Estimating the Reliability of Glaucomatous Visual Field for the Accurate Assessment of Progression Using the Gaze-Tracking and Reliability Indices
Estimating the Reliability of Glaucomatous Visual Field for the Accurate Assessment of Progression Using the Gaze-Tracking and Reliability Indices
复制标题
DOI：
10.1016/j.ogla.2019.02.001
复制
发表时间：
2019-03-01
影响因子：
2.9
通讯作者：
Murata, Hiroshi
中科院分区：
文献类型：
--
作者：
Asaoka, Ryo;Fujino, Yuri;Murata, Hiroshi
目的：探讨视野 (VF) 可靠性指数和注视追踪 (GT) 在准确评估青光眼进展方面的有用性。设计：机构实践。参与者：304 名开角型青光眼患者的 483 只眼睛。方法：在每只眼睛中检查一系列 10 个 VF。使用VF打印输出底部的GT图表（汉弗莱场分析仪，瑞典交互式阈值算法标准），计算平均跟踪失败频率（TFF）、平均眨眼频率（BF）和每次刺激的眼球运动总量（MPS）。平均总偏差 (mTD) 是根据每个 24-2 VF 中的 52 个总偏差值计算的，并对随时间推移的一系列 mTD 测量值应用线性回归。进展被认为是基于 2 个单独的定义发生的：(1) 所有 10 个 VF (VF1-10) 的 mTD 率均显着 (P < 0.05)，且小于 0 dB/年；(2) mTD 率小于 -0.5 dB/年，且 P < 0.05 显着。 mTD 率也在较短的 VF 系列 (VF1-5-VF1-9) 中计算。根据较短 VF 系列 (n < 10) 的 mTD 率与所有 10 个 VF 进展诊断之间的关系计算受试者工作特征曲线下面积 (AUC)。在排除标准 VF 可靠性指数（注视丧失 [FL]、假阳性 [FP] 率和假阴性 [FN] 率）和 GT 参数（TFF、BF 和 MPS）的截止值越来越严格的眼睛后，重复进行该分析。主要结果指标：不同 GT 和标准可靠性指数排除标准的受试者工作特征曲线值下面积。结果：根据更严格的 FL 和 FP 标准，AUC 基本没有变化。 AUC 随着更严格的 FN 标准而增加，但仅当使用第二个定义描述进展时（小于 -0.5 dB/年的比率）。随着 TFF、BF 和某种程度上 MPS 标准的更严格，AUC 增加。结论：当 FN、TFF、BF 和 MPS 指数更严格时，平均总偏差进展率更可靠。评估青光眼进展时应考虑注视追踪结果。 (C) 2019 年美国眼科学会
Purpose: To investigate the usefulness of visual field (VF) reliability indices and gaze tracking (GT) for the accurate assessment of progression in glaucoma.Design: Institutional practice.Participants: Four hundred eighty-three eyes of 304 patients with open-angle glaucoma.Methods: In each eye, a series of 10 VFs were examined. Using the GT chart at the bottom of the VF printout (Humphrey Field Analyzer, Swedish interactive threshold algorithm standard), average tracking failure frequency (TFF), average blinking frequency (BF), and the total amount of eye movement per stimulus (MPS) were calculated. Mean total deviation (mTD) was calculated based on the 52 total deviation values in each 24-2 VF, and linear regression was applied to the series of mTD measurements overtime. Progression was deemed to have occurred based on 2 separate definitions: (1) the mTD rate across all 10 VFs (VF1-10) was significant (P < 0.05) and less than 0 dB/year and (2) the mTD rate was less than -0.5 dB/year and significant at P < 0.05. The mTD rate also was calculated in shorter VF series (VF1-5-VF1-9). The area under the receiver operating characteristic curve (AUC) was calculated from the relationship between mTD rates of shorter VF series (n < 10) and the diagnosis of progression in all 10 VFs. This analysis was iterated after excluding eyes with increasingly strict cutoff values for the standard VF reliability indices (fixation loss [FL], false-positive [FP] rate, and false-negative [FN] rate) and the GT parameters (TFF, BF, and MPS).Main Outcome Measures: Area under the receiver operating characteristic curve values with different exclusion criteria for GT and standard reliability indices.Results: The AUC was largely unchanged according to stricter FL and FP criteria. The AUC increased with a more stringent FN criterion, but only when progression was described using the second definition (a rate of less than -0.5 dB/year). The AUC increased with stricter criteria for TFF, BF, and, to some extent, MPS.Conclusions: Mean total deviation progression rates are more reliable when FN, TFF, BF, and MPS indices are stricter. Gaze-tracking results should be considered when assessing glaucomatous progression. (C) 2019 by the American Academy of Ophthalmology