倍频视野计检测青光眼性视野缺损能力的研究

Studies of frequency doubling perimetry in detection of glaucomatous visual field loss

目的评估倍频视野计检测青光眼性视野缺损的能力以及与OCTOPUS101全自动视野计检查结果之间的相关性。方法应用倍频视野计的C-20-5筛选程序以及OC-TOPUS101全自动视野计的G2-TOP程序对23例正常对照者、20例早期青光眼患者、35例中晚期青光眼患者、11例高眼压患者及13例疑似青光眼患者进行视野检测。正常对照组、高眼压组及疑似青光眼组随机选择一眼进行测试,青光眼组选择具有较严重视野缺损的一眼进行测试。结果倍频视野计的C-20-5筛选程序在检测青光眼时ROC曲线下面积为0.925(敏感性85%,特异性91%),与OCTOPUS视野指数-平均缺损、偏离缺失之间的Pearson系数分别为0.702与0.429(P<0.001),倍频视野计与OCTOPUS101视野计检查平均所需时间分别为1.00min与2.33min,2者之间有明显差异(P<0.001)。结论倍频视野计检测青光眼性视野缺损具有良好的敏感性与特异性,与OCTOPUS101视野计的视野指数之间亦存在理想的相关性,而且前者比后者检测速度更快,使大规模人群筛查成为可能。

Objective To evaluate the ability of frequency doubling perimetry in detecting glaucomatous visual field loss, and to ascertain whether FDP could provide results comparable with those of automated OCTOPUS 101 perimetry. Methods The visual fields of the participants were assessed by OCTOPUS program G2-TOP and FDP program C-20-5 screening. The participants include 23 normal, 20 early glaucoma, 35 moderate-advanced glaucoma, 11 high ocular hypertension and 13 suspected glaucoma. Only 1 eye of each participants was selected randomly except the glaucoma group, in which the eye with more severe visual field loss was selected. Results For the FDP program C-20-5 screening, area under the ROC curve is 0. 925 （ sensitivity 85 %, specificity 91% ） in detecting glaucomatous visual field loss. The pearson coefficient between FDP results and OCTOPUS indices-mean defect （MD） and loss variance （LV） are 0. 702 and 0. 429（P 〈 0. 001 ）, respectively. The time cost by FDP and OCTOPUS101 perimetry are 1.00 min and 2.33 min, respectively. A significant difference was found between the two visual field tests （P 〈 0.001）. Conclusion Frequency doubling perimetry demonstrates high sensitivity and specificity in the detection of glaucomatous visual field loss. A significant correlation was fould between FDP and OCTOPUS indices. The FDP is a faster way to analyze the visual field than OCTOPUS101 perimetry. All these foundings make a big population screening possible.