Asymmetry of Peak Thicknesses between the Superior and Inferior Retinal Nerve Fiber Layers for Early Glaucoma Detection: A Simple Screening Method
10.3349/ymj.2018.59.1.135
- Author:
Hyoung Won BAE
1
;
Sang Yeop LEE
;
Sangah KIM
;
Chan Keum PARK
;
Kwanghyun LEE
;
Chan Yun KIM
;
Gong Je SEONG
Author Information
1. Department of Ophthalmology, Severance Hospital, Institute of Vision Research, Yonsei University College of Medicine, Seoul, Korea. gjseong@yuhs.ac
- Publication Type:Original Article
- Keywords:
Glaucoma;
diagnosis;
optical coherence tomography
- MeSH:
Area Under Curve;
Color;
Early Diagnosis;
Female;
Glaucoma/diagnosis;
Humans;
Male;
Mass Screening/methods;
Middle Aged;
Nerve Fibers/pathology;
ROC Curve;
Reproducibility of Results;
Retina/pathology;
Retinal Ganglion Cells;
Sensitivity and Specificity
- From:Yonsei Medical Journal
2018;59(1):135-140
- CountryRepublic of Korea
- Language:English
-
Abstract:
PURPOSE: To assess whether the asymmetry in the peripapillary retinal nerve fiber layer (pRNFL) thickness between superior and inferior hemispheres on optical coherence tomography (OCT) is useful for early detection of glaucoma. MATERIALS AND METHODS: The patient population consisted of Training set (a total of 60 subjects with early glaucoma and 59 normal subjects) and Validation set (30 subjects with early glaucoma and 30 normal subjects). Two kinds of ratios were employed to measure the asymmetry between the superior and inferior pRNFL thickness using OCT. One was the ratio of the superior to inferior peak thicknesses (peak pRNFL thickness ratio; PTR), and the other was the ratio of the superior to inferior average thickness (average pRNFL thickness ratio; ATR). The diagnostic abilities of the PTR and ATR were compared to the color code classification in OCT. Using the optimal cut-off values of the PTR and ATR obtained from the Training set, the two ratios were independently validated for diagnostic capability. RESULTS: For the Training set, the sensitivities/specificities of the PTR, ATR, quadrants color code classification, and clock-hour color code classification were 81.7%/93.2%, 71.7%/74.6%, 75.0%/93.2%, and 75.0%/79.7%, respectively. The PTR showed a better diagnostic performance for early glaucoma detection than the ATR and the clock-hour color code classification in terms of areas under the receiver operating characteristic curves (AUCs) (0.898, 0.765, and 0.773, respectively). For the Validation set, the PTR also showed the best sensitivity and AUC. CONCLUSION: The PTR is a simple method with considerable diagnostic ability for early glaucoma detection. It can, therefore, be widely used as a new screening method for early glaucoma.
