Pentacam眼前节分析系统在圆锥角膜早期诊断中的应用

Application of Pentacam in the early diagnosis of keratoconus

目的探讨临床期圆锥角膜和亚临床期圆锥角膜与正常角膜的形态变化以及Pentacam眼前节分析系统在圆锥角膜早期诊断中的意义。方法病例对照研究。根据Rabinowitz诊断标准将35例圆锥角膜患者分成临床期圆锥角膜组16例（16眼）、亚临床期圆锥角膜组19例（19眼）．并选取有近视散光的患者29例（29眼）作为正常对照组。应用Pentacam眼前节分析系统检测3组患者的角膜前表面形态学参数：表面变异指数（ISV）、垂直不对称指数（IVA）、圆锥角膜指数（KI）、中央圆锥角膜指数（CKI）、高度不对称性指数（IHA）、高度离心指数（IHD）、最小曲率半径（Rmin）和不规则指数（ABR）。利用方差分析、Dunnett．t检验和受检者工作特征曲线（ROC曲线）分析方法对各参数进行比较研究，最终利用偏最小二乘（Partial Least Squares，PLS）方法构建临床期圆锥角膜和亚临床期圆锥角膜的早期诊断模型。结果角膜前表面形态学参数在临床期圆锥角膜组、亚临床期圆锥角膜组和正常对照组之间存在差异。ROC曲线分析结果显示临床期圆锥角膜组、亚临床期圆锥角膜组中各参数的曲线下面积（AUC）均接近于1，说明其诊断准确性较高，并且分别得出临床期圆锥角膜和亚临床期圆锥角膜的诊断界值：ISV为39和22．4，Rmin为6．765和7．11，KI为1．085和1．045。应用PLS方法构建临床期圆锥角膜的早期诊断模型：Yhat=8．231＋0．006ISV＋0．843IVA＋0．869KI-3．186CKI-0．010IHA＋0．679IHD-0．649Rmin-0．057ABR。亚临床期圆锥角膜早期诊断的PLS模型：Yhat=7．861—0．563Rmin＋0．018ISV＋0．390IVA-3．281KI＋0．571CKI-0．010IHA＋0．727IHD-0．009ABR。结论Pentacam眼前节分析系统对临床期圆锥角膜及亚临床期圆锥角膜的早期诊断有重要作用。

Objective To investigate the change in morphologic parameters using the Pentacam to discriminate between clinical keratoconus eyes, subclinical keratoconus eyes and normal eyes for early diagnosis. Methods This was a case control study. Adopting the Rabinowitz diagnosis standards, we divided 35 patients with keratoconus into 2 subgroups： 16 patients （16 eyes） with clinical keratoconus and 19 patients （19 eyes） with subclinical keratoconus. Twenty-nine patients （29 eyes） with myopic astigmatism were used as a control group. The following morphologic parameters of the anterior corneal surface provided by Pentacam in the clinical keratoconus group, subclinical keratoconus group and control group were recorded： index of surface variance （ISV）, index of vertical asymmetry （IVA）, keratoconus index （KI）, central keratoconus index （CKI）, index of highest asymmetry （IHA）, index of highest decentration （IHD）, minimum sagittal curvature （Rmin） and aberration coefficient （ABR）. An analysis of variance, Dunnett-t test and receiver operating characteristic （ROC） curves were used for analysis and partial least squares （PLS） was used to construct models with corneal morphologic parameters of the anterior corneal surface for early diagnosis. Results There were significant differences between the clinical keratoconus group, subclinical keratoconus group and control group. The values of the area under the curve （AUC） were all close to 1, and the diagnosis accuracy was high. The optimum cutoff points of the ISV for clinical keratoconus and subclinical keratoconus were 39 and 22.4, Rmin was 6.765 and 7.11, and KI was 1.085 and 1.045. The PLS model of clinical keratoconus was Yhat=8.231＋0.006xlSV＋ 0.843x IVA＋ 0.869xKI- 3.186xCKI- 0.010xlHA＋ 0.679xlHD- 0.649xRmin- 0.057xABR,and the PLS model of subclinical keratoconus was Yhat=7.861-0.563xRmin＋ 0.018x ISV＋ 0.390xlVA- 3.281＋KI＋ 0.571x CKI- 0.010xlHA＋ 0.727 xlHD- 0.009 xABR. Conclusion Pentacam is important in the early diagnosis of clinical keratoconus eyes and subclinical keratoconus.