Department of Ophthalmology and Visual Sciences, Dalhousie University and Nova Scotia Health Authority, Halifax, Canada.
Department of Ophthalmology and Visual Sciences, Dalhousie University and Nova Scotia Health Authority, Halifax, Canada.
Ophthalmology. 2021 Apr;128(4):545-553. doi: 10.1016/j.ophtha.2020.08.033. Epub 2020 Sep 6.
To determine whether the 10-2 test of the Humphrey Field Analyzer detected a higher proportion of abnormal visual fields compared with the 24-2 test in the central 10° of patients with early glaucomatous visual field damage.
Prospective observational study.
Patients with open-angle glaucoma and healthy control participants.
All participants underwent a 24-2 and 10-2 test. Only the 12 central test locations of the 24-2 test were included to analyze equivalent visual field areas. The performance of the 2 tests was compared across 4 pointwise criteria: total deviation (TD) and pattern deviation (PD) analyses at the 5% and 2% levels. Analyses also were conducted for 2 pairs of follow-up tests, each performed 4 months apart.
(1) Area under the receiver operating characteristic curve (AUC), (2) sensitivity at identically matched specificity for the 4 criteria, (3) overlap (entire field and by quadrant) of abnormal visual fields with both tests, and (4) repeatability of the findings in 2 subsequent follow-up tests.
One eye each of 97 glaucoma patients (median mean deviation, -2.31 dB) and 65 control participants were included in the study. The AUCs for the 24-2 and 10-2 tests were not significantly different for any of the 4 criteria and ranged from 0.88 to 0.93 and from 0.91 to 0.94, respectively. At matched specificity, the sensitivity of the 24-2 test was significantly higher for all criteria except for PD analysis at 5%. In patients with an abnormal field with either test, the overlap varied from 60% to 86% depending on the criterion, whereas by quadrant, concordance ranged from 70% to 87%. Over the follow-up, the repeatability of test results (both 24-2 and 10-2 abnormal, either abnormal, or both normal) was achieved in 55% to 70% of patients.
In this study of glaucoma patients with early damage with the 24-2 test, there was little evidence that adding the 10-2 test revealed additional undetected defects in the central visual field. It may be more prudent to reserve 10-2 testing for following up selected patients with higher risk of central visual field progression.
在早期青光眼视野损害患者的中央 10°，比较 Humphrey 视野分析仪的 10-2 测试与 24-2 测试，检测异常视野的比例是否更高。
前瞻性观察性研究。
开角型青光眼患者和健康对照参与者。
所有参与者均接受 24-2 和 10-2 测试。仅纳入 24-2 测试的 12 个中央测试位置，以分析等效视野区域。通过 5%和 2%两个水平的总偏差（TD）和模式偏差（PD）分析，比较两种测试的性能。还对每 4 个月进行一次的 2 对随访测试的结果进行了分析。
（1）接收者操作特性曲线下的面积（AUC）；（2）4 项标准中完全匹配特异性的敏感性；（3）两种测试异常视野的重叠（整个视野和象限）；（4）2 次后续随访测试结果的可重复性。
本研究共纳入 97 例青光眼患者（中位数平均偏差为-2.31dB）和 65 例对照参与者的每只眼各 1 只。对于 4 项标准中的任何一项，24-2 和 10-2 测试的 AUC 均无显著差异，范围分别为 0.88 至 0.93 和 0.91 至 0.94。在匹配特异性的情况下，除 PD 分析在 5%水平外，24-2 测试的敏感性在所有标准中均显著更高。在两种测试中均存在异常视野的患者中，根据标准，视野重叠率从 60%到 86%不等，而按象限，一致性从 70%到 87%不等。在随访期间，55%至 70%的患者达到了 24-2 和 10-2 测试结果（两者均异常、任何一个异常或两者均正常）的可重复性。
在这项对 24-2 测试显示早期损害的青光眼患者的研究中，几乎没有证据表明增加 10-2 测试会发现中央视野中更多未被检测到的缺陷。对于中央视野进展风险较高的患者，保留 10-2 测试进行随访可能更为谨慎。
