Department of Ophthalmology, San Raffaele Scientific Institute, University Vita-Salute San Raffaele, Milan, Italy.
Eye (Lond). 2022 May;36(5):1027-1033. doi: 10.1038/s41433-021-01555-5. Epub 2021 May 10.
To inspect the inter-reader agreement of different diagnostic modalities in identifying choroidal neovascularization (CNV) activity secondary to angioid streaks (AS) and to analyze the prevalence of subretinal hyper-reflective material (SHRM) in active CNV.
Retrospective study of patients with AS with active CNV; optical coherence tomography (OCT), OCT angiography (OCTA), fundus fluorescein angiography (FFA), and indocyanine green angiography (ICGA) from each patient were collected. Agreement between two readers using different diagnostic modalities is presented as free-marginal kappa (k) and 95% confidence interval (CI).
This study included 19 eyes of 12 patients with active CNV (5 naive and 14 previously treated). Agreement among readers on CNV activity was excellent for OCT (k =0.88; 95% CI 0.71-1.00), good for FFA (k = 0.70; 95% CI 0.46-0.94) and ICGA (k = 0.58; 95% CI 0.31-0.84), and poor using OCTA (k = 0.39; 95% CI 0.11-0.68). SHRM was the most common OCT finding associated with active CNV (100%); fuzzy borders were present in 53% of SHRM cases at baseline.
Identification of CNV activity in AS is challenging; OCT was the best modality to inspect active CNV. The identification of SHRM contributed to recognizing active CNV. Further studies are needed to assess the role of SHRM in anticipating prognosis and guiding treatment of CNV secondary to AS.
检查不同诊断方式在识别继发于脉络膜骨瘤(angioid streaks,AS)的脉络膜新生血管(choroidal neovascularization,CNV)活性方面的读者间一致性,并分析活性 CNV 中视网膜下高反射物质(subretinal hyper-reflective material,SHRM)的患病率。
回顾性研究伴有活动 CNV 的 AS 患者;收集每位患者的光学相干断层扫描(optical coherence tomography,OCT)、OCT 血管造影(OCT angiography,OCTA)、眼底荧光素血管造影(fundus fluorescein angiography,FFA)和吲哚青绿血管造影(indocyanine green angiography,ICGA)。使用不同诊断方式的两位读者之间的一致性以边缘自由kappa(k)和 95%置信区间(95% confidence interval,CI)表示。
本研究纳入了 19 只眼的 12 例患者(5 例为初次,14 例为既往治疗)。在评估 CNV 活性方面,两位读者之间的一致性在 OCT 中极好(k=0.88;95%CI 0.71-1.00),在 FFA 中良好(k=0.70;95%CI 0.46-0.94),在 ICGA 中良好(k=0.58;95%CI 0.31-0.84),在 OCTA 中较差(k=0.39;95%CI 0.11-0.68)。SHRM 是与活性 CNV 最常见的 OCT 发现(100%);在基线时,53%的 SHRM 病例存在模糊边界。
在 AS 中识别 CNV 活性具有挑战性;OCT 是检查活性 CNV 的最佳方式。SHRM 的识别有助于识别活性 CNV。需要进一步研究来评估 SHRM 在预测 AS 继发 CNV 的预后和指导治疗中的作用。