[Influence factors and differences of posterior corneal elevation measured by Pentacam system combined with Corvis ST].

Objective: To investigate the influence factors and differences of abnormal posterior corneal elevation by Pentacam system and Corvis ST. Methods: This retrospective case series study included 227 eyes of 144 patients (90 males, 139 eyes; 54 females, 88 eyes) from December 2017 to October 2018 who were going to receive corneal refractive surgery at the Corneal Refraction Department of Qingdao Eye Hospital. The general data of the patients including gender, age, refractive parameters, optimal correction of spherical and cylindrical diopters were collected. All patients underwent Pentacam system and Corvis ST measurement. According to the back difference (BD) of Pentacam parameters, BD<12 μm was set as the control group (59 patients, 118 eyes) and BD≥12 μm as the high BD group (85 patients, 109 eyes). In the high BD group, BD≤16 μm was set as the suspicious group (44 patients, 53 eyes), while BD>16 μm was set as the abnormal group (41 patients, 56 eyes). Seven parameters of Pentacam and 15 parameters of Corvis ST were selected. The Pentacam parameters included BD, anterior surface keratometry (ASK), posterior surface keratometry (PSK), anterior surface astigmatism (AAstig), posterior surface astigmatism (PAstig), central corneal thickness (CCT), and corneal diameter (W-W). The parameters of Corvis ST included the first applanation time (AT(1)), the first applanation length (AL(1)), the first applanation velocity (AV(1)), the second applanation time (AT(2)), the second applanation length (AL(2)), the second applanation velocity (AV(2)), highest concavity time (HCT), highest concavity peak distance (HC-PD), highest concavity deformation amplitude (HC-DA), highest concavity radius (HC-R), the ratio of deformation amplitude (DA ratio), Integr. Radius, corneal thickness thinnest/pachymetric progression (ARTh), SPA1 (resultant pressure divided by deflection amplitude at the first applanation), and the Corvis Biomechanical Index (CBI). The comparison between the groups was analyzed with Independent sample t test, Kruskal-Wallis H test, and Bonferroni test. Spearman rank correlation analysis was used to explore the correlation factors of BD, and the main factors affecting BD were found through multiple linear regression. Results: There were no statistically significant differences between the control group and the high BD group in age, spherical diopters, and cylindrical diopters (t=-3.311, -1.808, -2.359; P=0.071, 0.072, 0.121, respectively). In Pentacam parameters, ASK, PSK, PAstig, and W-W showed significant differences among groups (Z=18.492, 31.547, 10.773, 70.167; P<0.05). AAstig and CCT showed no statistical difference between groups (P>0.05). Compared with the control group [42.80 (41.98, 44.00)], ASK increased in the abnormal group [43.40 (42.20, 44.40)] significantly (t=-4.292; P<0.05). PSK of the suspicious group [-6.50 (-6.60, -6.35)] and the abnormal group [-6.50 (-6.70, -6.33)] increased significantly compared with the control group [-6.30 (-6.50, -6.20)] (t=4.492, 4.618; P<0.05). Compared with the control group [0.40 (0.30, 0.50)], PAstig of the suspicious group [0.40 (0.30, 0.40)] and the abnormal group [0.40 (0.30, 0.40)] increased significantly (t=2.796, 2.515; P=0.016, 0.036). Compared with the control group [11.50 (11.40, 11.80)], W-W of the suspicious group [11.40 (11.00, 11.60)] and the abnormal group [11.10 (10.90, 11.30)] decreased, and W-W of the abnormal group also decreased significantly compared with the suspicious group (t=3.235, 8.353, 4.282; P<0.05). The correlation analysis between BD and Pentacam parameters of patients in each group showed that BD was negatively correlated with W-W (r=-0.614, -0.304, -0.396, -0.661, P<0.05) in the control group, the suspicious group, the abnormal group, and all patients, while BD had a low correlation with other parameters or no significant correlation. The correlation analysis of BD and Corvis ST parameters in patients showed that only in the suspicious group, BD was positively correlated with AV(1), HCT, and HC-DA (r=0.332, 0.361, 0.382, P<0.05), while no significant correlation was found between BD and other Corvis ST parameters in each group. In order to further explore the main factors affecting BD, Pentacam parameters and Corvis ST parameters were selected as independent variables with BD as the dependent variable to establish a multivariate linear regression analysis model. There was no collinearity between variables W-W, ASK, PSK, HC-PD, SPA1, and CCT (tolerance<0.100). The equation test result was F=37.221, P<0.001, adjusted r(2)=0.504, and the fitting was good. Conclusions: Among the Pentacam parameters, W-W, ASK, and PSK are the main factors affecting the change of BD. HC-PD and SPA1 in the Corvis ST parameters may also have some influence on BD. The Pentacam system combined with Corvis ST is a very useful differential diagnosis system for patients with abnormal BD. (Chin J Ophthalmol, 2020, 56:110-117).目的： 探讨角膜形态学和生物力学参数对近视眼人群角膜后表面高度异常的影响及其辅助诊断角膜后表面高度异常的可行性。 方法： 病例对照研究。选取2017年12月至2018年10月在青岛眼科医院拟行角膜屈光手术的屈光不正患者144例（227只眼），其中男性90例（139只眼），女性54例（88只眼）；年龄（22.8±5.6）岁。所有患者检查最佳矫正视力下的球镜和柱镜度数，并行Pentacam眼前节分析系统和Corvis ST角膜生物力学检查。根据Pentacam参数角膜后表面高度变化值（BD）进行分组，将BD<12 μm定为对照组（59例，118只眼）；将BD≥12 μm定为高BD组（85例，109只眼），其中BD≤16 μm者为可疑组（44例，53只眼），>16 μm者为异常组（41例，56只眼）。选取Pentacam眼前节分析系统中BD、前表面曲率（ASK）、后表面曲率（PSK）、前表面散光（AAstig）、后表面散光（PAstig）、中央角膜厚度（CCT）、角膜直径（W-W）7个参数及Corvis ST角膜生物力学检测系统中第1次压平时间（AT(1)）、第1次压平长度（AL(1)）、第1次压平速度（AV(1)）、第2次压平时间（AT(2)）、第2次压平长度（AL(2)）、第2次压平速度（AV(2)）、最大压陷时间（HCT）、最大压陷峰距（HC-PD）、最大压陷形变幅度（HC-DA）、最大压陷曲率半径（HC-R）、形变幅度比值（DA Ratio）、综合半径、最薄点厚度/厚度变化率（ARTH）、硬度参数（SPA1）、Corvis生物力学指数（CBI）15个参数纳入研究。采用独立样本t检验、Kruskal-Wallis H检验及Bonferroni法进行组间差异性比较，利用Spearman秩相关分析探讨BD的相关影响因素，并建立多元线性回归模型寻找影响BD的主要因素。 结果： 对照组和高BD组在年龄、球镜度数、柱镜度数差异无统计学意义（t=-3.311，-1.808，-2.359；P=0.071，0.072，0.121）；Pentacam测量发现在对照组、可疑组和异常组中，ASK、PSK、PAstig、W-W组间差异有统计学意义（Z=18.492，31.547，10.773，70.167；P<0.05），AAstig、CCT组间差异无统计学意义（Z=2.204，1.108；P>0.05）。其中异常组[43.40（42.20，44.40）]较对照组[42.80（41.98，44.00）]ASK增大（t=-4.292；P<0.05），可疑组[-6.50（-6.60，-6.35）]与异常组[-6.50（-6.70，-6.33）]较对照组[-6.30（-6.50，-6.20）]PSK均增大（t=4.492，4.618；P<0.05）；可疑组[0.40（0.30，0.40）]与异常组[0.40（0.30，0.40]较对照组[0.40（0.30，0.50）] PAsting均增大（t=2.796，2.515；P=0.016，0.036）；可疑组[11.40（11.00，11.60）]和异常组[11.10（10.90，11.30）]较对照组[11.50（11.40，11.80）]W-W减小，同时异常组较可疑组W-W也明显减小（t=3.235，8.353，4.282；P<0.05）；Corvis ST参数在3个组间差异无统计学意义（P>0.05）。将每组患者BD与Pentacam参数进行相关性分析发现，在对照组、可疑组、异常组以及所有患者中，BD与W-W呈低至中度负相关（r=-0.614，-0.304，-0.396，-0.661；P<0.05），BD与其他参数相关性较低或未见明显相关性；将患者BD与Corvis ST参数进行相关性分析发现，仅在可疑组中，BD与AV(1)、HCT、HC-DA间存在低度正相关（r=0.332，0.361，0.382；P<0.05），BD与Corvis ST其他参数在各组中未见明显相关性。以BD为因变量，选取Pentacam参数与Corvis ST参数建立多元线性回归分析模型：变量W-W、ASK、PSK、HC-PD、SPA1、CCT间不存在共线性（容忍度<0.100），方程检验结果F=37.221，P<0.001，调整r(2)=0.504，拟合较好。 结论： 角膜形态学参数中角膜直径、前表面曲率、后表面曲率是影响BD的主要因素，角膜生物力学参数中HC-PD和SPA1也可能对BD产生一定程度影响。角膜形态学和生物力学参数的联合评估有助于角膜后表面高度异常的鉴别诊断。（中华眼科杂志，2020，56：110-117）.