Chang Yo-Chen, Tsai Rong-Kung
Department of Ophthalmology, Kaohsiung Municipal Hsiao-Kang Hospital, Kaohsiung Medical University, Kaohsiung, Taiwan.
Kaohsiung J Med Sci. 2008 May;24(5):233-9. doi: 10.1016/S1607-551X(08)70147-5.
The purpose of this study was to correlate quadrant specific Humphrey visual field mean deviation (MD) with retinal nerve fiber layer (RNFL) thickness as measured by scanning laser polarimetry (GDx), and to determine whether there is a difference in the correlation with visual field defect between the Asian normative database provided by GDx (GDx database) and our native normative database (KMU database). In an age-matched study, a control group of 240 normal eyes underwent GDx. Another 60 eyes with visual field defect due to primary angle glaucoma underwent autoperimetry and GDx examination. First, we compared four GDx measurements between the control and study groups. Next, we divided the visual field into four quadrants (superior, inferior, temporal, nasal) and calculated the MD of each quadrant. We correlated the MD of superior, inferior and overall visual field with RNFL thickness judged by two databases (the GDx Asian internal normative database and the database from our control group). GDx detected abnormal RNFL thickness significantly more accurately when using the KMU database (p = 0.0473 for superior quadrant; p = 0.0074 for inferior quadrant; p = 0.0011 for average thickness) than when using the GDx database. There was no significant difference in the specificity between these two databases. The normal ranges in the GDx internal normative database for Asians are too wide. By using our own GDx normative database, the correlations with MD of autoperimetry were significantly improved. We suggest that every laboratory and clinic establish its own normative database of GDx in Asia.
本研究的目的是将象限特异性的 Humphrey 视野平均偏差(MD)与通过扫描激光偏振仪(GDx)测量的视网膜神经纤维层(RNFL)厚度相关联,并确定 GDx 提供的亚洲标准数据库(GDx 数据库)与我们本地的标准数据库(KMU 数据库)在与视野缺损的相关性方面是否存在差异。在一项年龄匹配的研究中,240 只正常眼的对照组接受了 GDx 检查。另外 60 只因原发性闭角型青光眼导致视野缺损的眼睛接受了自动视野计检查和 GDx 检查。首先,我们比较了对照组和研究组之间的四项 GDx 测量值。接下来,我们将视野分为四个象限(上、下、颞、鼻),并计算每个象限的 MD。我们将上、下象限以及整个视野的 MD 与两个数据库(GDx 亚洲内部标准数据库和我们对照组的数据库)判断的 RNFL 厚度相关联。与使用 GDx 数据库相比,使用 KMU 数据库时,GDx 检测到异常 RNFL 厚度的准确性明显更高(上象限 p = 0.0473;下象限 p = 0.0074;平均厚度 p = 0.0011)。这两个数据库之间的特异性没有显著差异。GDx 亚洲内部标准数据库中亚洲人的正常范围太宽。通过使用我们自己的 GDx 标准数据库,与自动视野计 MD 的相关性得到了显著改善。我们建议每个实验室和诊所都建立自己的亚洲 GDx 标准数据库。