Department of Ophthalmology, University of California, San Francisco, CA, USA.
J Glaucoma. 2011 Feb;20(2):87-94. doi: 10.1097/IJG.0b013e3181d787b6.
To evaluate the capability of the optic disc, peripapillary retinal nerve fiber layer (P-RNFL), macular inner retinal layer (M-IRL) parameters, and their combination obtained by Fourier-domain optical coherent tomography (OCT) in differentiating a glaucoma suspect from perimetric glaucoma.
Two hundred and twenty eyes from 220 patients were enrolled in this study. The optic disc morphology, P-RNFL, and M-IRL were assessed by the Fourier-domain OCT (RTVue OCT, Model RT100, Optovue, Fremont, CA). A linear discriminant function was generated by stepwise linear discriminant analysis on the basis of OCT parameters and demographic factors. The diagnostic power of these parameters was evaluated with receiver operating characteristic (ROC) curve analysis. The diagnostic power in the clinically relevant range (specificity ≥ 80%) was presented as the partial area under the ROC curve (partial AROC).
The individual OCT parameter with the largest AROC and partial AROC in the high specificity (≥ 80%) range were cup/disc vertical ratio (AROC = 0.854 and partial AROC = 0.142) for the optic disc parameters, average thickness (AROC = 0.919 and partial AROC = 0.147) for P-RNFL parameters, inferior hemisphere thickness (AROC = 0.871 and partial AROC = 0.138) for M-IRL parameters, respectively. The linear discriminant function further enhanced the ability in detecting perimetric glaucoma (AROC = 0.970 and partial AROC = 0.172).
Average P-RNFL thickness is the optimal individual OCT parameter to detect perimetric glaucoma. Simultaneous evaluation on disc morphology, P-RNFL, and M-IRL thickness can improve the diagnostic accuracy in diagnosing glaucoma.
评估频域光相干断层扫描(OCT)获得的视盘、视盘周围视网膜神经纤维层(P-RNFL)和黄斑内层视网膜(M-IRL)参数及其组合在鉴别青光眼疑似患者和周边视野青光眼中的能力。
本研究纳入了 220 例 220 只眼。使用频域 OCT(RTVue OCT,Model RT100,Optovue,加利福尼亚州弗里蒙特)评估视盘形态、P-RNFL 和 M-IRL。基于 OCT 参数和人口统计学因素,通过逐步线性判别分析生成线性判别函数。使用受试者工作特征(ROC)曲线分析评估这些参数的诊断能力。以 ROC 曲线下面积的部分(partial AROC)表示在临床相关范围内(特异性≥80%)的诊断能力。
在高特异性(≥80%)范围内,ROC 曲线下面积和部分 AROC 最大的单个 OCT 参数分别为视盘参数中的杯盘垂直比(AROC=0.854,partial AROC=0.142)、P-RNFL 参数中的平均厚度(AROC=0.919,partial AROC=0.147)、M-IRL 参数中的下半球厚度(AROC=0.871,partial AROC=0.138)。线性判别函数进一步提高了检测周边视野青光眼的能力(AROC=0.970,partial AROC=0.172)。
平均 P-RNFL 厚度是检测周边视野青光眼的最佳 OCT 单项参数。同时评估视盘形态、P-RNFL 和 M-IRL 厚度可以提高诊断青光眼的准确性。