超广角扫频OCTA评估糖尿病视网膜病变血管生物标志物:DRIVE研究
该横断面研究基于单次采集29×24 mm超广角扫频OCTA图像,量化165眼非灌注指数与视网膜内微血管异常计数,并按中央区、中间区、外围区分析。结果显示两项指标随DR严重程度增加,外围区NPI判别PDR与NPDR效能与总NPI相当,中间区指标判别晚期与早期DR效能与全域测量相当。
要点
- 总NPI与IRMA计数随DR严重程度递增(P < 0.001)。
- NPI从中央区向中间区递增,PDR进一步延伸至外围区。
- 外围区NPI判别PDR与NPDR效能与总NPI相当。
核心结论
- 单次采集超广角扫频OCTA支持糖尿病视网膜病变的区域性生物标志物框架。
- 更外围的缺血可能有助于识别新生血管转化。
- 当无法获得完整超广角成像时,延伸至中间视网膜的成像对区分晚期与早期DR仍具诊断价值。
研究设计与样本
- 横断面研究,纳入120例患者共165眼(73例男性;平均年龄57.0 ± 11.1岁),覆盖不同DR严重程度。
- 在单次采集29 × 24 mm OCTA图像上分析非灌注指数与IRMA计数,并分中央区、中间区、外围区进行区域分析。
- 采用受试者工作特征曲线下面积评估区域及总指标判别DR严重程度的能力。
主要发现
- 总NPI和IRMA计数随DR严重程度加重而增加(P < 0.001)。
- NPI在所有DR分期中从中央区到中间区显著增加(P < 0.001),增殖性DR进一步延伸至外围区(P < 0.001)。
- IRMA计数在重度非增殖性DR、重度NPDR和PDR中于中间区最高(P < 0.001)。
- 外围区NPI区分PDR与NPDR的AUC为0.88,总NPI为0.86(P = 0.21)。
- 中间区NPI区分晚期与早期DR的AUC为0.95,总NPI为0.95(P = 0.83);IRMA为0.86 vs. 0.89(P = 0.12)。
关键数值与推荐
- 外围区NPI判别PDR与NPDR的AUC为0.88,与总NPI的0.86相当(P = 0.21)。
- 中间区NPI判别晚期与早期DR的AUC为0.95,与总NPI的0.95相当(P = 0.83)。
- 中间区IRMA计数判别晚期与早期DR的AUC为0.86,与总IRMA的0.89相当(P = 0.12)。
临床怎么用
- 可利用单次超广角扫频OCTA的区域NPI和IRMA计数评估DR严重程度。
- 关注外围区NPI有助于识别可能向增殖性转化的情况。
- 若无法完成完整超广角成像,中间区NPI和IRMA计数仍可用于区分晚期与早期DR。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
PURPOSE: To quantify nonperfusion areas (NPA) and intraretinal microvascular abnormalities (IRMAs) and evaluate their diagnostic performance in diabetic retinopathy (DR) using ultrawide-field (UWF) swept-source optical coherence tomography angiography (SS-OCTA). METHODS: In this cross-sectional study, nonperfusion index (NPI) and IRMA counts were analyzed on single-capture 29 × 24 mm optical coherence tomography angiography (OCTA) images. Regional analyses across the central zone (CZ), intermediate zone (IZ), and outer zone (OZ) were performed. Area under the receiver operating characteristic analysis evaluated the ability of regional and total metrics to discriminate DR severities. RESULTS: A total of 165 eyes from 120 patients (73 male; mean age, 57.0 ± 11.1 years) across DR severities were included. Total NPI and IRMA counts increased with advancing DR severity (P < 0.001). NPI increased significantly from CZ to IZ across all DR stages (P < 0.001), with further extension into OZ in proliferative DR (PDR) (P < 0.001). IRMA counts were highest in the IZ in moderate nonproliferative DR (NPDR), severe NPDR, and PDR (P < 0.001). OZ NPI performed comparably with total NPI for distinguishing PDR from NPDR (area under the receiver operating characteristic, 0.88 vs. 0.86; P = 0.21), whereas IZ NPI and IRMA counts performed comparably with their respective total field measurements for distinguishing advanced DR from early DR (NPI, 0.95 vs. 0.95 [P = 0.83]; IRMA, 0.86 vs. 0.89 [P = 0.12]). CONCLUSIONS: Single-capture UWF SS-OCTA supports a regional biomarker framework for DR. More peripheral ischemia may add value for identifying neovascular transformation, whereas imaging extending into the intermediate retina may retain useful diagnostic value for distinguishing advanced from early DR when full UWF imaging is unavailable.
引用
- Seah WH, Song W, Zhuang L, et al. Evaluation of Vascular Biomarkers in Diabetic Retinopathy Using Ultrawide-Field Swept-Source Optical Coherence Tomography Angiography: The DRIVE Study. Invest Ophthalmol Vis Sci. 2026. PMID: 42187192. doi:10.1167/iovs.67.5.65
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。