PDF(Sci-hub)
Abstract：
OBJECTIVE: To evaluate in vivo confocal microscopy (IVCM) features of corneal subbasal nerve plexus (SNP) in the setting of dry eye disease (DED) using fully automated software \"ACCMetrics,\" and to further investigate its diagnostic performance in discriminating DED patients.
METHODS: IVCM exams of SNP in DED patients and matched control subjects were performed using Heidelberg Retina Tomograph with the Rostock Cornea Module. The following parameters were obtained with ACCMetrics: corneal nerve fiber density (CNFD), corneal nerve branch density (CNBD), corneal nerve fiber length (CNFL), corneal nerve total branch density (CTBD), corneal nerve fiber area (CNFA), corneal nerve fiber width (CNFW), and corneal nerve fractal dimension (CNFrD). The Mann-Whitney U test was used to compare variables. Receiver operating characteristic curves with calculations of the area under the curve (AUC) were used to describe the accuracy of IVCM parameters for discriminating DED patients from controls.
RESULTS: Thirty-nine DED patients and 30 control subjects were included. Significantly, lower values of CNFD, CNBD, and CNFL and higher value of CNFW were found in DED patients compared to controls (respectively, 20.5 ± 8.7 vs 25.4 ± 6.7 n/mm2; 25.6 ± 20.1 vs 37.6 ± 21.5 n/mm2; 12.6 ± 4.4 vs 14.5 ± 2.9 mm/mm2; 0.021 ± 0.001 vs 0.019 ± 0.001 mm/mm2; always p < 0.024). CNFW value had the highest diagnostic power in discriminating DED patients (AUC = 0.828). When the diagnosis of DED was made based on either CNFW or CNBD, the sensitivity was 97.4% and the specificity 46.7%.
CONCLUSIONS: The software ACCMetrics was able to rapidly detect SNP alterations occurring in the setting of DED and showed good diagnostic performance in discriminating DED patients.
METHODS: IVCM exams of SNP in DED patients and matched control subjects were performed using Heidelberg Retina Tomograph with the Rostock Cornea Module. The following parameters were obtained with ACCMetrics: corneal nerve fiber density (CNFD), corneal nerve branch density (CNBD), corneal nerve fiber length (CNFL), corneal nerve total branch density (CTBD), corneal nerve fiber area (CNFA), corneal nerve fiber width (CNFW), and corneal nerve fractal dimension (CNFrD). The Mann-Whitney U test was used to compare variables. Receiver operating characteristic curves with calculations of the area under the curve (AUC) were used to describe the accuracy of IVCM parameters for discriminating DED patients from controls.
RESULTS: Thirty-nine DED patients and 30 control subjects were included. Significantly, lower values of CNFD, CNBD, and CNFL and higher value of CNFW were found in DED patients compared to controls (respectively, 20.5 ± 8.7 vs 25.4 ± 6.7 n/mm2; 25.6 ± 20.1 vs 37.6 ± 21.5 n/mm2; 12.6 ± 4.4 vs 14.5 ± 2.9 mm/mm2; 0.021 ± 0.001 vs 0.019 ± 0.001 mm/mm2; always p < 0.024). CNFW value had the highest diagnostic power in discriminating DED patients (AUC = 0.828). When the diagnosis of DED was made based on either CNFW or CNBD, the sensitivity was 97.4% and the specificity 46.7%.
CONCLUSIONS: The software ACCMetrics was able to rapidly detect SNP alterations occurring in the setting of DED and showed good diagnostic performance in discriminating DED patients.
摘要：
目的：使用全自动软件“ACCMetrics”评估干眼病（DED）背景下角膜基底下神经丛（SNP）的体内共聚焦显微镜（IVCM）特征，并进一步研究其在鉴别DED患者中的诊断性能。
方法：使用带有Rostock角膜模块的海德堡视网膜断层扫描仪对DED患者和匹配的对照受试者进行IVCM检查。使用ACCMetrics获得以下参数：角膜神经纤维密度（CNFD），角膜神经分支密度（CNBD），角膜神经纤维长度（CNFL），角膜神经总支密度（CTBD），角膜神经纤维面积(CNFA),角膜神经纤维宽度(CNFW),和角膜神经分形维数(CNFrD)。Mann-WhitneyU检验用于比较变量。使用具有曲线下面积（AUC）计算结果的受试者工作特征曲线来描述IVCM参数的准确性，以区分DED患者与对照。
结果：纳入39名DED患者和30名对照受试者。重要的是,较低的CNFD值，CNBD,与对照组相比，DED患者的CNFL和CNFW值较高（分别为，20.5±8.7vs25.4±6.7n/mm2；25.6±20.1vs37.6±21.5n/mm2；12.6±4.4vs14.5±2.9mm/mm2；0.021±0.001vs0.019±0.001mm/mm2；始终p<0.024）。CNFW值在区分DED患者中具有最高的诊断能力（AUC=0.828）。当根据CNFW或CNBD诊断DED时，敏感性为97.4%,特异性为46.7%.
结论：软件ACCMetrics能够快速检测DED设置中发生的SNP改变，并在区分DED患者方面显示出良好的诊断性能。
方法：使用带有Rostock角膜模块的海德堡视网膜断层扫描仪对DED患者和匹配的对照受试者进行IVCM检查。使用ACCMetrics获得以下参数：角膜神经纤维密度（CNFD），角膜神经分支密度（CNBD），角膜神经纤维长度（CNFL），角膜神经总支密度（CTBD），角膜神经纤维面积(CNFA),角膜神经纤维宽度(CNFW),和角膜神经分形维数(CNFrD)。Mann-WhitneyU检验用于比较变量。使用具有曲线下面积（AUC）计算结果的受试者工作特征曲线来描述IVCM参数的准确性，以区分DED患者与对照。
结果：纳入39名DED患者和30名对照受试者。重要的是,较低的CNFD值，CNBD,与对照组相比，DED患者的CNFL和CNFW值较高（分别为，20.5±8.7vs25.4±6.7n/mm2；25.6±20.1vs37.6±21.5n/mm2；12.6±4.4vs14.5±2.9mm/mm2；0.021±0.001vs0.019±0.001mm/mm2；始终p<0.024）。CNFW值在区分DED患者中具有最高的诊断能力（AUC=0.828）。当根据CNFW或CNBD诊断DED时，敏感性为97.4%,特异性为46.7%.
结论：软件ACCMetrics能够快速检测DED设置中发生的SNP改变，并在区分DED患者方面显示出良好的诊断性能。
