BackgroundTo map and compare the three-dimensional choroidal vascularity index (3D-CVI) in eyes with unilateral central serous chorioretinopathy (CSC), fellow eyes and control eyes using ultra-widefield swept source optical coherence tomography (UWF SS-OCTA).MethodsIn this prospective observational study, the 3D-CVIs were measured in 9 subfields or 1 × 1 mm grids by the UWF SS-OCTA with a viewing angle of horizontal 24 × vertical 20 mm. The proportions of vortex vein anastomoses and their corresponding CVI in the central regions were compared among the CSC, fellow and control eyes. Correlations of CVI and vascular density of the large choroidal vessel layer/choriocapillaris layer/choroidal thickness (CT) were also assessed.ResultsThirty-two eyes in 32 patients with unilateral CSC and 32 normal eyes were included in the study. The mean CVI in the eyes with CSC was significantly greater than that in the fellow eyes of CSC and control eyes (41.99 ± 3.56% vs. 40.38 ± 3.855%, P = 0.003; 41.99 ± 3.56% vs. 38.93 ± 4.067%, P = 0.004, respectively). The CVIs in superotemporal, inferotemporal and inferonasal regions were significantly higher in CSC eyes than control eyes (P = 0.03, P = 0.02, P = 0.008). In CSC eyes, there was a linear positive correlation between 3D-CVI and vascular density of the large choroidal vessel layer and CT in all subfields. The proportion of vortex vein anastomoses in CSC was 25/32 (78.1%), and significantly higher in fellow and control eyes (P < 0.001). The average central CVI was significantly higher in CSC eyes with anastomoses than in CSC eyes without anastomoses (42.8 ± 5.1% vs. 38.4 ± 2.7%, P = 0.039). CVIs in superior, central, inferior, superonasal, nasal and inferonasal regions were significantly correlated with vortex vein anastomoses (P < 0.05), regardless of CSC, fellow or healthy eyes. In addition, whether there were vortex vein anastomoses, CVI in superotemporal region was significantly higher in eyes with CSC (P = 0.002) and fellow eyes (P = 0.014), compared to control eyes. No significant correlation was found between hypertension and CVIs in the three groups.ConclusionRemodeling of choroidal drainage routes by venous anastomosis between superior and inferior vortex veins may be common in CSC. The 3D-CVI could be a comprehensive parameter to evaluate the choroid vasculature and help understand the pathogenesis of pachychoroid spectrum disease.

We aimed to quantitatively compare the choroid blood flow and choroid thickness at the periphery among eyes with central serous chorioretinopathy (CSC), fellow eyes and healthy eyes using ultra-widefield swept-source optical coherence tomography angiography (UWF SS-OCTA). Retrospective analysis of 49 patients with CSC (98 eyes, including unaffected fellow eyes) and 49 age and sex matched controls were included. We obtained 3-dimensional data of vertical 20 mm × horizontal 24 mm × scan depth 6 mm, comprising 9 subfields (superotemporal, upper, superonasal, temporal, central, nasal, inferotemporal, lower, inferonasal regions). CSC eyes presented with greater density of large-vessel choroidal layer in all the 9 subfields compared with controls. Compared with normal eyes, CSC eyes had greater choroidal thickness (superotemporal, upper, superonasal, temporal, central, nasal, inferotemporal, and inferonasal subfields) and choroidal volume (superotemporal, upper, superonasal, temporal, central, and nasal subfields). Compared with control eyes, the choriocapillaris density in the superotemporal, inferotemporal and inferonasal subfields was greater in patients with CSC. Our study may provide further evidence for the congestion of vortex vein in the pathogenesis of CSC. UWF SS-OCTA can be used to evaluate the abnormalities of the choroidal structures even at the periphery in eyes with CSC.