Department of Biostatistics, Graduate School of Medical Science, Kyoto Prefectural University of Medicine, Kyoto, Japan.
Department of Biostatistics, Graduate School of Medical Science, Kyoto Prefectural University of Medicine.
J Glaucoma. 2024 Nov 1;33(11):849-854. doi: 10.1097/IJG.0000000000002476. Epub 2024 Aug 1.
Analysis of Ambient Interactive Zippy Estimation of Sequential Testing (AIZE) Rapid test variability in patients with stable glaucoma showed that the 95% prediction interval of the mean deviation (MD) value, potentially an index for judging progression, was ±1.63 to ±1.78 dB in early-to-moderate-stage patients.
To explore the 95% prediction interval of the MD value using the AIZE Rapid test strategy for glaucoma observation.
This study included 72 patients with stable or suspected glaucoma who underwent the imo AIZE Rapid test 3 times or more within 2 years. Both eyes from each patient were classified as better or worse eyes. They were divided based on baseline MD values into the following 4 groups: MD > -3 dB, -6 dB < MD ≤ -3 dB, -12 dB < MD ≤ -6 dB, and MD ≤ -12 dB. The variability of MD during the observation period and the 95% prediction intervals were analyzed. Pointwise variability of limits at each test location was also calculated.
The numbers of better and worse eyes included in the study were 46 and 33. The median follow-up period was 1.3 years (range: 0.5 to 1.9 y). The 95% prediction intervals for MD values were ±1.41 dB for better eyes (n = 46) and ±1.47 dB for worse eyes (n = 33). The 95% prediction intervals in the MD > -3 dB, -6 dB < MD ≤ -3 dB, -12 dB < MD ≤ -6 dB, and MD ≤ -12 dB groups were ±1.63 dB, ±1.34 dB, ±1.78 dB, and ±1.33 dB, respectively. Pointwise variability of worse eyes was larger than that of better eyes, especially between 10 to 15 dB.
In the case of a difference in MD greater than the 95% prediction intervals when compared with the previous visual field result, we should pay much attention to the possibility of progression of the glaucomatous visual field in patients with stable glaucoma.
对稳定青光眼患者的环境交互 Zippy 估计序贯测试（AIZE）快速检测变异性进行分析显示，平均偏差（MD）值的 95%预测区间可能是判断进展的指标，在早期至中期患者中为±1.63 至±1.78 dB。
使用 AIZE 快速检测策略探讨 MD 值的 95%预测区间，以观察青光眼。
本研究纳入了 72 名稳定或疑似青光眼患者，他们在 2 年内接受了 imo AIZE Rapid 测试 3 次或以上。每位患者的双眼均分为较好眼或较差眼。根据基线 MD 值将患者分为以下 4 组：MD > -3 dB、-6 dB < MD ≤ -3 dB、-12 dB < MD ≤ -6 dB 和 MD ≤ -12 dB。分析观察期间 MD 的变异性和 95%预测区间。还计算了每个测试位置的点预测限的变异性。
研究中包括 46 只较好眼和 33 只较差眼。中位随访时间为 1.3 年（范围：0.5 至 1.9 年）。较好眼（n = 46）的 MD 值 95%预测区间为±1.41 dB，较差眼（n = 33）的 MD 值 95%预测区间为±1.47 dB。MD > -3 dB、-6 dB < MD ≤ -3 dB、-12 dB < MD ≤ -6 dB 和 MD ≤ -12 dB 组的 MD 值 95%预测区间分别为±1.63 dB、±1.34 dB、±1.78 dB 和±1.33 dB。较差眼的点预测限变异性大于较好眼，尤其是在 10 至 15 dB 之间。
与之前的视野结果相比，如果 MD 的差异大于 95%预测区间，我们应该高度关注稳定青光眼患者的青光眼视野进展的可能性。
