Ophthalmic Surgery, Lasers and Imaging Retina

Clinical Science 

Comparison of Morphological Features of Type 1 CNV in AMD and Pachychoroid Neovasculopathy: An OCTA Study

Özlem Biçer, MD; Sibel Demirel, MD, FEBO; Zeynep Yavuz, MD; Figen Batioğlu, MD; Emin Özmert, MD

Abstract

BACKGROUND AND OBJECTİVE:

The purpose of this study is to compare the morphological features of type 1 choroidal neovascularization (CNV) in eyes with age-related macular degeneration (AMD) and pachychoroid neovasculopathy (PNV) using optical coherence tomography angiography (OCTA).

PATİENTS AND METHODS:

Nineteen eyes of 17 patients with PNV and 30 eyes of 30 patients with AMD were evaluated. The size and area of CNV and morphological patterns during a 6-month period were analyzed using optical coherence tomography angiography.

RESULTS:

The presence of a feeder vessel was more common in AMD than in PNV. Indistinct pattern was more common in PNV than AMD. Pruned vascular tree pattern was rare in PNV eyes during follow-up. The mean size and flow of selected CNV area was significantly smaller in PNV group.

CONCLUSİON:

This study demonstrated that type 1 CNVs in the PNV group is characterized by a smaller area. Morphologic pattern differences between them might be explained by different etiopathogenesis under these circumstances.

[Ophthalmic Surg Lasers Imaging Retina. 2020;51:640–647.]

Abstract

BACKGROUND AND OBJECTİVE:

The purpose of this study is to compare the morphological features of type 1 choroidal neovascularization (CNV) in eyes with age-related macular degeneration (AMD) and pachychoroid neovasculopathy (PNV) using optical coherence tomography angiography (OCTA).

PATİENTS AND METHODS:

Nineteen eyes of 17 patients with PNV and 30 eyes of 30 patients with AMD were evaluated. The size and area of CNV and morphological patterns during a 6-month period were analyzed using optical coherence tomography angiography.

RESULTS:

The presence of a feeder vessel was more common in AMD than in PNV. Indistinct pattern was more common in PNV than AMD. Pruned vascular tree pattern was rare in PNV eyes during follow-up. The mean size and flow of selected CNV area was significantly smaller in PNV group.

CONCLUSİON:

This study demonstrated that type 1 CNVs in the PNV group is characterized by a smaller area. Morphologic pattern differences between them might be explained by different etiopathogenesis under these circumstances.

[Ophthalmic Surg Lasers Imaging Retina. 2020;51:640–647.]

Introduction

Type 1 choroidal neovascularization (CNV) can occur not only in age-related macular degeneration (AMD) but also among pachychoroid diseases, including central serous chorioretinopathy, pachychoroid neovasculopathy (PNV), and polypoidal choroidal vasculopathy.1 Pang and Freund suggested that PNV should be considered in the differential diagnosis of eye features of type 1 CNV and choroidal thickening in the absence of characteristic AMD findings or degenerative changes.2 Miyake et al. reported that the genetic background, age, and choroidal thickness of PNV differ significantly from those of AMD.3

Traditional multimodal imaging, such as fluorescein angiography (FA), indocyanine green angiography (ICGA), and spectral-domain optical coherence tomography (SD-OCT), play an important role in the diagnosis of type 1 CNV.4,5 Type 1 CNV due to its location under the retinal pigment epithelium (RPE) makes it challenging to evaluate the neovascular complex using conventional angiography or SDOCT.6 OCT angiography (OCTA) is a new technique used to visualize the retinal and choroidal vascular structure and outer retina by detecting motion contrast from blood flow without the injection of any dye, unlike traditional angiographies.7 The evaluation of type 1 CNV using the conventional angiography method is highly difficult in practicality due to leakage on FA and choroidal hyperpermeability on ICGA. Owing to no staining and no leakage on OCTA, vascular structures are not masked by hyperfluorescence, and more detailed images are obtained.

Previous studies suggest that OCTA can be used to identify the vascular network of CNVs.7–11 To our knowledge, there have been no studies comparing the qualitative and quantitative parameters of type 1 CNV on OCTA between AMD and PNV in patients who are under treatment. The aim of this study was to use OCTA to describe the morphologic features of type 1 CNV in eyes with AMD and PNV and to compare the treatment responses of type 1 lesions in AMD and in PNV. We hypothesized that the novel findings of this comparison would inspire ophthalmologists to use OCTA for the differential diagnosis of type 1 CNV and to understand differences of these diseases.

Patients and Methods

This was a retrospective cohort study of eyes with AMD and PNV previously treated at Ankara University Faculty of Medicine, Department of Ophthalmology, Retina Unit, between October 2017 and October 2018. The study received institutional review board approval for retrospective OCTA study and was performed in accordance with the Declaration of Helsinki and the Health Insurance Portability and Accountability Act (19-1295-18/26.11.2018). We retrospectively reviewed 19 eyes of 17 patients with PNV and 30 eyes of 30 patients with type 1 CNV in AMD.

Only eyes that were active according to OCT criteria with a baseline presence of intraretinal fluid or subretinal fluid, intraretinal cyst, and with pigment epithelial detachment (PED) during their baseline visit were included. All eyes had at least 6 months' follow-up with good quality OCTA images at all visits. The diagnosis of PNV was based on the presence of type 1 CNV without any features of AMD or other degenerative changes. We enrolled cases that exhibited localized areas of choroidal thickening directly below the neovascular tissue and/or pachyvessels with attenuation of overlying choriocapillaris. Patients who had a history of active neovascularization treated with anti-VEGF for at least the last 6 months and exudative signs of activity at the initial evaluation were included. Exclusion criteria included other forms of neovascular AMD (type 2 or 3 neovascularization) or secondary CNV other than AMD and PNV; end-stage neovascular AMD with macular scarring; evidence of diabetic retinopathy or any other macular or retinal vascular disease; signs or history of hereditary retinal dystrophy or myopic CNV and history of treatments other than intravit-real anti-VEGF drug administration, such as photo-dynamic therapy for PNV. Poor scan quality with a signal strength index less than 40, the absence of CNV on initial OCTA, and the inability to obtain serial imaging were additional exclusion criteria.

Using the AngioAnalytics software (Optovue, Fremont, CA), the choriocapillaris slab was selected and the segmentation level was manually adjusted to show the most detailed image of the type 1 lesion. The greatest linear diameter (GLD) of CNV, selected CNV area, and CNV flow area values were automatically measured as only detected flow signals within the selected area of the CNV. We focused on the comparison of these parameters on OCTA between the type 1 CNV in AMD and PNV at each of three visits (baseline, and at months 3 and 6).

On the en face images, the morphological patterns of the neovascular membranes were categorized using previously described methods.10,11 A comparison of the morphologic patterns and qualitative features between the groups was also performed at baseline, as well as at Months 3 and 6.

Biomarkers of CNV activity on OCTA were described by Coscas et al.10 A lesion was assessed as active if it showed at least three of the following five features: 1) shape (lacy-wheel or sea-fan); 2) branching (tiny numerous capillaries); 3) the presence of anastomoses and loops; 4) the presence of a peripheral arcade; and 5) the presence of a perilesional hypointense halo.10 In our study, we had accepted the medusa, sea-fan, and indistinct network patterns instead of the lacy-wheel pattern to define a membrane as active. A concordance regarding activity of the lesions between OCTA and the B-scans of the OCT was analyzed and compared between groups.

Descriptive statistics of the type 1 CNV in AMD and PNV patient groups were presented as; frequency (percentage) for quantitative variables, mean ± standard deviation and median (minimum-maximum) for qualitative variables respectively. For statistical analysis of changes in the quantitative parameters on OCT and OCTA after treatment, we used the Friedman test and the Wilcoxon signed-rank test with Bonferroni correction for pairwise comparisons between these time points (baseline, 3 months, 6 months). Qualitative variables between the type 1 AMD and PNV patients were compared with Chi-Square or Fisher's exact test, whereas Mann-Whitney U Test was used to compare quantitative variables. Agreements between OCT and OCTA at baseline, 3 months, and 6 months were assessed with Kappa (κ) and prevalence-adjusted bias-adjusted Kappa (PABAK) coefficients. The imbalance between the frequencies of positive and negative values causes κ coefficient to be underestimated; therefore, κ and PABAK proposed by Byrt et al.,12 which adjusts for the imbalance, were both presented. Strength of the agreement was assessed using the range values by Landis and Koch.13 Statistical analyses were performed using Statistical Package for Social Sciences (version 15.0; SPSS Inc., Chicago, IL.) and “epiR” package14 for R (RStudio, Version 1.1.463; RStudio, Inc., Boston, MA). P values less than .05 were considered statistically significant.

Results

The demographic and clinical characteristics of the patients are shown in Table 1. Age was significantly higher in the AMD group than in the PNV group (P < .001). The quantitative measurement of OCT and OCTA characteristics, such as the mean central macular thickness, central choroidal thickness, maximum PED height, GLD of CNV, selected CNV area, and CNV flow area are summarized in Table 2. The feeder vessel was more common in the AMD group than the PNV group at all visits; however, peripheral arcade and anastomoses were more frequent in the PNV group than the AMD group (Table 3).

Demographics and Clinical Data From the Study Population

Table 1:

Demographics and Clinical Data From the Study Population

Changes in Quantitative Parameters on OCT and OCTA During Six-Month Treatment for Type 1 Neovascularization in AMD and PNVChanges in Quantitative Parameters on OCT and OCTA During Six-Month Treatment for Type 1 Neovascularization in AMD and PNV

Table 2:

Changes in Quantitative Parameters on OCT and OCTA During Six-Month Treatment for Type 1 Neovascularization in AMD and PNV

Baseline Qualitative Characteristics of the İncluded Patients

Table 3:

Baseline Qualitative Characteristics of the İncluded Patients

Lesion patterns of each of the groups with type 1 CNV are summarized in Table 4. The indistinct pattern was more common in the PNV group than the AMD group at all visits. At 3 months, a pruned vascular tree pattern was seen only in eyes with AMD. At 6 months, a pruned vascular tree pattern was detected in 23.3% of AMD eyes and 5.3% of PNV eyes (Table 4).

Morphologic Pattern on OCTA of the İncluded Patients

Table 4:

Morphologic Pattern on OCTA of the İncluded Patients

We found no relationship between OCT and OCTA regarding lesion activity in the AMD group at baseline, 3 months, and 6 months. Agreement with OCT and OCTA was moderate at baseline and weak at 3 and 6 months in PNV group (Table 5).

Agreement of OCT and OCTA

Table 5:

Agreement of OCT and OCTA

Discussion

In the present study, we compared the morphology characteristics of type 1 CNV to distinguish PNV from AMD and found that type 1 CNV in PNV has a smaller linear dimension and area of CNV in addition to smaller flow area than type 1 CNV in AMD. No significant change was found in the size and flow area during follow-up in both groups. Qualitative analysis of OCTA images revealed an indistinct pattern was more common in the PNV group than the AMD group, as well as the rare appearance of a pruned vascular tree in the PNV group during follow-up. These findings highlight the different responses to the vascular net of type 1 CNV.

Previous studies have shown different clinical findings between PNV and AMD.2,3 PNV can be distinguished from AMD by a relative absence of drusen, younger age at onset of CNV, and thick choroid with pachyvessels.2 Miyake et al. reported that the genetic susceptibility (ARMS2 rs10490924 and CFH rs800292) of PNV to AMD was significantly lower than that of AMD.3 Thus, genetic backgrounds might be different between PNV and AMD. However, a few studies mention differences between PNV and AMD regarding imaging properties with dye angiography, genetic predisposition, and treatment responses. One reported by Terao et al. demonstrated that neither GLD on FA nor CNV lesion size on ICGA differed between PNV and AMD.15 However, we showed that the difference in quantitative features on OCTA between the PNV group and AMD group was obvious. It might be explained that OCTA could show the exact size of the neovascular membrane, and unlike conventional dye angiographies, it is not affected by leakage, staining, and/or choroidal hyperpermeability. Therefore, studies performed with dye angiography and comparing the lesion size may contain flaws regarding the measurement of lesion size. In addition, angiogenic factors and proinflammatory cytokines may play an important role in the difference in lesion size between the groups. Previously, Hata et al. evaluated nine eyes with treatment-naïve PNV and 21 eyes with treatment-naïve AMD and compared initial concentration of VEGF in the aqueous humor using enzyme-linked immunosorbent assay.16 They concluded that the mean VEGF concentration in PNV was lower than that in exudative AMD. Terao et al. reported that in the AMD group, VEGF-A, bFGF, GM-CSF, and MCP-1 were significantly higher than in the PNV group.15 Therefore, a prominent increase in inflammation and angiogenesis in AMD may be associated with lesion growth.

Several studies have reported changes in CNV morphology and changes in CNV size in response to anti-VEGF treatment of different subtypes of AMD.7–11 Coscas et al. reported that quiescent lesions showed typical long, filamentous linear vessels branching into other large mature vessels, with rare or absent anastomoses and characterized by a dead tree appearance at the vessel's termini.10 These similar findings were described for type 1 CNV in the PNV group as quiescent lesions in a small cohort of five PNV patients by Azar et al.8 In our series, the indistinct pattern and the peripheral arcade and/or anastomose were more common in the PNV group than the AMD group. However, a prominent feeder vessel that helped to name the CNV appearance (such as medusa, sea-fan, etc.) was more common in the AMD group than the PNV group. A feeder vessel was visible in 21 eyes (70%) with AMD lesions in our study but was seen in only four eyes (21.1%) with PNV. This difference might be related to the relatively large number of cases in the series on the contrary to small series reported by Azar et al.

The reason for the difference in the response to anti-VEGF treatment regarding membrane morphology between PNV and AMD is unclear.17 Lesion area and flow area showed no significant changes for both type 1 CNV in the study. However, the pruned vascular tree pattern was more common in the AMD group than the PNV group during follow-up. This suggests that the morphologic pattern reveals the real pathophysiological features of type 1 CNV. This result may be associated with the pruning process by repeated anti-VEGF treatment for CNV, characterized by the appearance of large diameter vessels, loss of capillaries, and prominent anastomoses of vessels.18 On the other hand, the adequate pericyte coverage of the vessel renders it resistant to anti-VEGF therapy alone, as the pericytes supply VEGF locally to their associated vascular endothelial cells. Pruning may be rarely seen in eyes with PNV because of the difference in the level of cytokine seen in the PNV group compared to the AMD group, which may play a role in the proliferation, maturation, and migration of pericytes. Terao et al. recently demonstrated a significant association between the levels of VEGF-A and PIGF and the resolution of exudative change in the AMD group, whereas there was no significant association between cytokine and treatment response in the PNV group.15 This finding seems to confirm our results, suggesting a different pruning process caused by the cytokine profile of the two conditions. Even if there is no difference in anti-VEGF response between the groups regarding treatment efficacy, the morphologic pattern during follow-up between groups seems different, as we rarely observed the pruned tree in the PNV group.

A reduction in the visibility of the tiny branching vessels, peripheral arcades, anastomoses, and loops was demonstrated by qualitative analysis of the OCTA images in the literature.18,19 These data do not seem concordant with the PNV cases in our series. We demonstrated no pronounced reduction in these morphological findings in PNV, at least in the short term. This indicates that this potential biomarker may not differentiate active lesions during follow-up visits. As the agreement between OCTA and OCT was low, we suggest that combining OCTA and structural OCT data may be a useful technique for the noninvasive, clinical diagnosis and follow-up treatment of type 1 CNV.

The study has several limitations due to its retrospective nature, single-center design, small sample size, heterogenous nature (treatment naïve and treated eyes), and application of different treatment protocol to patients. In addition, the need for a high imaging quality for analysis may have resulted in a selection bias. The limitations of OCTA, including projection, motion, and image processing artifacts resulting in a high number of non-evaluable images, remain considerable. However, despite the drawbacks of the studies, the differences we have observed are striking enough to have an idea about differences between these diseases regarding imaging properties.

In conclusion, our results demonstrated that type 1 CNVs in PNV were characterized by a smaller CNV area and flow compared with type 1 CNV in AMD. Their etiologies are likely to differ because of the rare appearance of a pruned vascular tree and mostly different morphologic pattern in the PNV group. OCTA might be a useful technology for the noninvasive monitoring of type 1 CNV subtypes and differential diagnosis of two conditions with further prospective studies on larger cohorts and long-term follow-up to validate our findings.

References

  1. Gallego-Pinazo R, Dolz-Marco R, Gómez-Ulla F, Mrejen S, Freund KB. Pachychoroid diseases of the macula. Med Hypothesis Discov Innov Ophthalmol. 2014;3(4):111–115. PMID:25756060
  2. Pang CE, Freund KB. Pachychoroid neovasculopathy. Retina. 2015;35(1):1–9. doi:10.1097/IAE.0000000000000331 [CrossRef] PMID:25158945
  3. Miyake M, Ooto S, Yamashiro K, et al. Pachychoroid neovasculopathy and age-related macular degeneration. Sci Rep. 2015;5(1):16204 doi:10.1038/srep16204 [CrossRef] PMID:26542071
  4. Sulzbacher F, Kiss C, Munk M, Deak G, Sacu S, Schmidt-Erfurth U. Diagnostic evaluation of type 2 (classic) choroidal neovascularization: optical coherence tomography, indocyanine green angiography, and fluorescein angiography. Am J Ophthalmol. 2011;152(5):799–806.e1. doi:10.1016/j.ajo.2011.04.011 [CrossRef] PMID:21742302
  5. Wilde C, Patel M, Lakshmanan A, Amankwah R, Dhar-Munshi S, Amoaku WMedscape. The diagnostic accuracy of spectral-domain optical coherence tomography for neovascular age-related macular degeneration: a comparison with fundus fluorescein angiography. Eye (Lond). 2015;29(5):602–609. doi:10.1038/eye.2015.44 [CrossRef] PMID:25907206
  6. Kuehlewein L, Bansal M, Lenis TL, et al. Optical Coherence Tomography Angiography of Type 1 Neovascularization in Age-Related Macular Degeneration. Am J Ophthalmol. 2015;160(4):739–48.e2. doi:10.1016/j.ajo.2015.06.030 [CrossRef] PMID:26164826
  7. Jia Y, Bailey ST, Wilson DJ, et al. Quantitative optical coherence tomography angiography of choroidal neovascularization in age-related macular degeneration. Ophthalmology. 2014;121(7):1435–1444. doi:10.1016/j.ophtha.2014.01.034 [CrossRef] PMID:24679442
  8. Azar G, Wolff B, Mauget-Faÿsse M, Rispoli M, Savastano MC, Lumbroso B. Pachychoroid neovasculopathy: aspect on optical coherence tomography angiography. Acta Ophthalmol. 2017;95(4):421–427. doi:10.1111/aos.13221 [CrossRef] PMID:27597633
  9. Dansingani KK, Balaratnasingam C, Naysan J, Freund KB. En Face Imaging of Pachychoroid Spectrum Disorders with Swept-Source Optical Coherence Tomography. Retina. 2016;36(3):499–516. doi:10.1097/IAE.0000000000000742 [CrossRef] PMID:26335436
  10. Coscas GJ, Lupidi M, Coscas F, Cagini C, Souied EH. Optıcal coherence tomography angiography versus traditional multımodal imaging in assessing the activity of exudative age-related macular degeneratıon: A new diagnostic challenge. Retina. 2015;35(11):2219–2228. doi:10.1097/IAE.0000000000000766 [CrossRef] PMID:26398697
  11. Miere A, Butori P, Cohen SY, et al. Vascular remodeling of choroidal neovascularization after anti-vascular endothelial growth factor therapy visualized on optical coherence tomography angiography. Retina. 2019;39(3):548–557. doi:10.1097/IAE.0000000000001964 [CrossRef] PMID:29210939
  12. Byrt T, Bishop J, Carlin JB. Bias, prevalence and kappa. J Clin Epidemiol. 1993;46(5):423–429. doi:10.1016/0895-4356(93)90018-V [CrossRef] PMID:8501467
  13. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33(1):159–174. doi:10.2307/2529310 [CrossRef] PMID:843571
  14. Heuer Cord, Mark Stevenson with contributions from Telmo Nunes et al. (2020). epiR: Tools for Analysis of Epidemiological Data. R package version 1.0-14. https://CRAN.R-project.org/package=epiR.
  15. Terao N, Koizumi H, Kojima K, et al. Distinct Aqueous Humour Cytokine Profiles of Patients with Pachychoroid Neovasculopathy and Neovascular Age-related Macular Degeneration. Sci Rep. 2018;8(1):10520 doi:10.1038/s41598-018-28484-w [CrossRef] PMID:30002400
  16. Hata M, Yamashiro K, Ooto S, et al. Intraocular Vascular Endothelial Growth Factor Levels in Pachychoroid Neovasculopathy and Neovascular Age-Related Macular Degeneration. Invest Ophthalmol Vis Sci. 2017;58(1):292–298. doi:10.1167/iovs.16-20967 [CrossRef] PMID:28114590
  17. Cho HJ, Jung SH, Cho S, Han JO, Park S, Kim JW. Intravitreal Anti-Vascular Endothelial Growth Factor Treatment for Pachychoroid Neovasculopathy. J Ocul Pharmacol Ther. 2019;35(3):174–181. doi:10.1089/jop.2018.0107 [CrossRef] PMID:30807236
  18. Spaide RF. Optical Coherence Tomography Angiography Signs of Vascular Abnormalization With Antiangiogenic Therapy for Choroidal Neovascularization. Am J Ophthalmol. 2015;160(1):6–16. doi:10.1016/j.ajo.2015.04.012 [CrossRef] PMID:25887628
  19. de Carlo TE, Bonini Filho MA, Chin AT, et al. Spectral-domain optical coherence tomography angiography of choroidal neovascularization. Ophthalmology. 2015;122(6):1228–1238. doi:10.1016/j.ophtha.2015.01.029 [CrossRef] PMID:25795476

Demographics and Clinical Data From the Study Population

AMD (n = 30)PNV (n = 19)P Value

Mean ± SD*, Median (Min-Max)Mean ± SD*, Median (Min-Max)

Age, Years71.53 ± 9.14, 72.5 (52–89)56.36 ± 7.54, 58.0 (40–68)< .001

Gender, n (%).489
  Male  21 (70.0%)  15 (78.9%)
  Female  9 (30.0%)  4 (21.1%)

Anti-VEGF Injections
  Prior to Baseline8.03 ± 3.21, 8.0 (3–15)6.47 ± 2.97, 6.0 (3–16).052
  Baseline to 3 Months1.7 ± 0.75, 2.0 (1–3)1.53 ± 0.84, 1.0 (1–3).304
  3 Months to 6 Months1.03 ± 0.96, 1.0 (0–3)1.0 ± 0.88, 1.0 (0–3).948

Changes in Quantitative Parameters on OCT and OCTA During Six-Month Treatment for Type 1 Neovascularization in AMD and PNV

Baseline3 Months6 Months

Mean ± SDMean ± SDMean ± SDP Value**

Median (Min-Max)Median (Min-Max)Median (Min-Max)

PED Max Height, μm
AMD161.03 ± 76.21a145.86 ± 66.36b148.2 ± 72.91b.012
146.0 (81.0–385.0)122.5 (66.0–337.0)118.0 (62–353.0)

PNV81.47 ± 48.4573.21 ± 47.5270.47 ± 50.96.132
60.0 (25.0–205.0)58.0 (0–179.0)57.0 (0–201.0)

P Value*< .001< .001< .001

CMT, μm

AMD316.93 ± 82.55a289.66 ± 68.51b300.4 ± 77.57a,b.019
304.5 (211.0–523.0)277.5 (191.0–484.0)265.5 (191.0–512.0)

PNV328.57 ± 97.49a255.21 ± 72.35b240.36 ± 58.48b< .001
336.0 (171.0–528.0)249.0 (167.0–452.0)229.0 (177.0–393.0)

P Value*.454.081.005

CCT, μm

AMD191.6 ± 53.09a182.33 ± 45.35a,b175.53 ± 45.15b< .001
188.0 (85.0–323.0)183.5 (90.0–299.0)169.0 (83.0–287.0)

PNV373.52 ± 112.63a320.63 ± 105.77b299.42 ± 87.59b< .001
347.0 (195.0–535.0)298.0 (168.0–495.0)284.0 (152.0–446.0)

P Value*< .001< .001< .001

GLD of CNV, μm

AMD1364.0 ± 571.481352.3 ± 576.531325.3 ± 503.02.471
1275.0 (310.0–2460.0)1310.0 (0–2370.0)1330.0 (270.0–2360.0)

PNV856.8 ± 488.26893.7 ± 463.49837.9 ± 456.08.287
680.0 (220.0–1910.0)760.0 (240.0–1780.0)710.0 (220.0–1710.0)

P Value*.002.003.002

Selected Area, mm2

AMD6.0 ± 4.875.54 ± 4.455.3 ± 4.09.393
4.82 (0.24–18.19)4.48 (0–17.54)4.51 (0.21–14.89)

PNV2.02 ± 2.182.11 ± 2.052.0 ± 2.11.504
1.30 (0.06–8.11)1.4 (0.09–7.43)1.08 (0.12–8.16)

P Value*< .001.003.001

Flow Area, mm2

AMD3.44 ± 2.743.15 ± 2.433.1 ± 2.28.195
2.93 (0.16–10.83)2.52 (0–9.95)2.6 (0.17–8.64)

PNV1.27 ± 1.321.34 ± 1.31.24 ± 1.24.431
0.76 (0.05–4.81)0.99 (0.07–4.66)0.65 (0.09–4.52)

P Value*.001.003.001

Baseline Qualitative Characteristics of the İncluded Patients

Morphologic PatternAMD (n = 30)PNV (n = 19)P Value

n (%)n (%)

Baseline
   Subretinal fluid19 (63.3%)18 (94.7%).017
   Intraretinal cyst21 (70.0%)3 (15.8%)< .001
   Feeder vessel21 (70.0%)4 (21.1%).001
   Branching numerous tiny capillaries19 (63.3%)15 (78.9%).248
   Anastomoses18 (60.0%)15 (78.9%).168
   Peripheral arcade8 (26.7%)11 (57.9%).029
   Loops2 (6.7%)1.0001.000
   Hypointense halo8 (26.7%)5 (26.3%).978

3 Months
   Subretinal fluid13 (43.3%)13 (68.4%).086
   Intraretinal cyst10 (33.3%)1 (5.3%).033
   Feeder vessel21 (70%)4 (21.1%).001
   Branching numerous tiny capillaries19 (63.3%)15 (78.9%).248
   Anastomoses12 (40.0%)15 (78.9%).008
   Peripheral arcade6 (20.0%)12 (63.2%).002
   Loops2 (6.7%)4 (21.1%).190
   Hypointense halo6 (20.0%)7 (36.8%).193

6 Months
   Subretinal fluid8 (26.7%)10 (52.6%).066
   Intraretinal cyst11 (36.7%)1 (5.3%).017
   Feeder vessel21 (70.0%)4 (21.1%).001
   Branching numerous tiny capillaries18 (60.0%)14 (73.7%).327
   Anastomoses13 (43.3%)14 (73.7%).037
   Peripheral arcade6 (20.0%)8 (42.1%).095
   Loops3 (10.0%)3(15.8%).665
   Hypointense halo10 (33.3%)5 (26.3%).604

Morphologic Pattern on OCTA of the İncluded Patients

Morphologic PatternAMD (n = 30)PNV (n = 19)P Value

n (%)n (%)

Baseline
   Seafan3 (10.0%)0 (0.0%).004
   Medusa13 (43.3%)4 (21.1%)
   Indisinct9 (30.0%)a15 (78.9%)b
   Pruned vascular tree5 (16.7%)0 (0.0%)

3 Months
   Seafan3 (10.3%)0 (0.0%).001
   Medusa10 (34.5%)4 (21.1%)
   Indisinct8 (27.6%)a15 (78.9%)b
   Pruned vascular tree8 (27.6%)a0 (0.0%)b

6 Months
   Seafan2 (6.7%)0 (0.0%).039
   Medusa11 (36.7%)4 (21.1%)
   Indisinct10 (33.3%)a14 (73.7%)b
   Pruned vascular tree7 (23.3%)1 (5.3%)

Agreement of OCT and OCTA

OCT-OCTAAMDPNV

Kappa (κ) (95% CI)PABAK (95% CI)Kappa (κ) (95% CI)PABAK (95% CI)

Baseline0.133 (−0.251; 0.490)0.578 (0.088; 0.878)

3 Months0 (−0.351; 0.351) 0 (−0.374; 0.374)0.015 (−0.393; 0.422) 0.263 (−0.232; 0.674)

6 Months−0.104 (−0.442; 0.234) −0.133 (−0.491; 0.251)−0.226 (−0.581; 0.129) −0.263 (−0.674; 0.232)
Authors

From Boğazlıyan State Hospital, Department of Ophthalmology, Yozgat, Turkey (OB); Ankara University Faculty of Medicine Department of Ophthalmology, Ankara, Turkey (SD, FB, EO); and Ankara University Faculty of Medicine, Department of Biostatistics, Ankara, Turkey (ZY).

The authors report no relevant financial disclosures.

Address correspondence to Sibel Demirel, MD, Department of Ophthalmology, Vehbi Koç Eye Hospital, Mamak Street, Dikimevi, Ankara, Turkey; email: drsibeldemireltr@yahoo.com.tr.

Received: April 16, 2020
Accepted: September 10, 2020

10.3928/23258160-20201104-06

Sign up to receive

Journal E-contents