VST Glaucoma Center, LV Prasad Eye Institute, Banjara Hills, Hyderabad, India.
Eye (Lond). 2012 Jan;26(1):133-9. doi: 10.1038/eye.2011.277. Epub 2011 Nov 11.
To compare the diagnostic abilities of peripapillary retinal nerve fiber layer (RNFL) and macular inner retina (MIR) measurements by spectral domain optical coherence tomography (SD-OCT) in Indian eyes early glaucoma.
In an observational, cross-sectional study, 125 eyes of 64 normal subjects and 91 eyes of 59 early glaucoma patients underwent RNFL and MIR imaging with SD-OCT. Glaucomatous eyes had characteristic optic nerve and RNFL abnormalities and correlating visual field defects and a mean deviation of better than or equal to -6 dB on standard automated perimetry. Areas under the receiver operating characteristic curves (AUC), sensitivities at a fixed specificity and likelihood ratios (LRs) were estimated for all RNFL and MIR parameters.
The AUCs for the RNFL parameters ranged from 0.537 for the temporal quadrant thickness to 0.821 for the inferior quadrant RNFL thickness. AUCs for the MIR parameters ranged from 0.603 for the superior minus inferior MIR thickness average to 0.908 for ganglion cell complex focal loss volume (GCC-FLV). AUC for the best MIR parameter (GCC-FLV) was significantly better (P<0.001) than that of the best RNFL parameter (inferior quadrant thickness). The sensitivities of these parameters at high specificity of 95%, however, were comparable (52.7% vs58.2%). Evaluation of the LRs showed that outside normal limits results of most of the RNFL and MIR parameters were associated with large effects on the post-test probability of disease.
MIR parameters with RTVue SD-OCT were as good as the RNFL parameters to detect early glaucoma.
比较频域光学相干断层扫描(SD-OCT)检测周边视网膜神经纤维层(RNFL)和黄斑内层视网膜(MIR)在印度早期青光眼患者中的诊断能力。
在一项观察性、横断面研究中,对 64 名正常受试者的 125 只眼和 59 名早期青光眼患者的 91 只眼进行了 SD-OCT 的 RNFL 和 MIR 成像。青光眼眼具有特征性视神经和 RNFL 异常,并伴有相应的视野缺损和标准自动视野计上平均偏差大于或等于-6dB。计算了所有 RNFL 和 MIR 参数的受试者工作特征曲线下面积(AUC)、固定特异性时的灵敏度以及似然比(LR)。
RNFL 参数的 AUC 范围从颞象限厚度的 0.537 到下象限 RNFL 厚度的 0.821。MIR 参数的 AUC 范围从上象限与下象限 MIR 厚度平均值的 0.603 到神经节细胞复合体焦点损失体积(GCC-FLV)的 0.908。最佳 MIR 参数(GCC-FLV)的 AUC 明显优于最佳 RNFL 参数(下象限厚度)(P<0.001)。然而,这些参数在高特异性为 95%时的灵敏度是相当的(52.7% vs 58.2%)。LR 的评估表明,大多数 RNFL 和 MIR 参数超出正常范围的结果与疾病后验概率的大影响有关。
与 RNFL 参数相比,RTVue SD-OCT 的 MIR 参数在检测早期青光眼方面同样有效。