Department of Ophthalmology and Visual Science, Eye, Ear, Nose, and Throat Hospital, Shanghai Medical College, Fudan University, Shanghai, China.
Key Laboratory of Myopia, Ministry of Health (Fudan University), Shanghai, China.
Eye (Lond). 2019 Oct;33(10):1596-1605. doi: 10.1038/s41433-019-0451-x. Epub 2019 May 3.
To investigate the diagnostic values of intereye or intraeye asymmetry of retinal perfused vessel density and neural structure parameters for detection of glaucoma.
In total, 152 healthy subjects and 72 bilateral primary open-angle glaucoma (POAG) patients were enrolled. Total POAG group contains all glaucoma patients. Early to moderate POAG group contains patients whose binocular mean defect values were larger than -12 dB. The retinal perfused vessel densities were acquired using optic coherence tomography angiography. The neural structure parameters include RNFL, GCC thickness and its derivative indices like focal loss volume percentage (FLV%) and global loss volume percentage (GLV%). Intereye asymmetry equaled to the absolute difference of parameters between paired eyes. Intraeye asymmetry was defined as absolute difference between the inferior and superior hemisphere values from one random selected eye. The areas under the receiver operating characteristic curves (AUROCs) were calculated to evaluate diagnostic ability.
From pairwise comparison analysis of ROC curves, the intereye asymmetric parameters with the largest diagnostic accuracy were FLV and GLV% (AUROC = 0.944), which were significantly superior to the intereye asymmetry of perfused vessel density in peripapillary area and parafoveal area (P < 0.05). Particularly, the intereye asymmetry of FLV% (AUROC = 0.926) and GLV% (AUROC = 0.950) showed excellent diagnostic precision for detecting early to moderate glaucoma patients. However, the intraeye asymmetry of microvascular parameters and neural structure parameters showed fair diagnostic ability for identifying POAG patients.
The intereye asymmetry of neural structure parameters, particularly the FLV% and GLV%, outperformed the microvascular parameters for identifying POAG patients.
探讨视网膜血流密度和神经结构参数的双眼或单眼不对称性在青光眼诊断中的价值。
共纳入 152 名健康受试者和 72 名双侧原发性开角型青光眼(POAG)患者。总 POAG 组包含所有青光眼患者。早期至中度 POAG 组包含双眼平均缺损值大于-12dB 的患者。使用光相干断层扫描血管造影术获得视网膜血流密度。神经结构参数包括 RNFL、GCC 厚度及其衍生指数,如焦点丢失体积百分比(FLV%)和全局丢失体积百分比(GLV%)。双眼不对称性等于双眼参数之间的绝对差值。单眼不对称性定义为一只随机选择的眼的下半球和上半球值之间的绝对差值。计算受试者工作特征曲线(ROC)下的面积(AUROC)以评估诊断能力。
从 ROC 曲线的成对比较分析来看,具有最大诊断准确性的双眼不对称参数是 FLV 和 GLV%(AUROC=0.944),明显优于视盘周围和中心凹旁血流密度的双眼不对称性(P<0.05)。特别是,FLV%(AUROC=0.926)和 GLV%(AUROC=0.950)的双眼不对称性对早期至中度青光眼患者的诊断具有出色的精度。然而,微血管参数和神经结构参数的单眼不对称性显示出识别 POAG 患者的良好诊断能力。
神经结构参数的双眼不对称性,特别是 FLV%和 GLV%,在识别 POAG 患者方面优于微血管参数。