Hackett Deborah A, Anderson Andrew J
Department of Optometry and Vision Sciences, The University of Melbourne, Melbourne, Victoria, Australia.
Optom Vis Sci. 2011 Jan;88(1):48-55. doi: 10.1097/OPX.0b013e3181fa6328.
Evidence for ganglion cell visual dysfunction in human glaucoma is often indirect, being either measured at the cellular level in animal models or being inferred from the pooled responses of a large number of ganglion cells in human observers. Rarebit perimetry (RBP) uses repeated, intense (150 cd/m2) stimuli-whose size is close to the spatial scale of a ganglion cell-to search for small retinal areas with zero sensitivity. Decreasing the stimulus luminance to 64 cd/m2 in normal observers does not alter the percentage of RBP stimuli detected [the mean hit rate (MHR)], and so we hypothesized that a similar response robustness should occur in glaucoma if the elements detecting the RBP target show no signs of visual dysfunction.
Nineteen glaucoma subjects and 19 age-matched controls were tested with a customized RBP test at 13 stimulus luminances (10 to 150 cd/m2, 0.14 log unit intervals). A four-parameter (threshold, spread, false positive proportion, and miss rate) cumulative Gaussian psychometric function was fitted to the response rate data from a glaucoma-affected region (glaucoma subjects; MHR >50% and <80%) and from the corresponding region in an age-matched normal control. Our hypothesis would predict that only the miss rates should differ between groups.
Glaucoma subjects showed significantly higher miss rates (0.18 vs. 0.04, p < 0.001), lower false positive proportions (0.009 vs. 0.025, p = 0.004), greater spreads (0.30 vs. 0.19, p = 0.002), and elevated thresholds [1.57 log(cd/m2) vs. 1.13 log(cd/m2), p < 0.001].
Responses to RBP stimuli are not robust to decreasing luminances in glaucoma. Our results more directly imply the presence of ganglion cell visual dysfunction in human glaucoma than studies using larger targets where contrast sensitivity losses could result through ganglion cell death alone. Such dysfunction may not be detected by Rarebit's MHR given that dysfunctional elements may still respond to the very intense RBP stimulus.
人类青光眼患者神经节细胞视觉功能障碍的证据往往是间接的，要么是在动物模型的细胞水平上测量，要么是从人类观察者大量神经节细胞的综合反应中推断出来。Rarebit视野计（RBP）使用重复的高强度（150 cd/m²）刺激——其大小接近神经节细胞的空间尺度——来寻找零敏感度的小视网膜区域。在正常观察者中，将刺激亮度降低到64 cd/m²不会改变检测到的RBP刺激的百分比[平均命中率（MHR）]，因此我们假设，如果检测RBP目标的元件没有视觉功能障碍的迹象，青光眼患者也会出现类似的反应稳健性。
对19名青光眼患者和19名年龄匹配的对照者进行定制的RBP测试，测试13种刺激亮度（10至150 cd/m²，0.14对数单位间隔）。将四参数（阈值、分布、假阳性比例和漏报率）累积高斯心理测量函数拟合到青光眼受累区域（青光眼患者；MHR>50%且<80%）和年龄匹配的正常对照相应区域的反应率数据。我们的假设预测，两组之间只有漏报率会有所不同。
青光眼患者的漏报率显著更高（0.18对0.04，p<0.001），假阳性比例更低（0.009对0.025，p = 0.004），分布更宽（0.30对0.19，p = 0.002），阈值更高[1.57 log（cd/m²）对1.13 log（cd/m²），p<0.001]。
青光眼患者对RBP刺激的反应在亮度降低时并不稳健。与使用较大目标的研究相比，我们的结果更直接地表明人类青光眼患者存在神经节细胞视觉功能障碍，在使用较大目标的研究中，对比度敏感度损失可能仅由神经节细胞死亡导致。鉴于功能失调的元件可能仍然对非常强烈的RBP刺激有反应，这种功能障碍可能无法通过Rarebit的MHR检测到。
