阈值视野计的可靠性指标在原发性开角型青光眼患者视野检查中的有效性研究

Study on the validity of threshold field of vision in the field of vision examination of POAG patients

目的评估Humphrey视野计的SITA standard和Full threshold程序的可靠性指标在原发性开角型青光眼患者视野检测中的有效性。方法从开角型青光眼出院患者中筛选两组各74例(74只眼)患者纳入本研究。患者一周内连续两次进行相同程序的视野检查,分别计算Full threshold组和SITA standard组两次重复检查差异值的标准差和可重复性系数,评价数据的可重复性;应用多元线性逐步回归的方法统计变异度与平均视野缺损(MD)、假阳性率(FP)、假阴性率(FN)及固视丢失率(FL)等可靠性检验指标的关联性,从而评价其有效性。结果两组中FL均显示了最高的重复系数分别为28.6%和27.8%(接近30%),即最低的可重复性。SITA standard组的FN的重复性系数为16.7%,比Full threshold组(FN的重复系数为24.7%)低,这可能也提示SITA standard程序的FN对患者的重复检查变异度有较好的预测性。两组的FP虽然均显示了更高的可重复性(更低的重复性系数),但其中较大部分原因是由于其拥有大量的0值,缺乏变异。虽然SITA standard组可靠性指标与重复检查变异度之间的相关性较Full threshold组有所提高,但总体相关性仍较弱,Full threshold组及SITA standard组分别只有5%及17%的重复检查变异度可以被三项可靠性指标所解释。其中,FN在两组中均占最高比例,分别为3%及13%。只有SITA standard组的FN与变异度呈显著相关(R^2=0.1324,P<0.05)。结论 SITA standard程序对视野检查结果可靠性检验的改进,在一定程度上提高了其有效性。但现有的可靠性指标,对重复检查变异度的预测性仍较弱,需要研发更有效的可靠性指标对视野检查结果进行判断。

Objective Evaluation of Humphrey perimetry of SITA standard program and Full threshold reliability index in primary open angle glaucoma detection effectiveness. Methods 74 patients （74 eyes）were selected from the outpatient clinics. Patients within a week twice in a row of visual field examination in the same procedure, the calculated SITA standard group and Full threshold group repeated twice to examine differences in values of standard deviation and coefficient of repeatability, data evaluation can repeat; application multiple linear stepwise regression method statistical variation degree and average visual field defect （MD）, false positive rate （FP） and false negative rate （FN） and the relevance of the fixation loss （FL） to check the reliability of indicators, so as to evaluate its effectiveness. Results Although SITA standard tests showed better correlation between reliability indices and test-retest variability than Full threshold tests, the total relationship was still poor. Only 5% and 17% of the variability in test-retest results being explained by the Full Threshold and SITA standard indices respectively. The false-negative index was the most significant, accounting for 3% and 13% of the variability in Full Threshold and SITA standard tests. Only false-negative in SITA standard group showed significant eorrelation with variability （R2 = 0. 132 4, P〈0.05 ）. Conclusion Standard SITA program to improve the reliability of the results of the field test results, to a certain extent, improve its effectiveness. However the existing reliability indexes are still weak, and it is necessary to develop more effective reliability index to judge the result of the visual field test.