomography Angiography Parameters in Study Groups
There was a statistically significant correlation between the number of resolved PAMM lesions and the number of choriocapillaris flow voids and the total area of choriocapillaris flow voids (r = 0.44; P = .002) and (r = 0.43; P = .002), respectively. Additionally, there was a statistically significant negative correlation between the number of resolved PAMM lesions and the device-specific choriocapillaris total flow area (r = −0.52; P < .001) (Figure 2). There was no statistically significant correlation between the choriocapillaris total flow area and any retinal OCTA parameter, including vessel density of SCP, vessel density of DCP, and area of FAZ.
Correlation between status of choriocapillaris and number of resolved paracentral acute middle maculopathy (PAMM) lesions. Arrowheads indicate resolved PAMM lesions. BRVO = branch retinal vein occlusion.
The multiple regression model, including the number of resolved PAMM lesion dependent on the age, gender, vessel density of the SCP, vessel density of the DCP, area of the FAZ, and choriocapillaris total flow area showed that only vessel density of SCP and choriocapillaris total flow area are independent factor related to the number of PAMM lesions with β = −0.37 (P = .015) and β = −0.39 (P = .009), respectively.
The device-specific choriocapillaris total flow area showed a statistically significant correlation with the number of choriocapillaris flow voids (r = −0.37; P = .012) and with the total area of choriocapillaris flow voids (r = −0.75; P < .001).
This study showed that the choriocapillaris in fellow eyes of unilateral RVO patients is substantially altered compared to age-matched controls and its status correlates significantly with the number of resolved PAMM lesions in those eyes. Other retinal OCTA parameters have no correlation with alteration of choriocapillaris and we suggest, therefore, that alteration of the choriocapillaris and deep retinal ischemia may be the earliest concurrent processes in the retinal microcirculation resulting from the effects of systemic risk factors.
Alteration of the choriocapillaris in fellow eyes of unilateral RVO patients may be expected since the main risk factor of RVO, hypertension,16 compromises the choroid.17 Moreover, alteration of choriocapillaris was recently described in association with other systemic cardiovascular risk factors such as diabetes mellitus early in the course of this disease.6,7 PAMM has also been described as one of the earliest and the mildest forms of retinal ischemia.7–10 Additionally, small resolved PAMM lesions are reported to be highly prevalent in healthy eyes of patients with unilateral RVO and in the eyes of hypertensive patients.7,10 It is reasonable to conclude therefore that deep retinal microcirculation and choriocapillaris may demonstrate some level of correlation in subclinical alteration.18
Although the choriocapillaris and deep vascular complex are supplied from a common arterial source, they represent separate microcirculatory networks. However, simultaneous alteration of these structures can be explained since both are terminal microvascular networks having lobular architecture.9,19
Healthy eyes of unilateral RVO patients also demonstrate a significant reduction of vessel density of the SCP and DCP compared to healthy age-matched controls; however, these parameters are independent of changes of the choriocapillaris. This is to be expected since detectable changes in vessel density likely reflect a more severe level of retinal microcirculatory alteration following the appearance of small PAMM lesions and flow voids in the choriocapillaris.
The evaluation of the choriocapillaris requires specific image processing to count the number of flow voids.12 However, we found a significant correlation between the flow voids metric (the number and the total area of flow voids) and the device-specific choriocapillaris total flow area. Moreover, the number of PAMM lesions also correlates significantly with this device-specific choriocapillaris total flow area. All of which suggests that this relatively simple parameter may be considered for use in further studies.
This study has two substantial limitations. Firstly, we used spectral-domain OCT, which is considered to be inferior to swept source OCT in the evaluation of the choriocapillaris. This distinction between two methods results in the difference of absolute values when characterizing the status of choriocapillaris perfusion. However, this is unlikely to affect the main conclusion of the study, which accounts for the correlation of angiographic parameters. Secondly, the problem of evaluating changes in retinal and choroidal microcirculation is their age-dependent character . It is possible, therefore, to find some level of correlation between the alteration of deep retinal microcirculation and the choriocapillaris even in a healthy population if the age of the study cohort ranges widely. In this study, we included relatively young individuals within a narrow range of age to minimize age-related changes in the choriocapillaris. Moreover, multiple regression showed that the status of the choriocapillaris is an independent predictor for the number of resolved PAMM lesions, while age demonstrates no statistically significant correlation.
In conclusion, this study showed the alteration of the choriocapillaris in healthy eyes of unilateral RVO patients, including reduction of the total choriocapillaris flow area, increase in the number of flow voids and their total area. Choriocapillaris total flow area, the number of flow voids, and their total area correlate significantly with the number of resolved PAMM lesions. Therefore, simultaneous alteration of deep retinal circulation and the choriocapillaris may be the earliest changes in retinal microcirculation resulting from the effects of systemic risk factors.
- Kim DY, Fingler J, Zawadzki RJ, et al. Optical imaging of the chorioretinal vasculature in the living human eye. Proc Natl Acad Sci USA. 2013;110(35):14354–14359. doi:10.1073/pnas.1307315110 [CrossRef] PMID:23918361
- Uji A, Balasubramanian S, Lei J, Baghdasaryan E, Al-Sheikh M, Sadda SR. Choriocapillaris Imaging Using Multiple En Face Optical Coherence Tomography Angiography Image Averaging. JAMA Ophthalmol. 2017;135(11):1197–1204. doi:10.1001/jamaophthalmol.2017.3904 [CrossRef] PMID:28983552
- Chua J, Chin CWL, Tan B, et al. Impact of systemic vascular risk factors on the choriocapillaris using optical coherence tomography angiography in patients with systemic hypertension. Sci Rep. 2019;9(1):5819. doi:10.1038/s41598-019-41917-4 [CrossRef] PMID:30967575
- Conti FF, Qin VL, Rodrigues EB, et al. Choriocapillaris and retinal vascular plexus density of diabetic eyes using split-spectrum amplitude decorrelation spectral-domain optical coherence tomography angiography. Br J Ophthalmol. 2019;103(4):452–456. doi:10.1136/bjophthalmol-2018-311903 [CrossRef] PMID:29793926
- Dai Y, Zhou H, Chu Z, et al. Microvascular Changes in the Choriocapillaris of Diabetic Patients Without Retinopathy Investigated by Swept-Source OCT Angiography. Invest Ophthalmol Vis Sci. 2020;61(3):50. doi:10.1167/iovs.61.3.50 [CrossRef] PMID:32232345
- Adhi M, Filho MA, Louzada RN, et al. Retinal capillary network and foveal avascular zone in eyes with vein occlusion and fellow eyes analyzed with optical coherence tomography angiography. Invest Ophthalmol Vis Sci. 2016;57(9):OCT486–OCT494. doi:10.1167/iovs.15-18907 [CrossRef] PMID:27442342
- Maltsev DS, Kulikov AN, Burnasheva MA, Chhablani J. Prevalence of resolved paracentral acute middle maculopathy lesions in fellow eyes of patients with unilateral retinal vein occlusion. Acta Ophthalmol. 2020;98(1):e22–e28. doi:10.1111/aos.14196 [CrossRef] PMID:31347293
- Sarraf D, Rahimy E, Fawzi AA, et al. Paracentral acute middle maculopathy: a new variant of acute macular neuroretinopathy associated with retinal capillary ischemia. JAMA Ophthalmol. 2013;131(10):1275–1287. doi:10.1001/jamaophthalmol.2013.4056 [CrossRef] PMID:23929382
- Bakhoum MF, Freund KB, Dolz-Marco R, et al. Paracentral acute middle maculopathy and the ischemic cascade associated with retinal vascular occlusion. Am J Ophthalmol. 2018;195:143–153. doi:10.1016/j.ajo.2018.07.031 [CrossRef] PMID:30081014
- Burnasheva MA, Maltsev DS, Kulikov AN, Sherbakova KA, Barsukov AV. Association of Chronic Paracentral Acute Middle Maculopathy Lesions with Hypertension. Ophthalmol Retina. 2020;4(5):504–509. doi:10.1016/j.oret.2019.12.001 [CrossRef] PMID:31948908
- Kulikov AN, Maltsev DS, Leongardt TA. Retinal Microvasculature Alteration in Paracentral Acute Middle Maculopathy and Acute Macular Neuroretinopathy: A Quantitative Optical Coherence Tomography Angiography Study. Retin Cases Brief Rep. 2020;14(4):343–351. doi:10.1097/ICB.0000000000000709 [CrossRef] PMID:29443808
- Spaide RF. Choriocapillaris flow features follow a power law distribution: implications for characterization and mechanisms of disease progression. Am J Ophthalmol. 2016;170:58–67. doi:10.1016/j.ajo.2016.07.023 [CrossRef] PMID:27496785
- Nassisi M, Baghdasaryan E, Tepelus T, Asanad S, Borrelli E, Sadda SR. Topographic distribution of choriocapillaris flow deficits in healthy eyes. PLoS One. 2018;13(11):e0207638. doi:10.1371/journal.pone.0207638 [CrossRef] PMID:30440050
- Spaide RF. Choriocapillaris signal voids in maternally inherited diabetes and deafness and in pseudoxanthoma elasticum. Retina. 2017;37(11):2008–2014. doi:10.1097/IAE.0000000000001497 [CrossRef] PMID:28092344
- Phansalkar N, More S, Sabale A, Joshi M. Paper ID 154: Adaptive local thresholding for detection of nuclei in diversely stained cytology images. Proc International Conf Commun Signal Process. 2011;218–220.
- Ponto KA, Elbaz H, Peto T, et al. Prevalence and risk factors of retinal vein occlusion: the Gutenberg Health Study. J Thromb Haemost. 2015;13(7):1254–1263. doi:10.1111/jth.12982 [CrossRef] PMID:25894549
- Sansom LT, Suter CA, McKibbin M. The association between systolic blood pressure, ocular perfusion pressure and subfoveal choroidal thickness in normal individuals. Acta Ophthalmol. 2016;94(2):e157–e158. doi:10.1111/aos.12794 [CrossRef] PMID:26173461
- Oh J, Baik DJ, Ahn J. Inter-relationship between retinal and choroidal vasculatures using optical coherence tomography angiography in normal eyes. Eur J Ophthalmol. 2020;30(1):48–57. doi:10.1177/1120672118816225 [CrossRef] PMID:30516077
- Zouache MA, Eames I, Klettner CA, Luthert PJ. Form, shape and function: segmented blood flow in the choriocapillaris. Sci Rep. 2016;6(1):35754. doi:10.1038/srep35754 [CrossRef] PMID:27779198
- Pettenkofer M, Scherm P, Feucht N, Wehrmann K, Lohmann CP, Maier M. Choriocapillaris flow density negatively correlates with advancing age on spectral-domain optical coherence tomography angiography. Ophthalmic Surg Lasers Imaging Retina. 2019;50(5):302–308. doi:10.3928/23258160-20190503-07 [CrossRef] PMID:31100161
Demographic and Clinical Characteristics of Study Groups
|Age, y||53.9 ± 4.2||55.4 ± 6.2||.39|
|Hypertension, n (%)||0||23 (71.8%)||NA|
|Smoking Status||3 (18.8%)||7 (21.8%)||.9|
Optical Coherence Tomography Angiography Parameters in Study Groups
|FAZ Area, mm2||0.25 ± 0.1||0.27 ± 0.1||.67|
|Vessel Density SCP, %||49.0 ± 2.4||46.8 ± 3.6||.03|
|Vessel Density DCP, %||55.0 ± 1.9||53.2 ± 3.2||.04|
|Number of Resolved PAMM Lesions||0||5.1 ± 5.3||NA|
|Choriocapillaris Total Flow Area, mm2||6.22 ± 0.13||6.0 ± 0.34||.005|
|Flow Voids Number||13.4 ± 5.4||20.8 ± 5.5||.000|
|Flow Voids Area, %||6.3 ± 2.7||12.4 ± 7.1||.002|