原发性开角型青光眼与正常人视乳头周围脉络膜厚度的比较

Peripapillary choroidal thickness in primary open angle glaucoma and normal subjects measured by enhanced depth imaging optical coherence tomography

目的研究原发性开角型青光眼（POAG）患者视乳头周围脉络膜厚度与正常人之间的差异及其与视野缺损间的关系。方法横断面研究。应用增强深部成像的相干光断层扫描（EDIOCT）仪对40例（40只眼）POAG患者和41例（41只眼）正常人进行以视乳头中心为圆心、直径为3．4mm圆周的视网膜神经纤维层（RNFL）厚度测量。再由检测者手动描绘视网膜色素上皮外缘和巩膜内界面，由软件自动获得相应部位的脉络膜厚度；信度分析采用组内相关系数（ICC），其值越大表明测量结果越稳定、可重复性越好。POAG患者与正常人视乳头周围各象限RNFL和脉络膜厚度比较，采用成组设计t检验；视乳头周围各象限脉络膜厚度比较，采用重复测量的方差分析和LSD-t检验（该软件仅提供P值）；脉络膜厚度与年龄和RNFL厚度的相关性分析，采用Pearson相关分析法；视乳头全周RNFL和脉络膜平均厚度与视野平均缺损（非正态分布）之间的相关性分析，采用Spearman秩相关分析法。结果测量各象限脉络膜厚度的信度分析，检测者内ICC值为0．993～0．999，检测者间ICC值为0．993～0．998。POAG患者和正常人各象限脉络膜厚度均随年龄增加而变薄（r=-0．509～-0．608，-0．549～-0．668；P〈0．05），但与相应部位的RNFL厚度之间无相关性（r=-0．186～0．273，-0．090～0．132；P〉0．05）；视乳头周围脉络膜厚度特征为颞上、鼻上象限最厚，其次为鼻、颞侧，颞下和鼻下象限最薄（P=0．000-0．036）。POAG患者视乳头全周RNFL厚度为（64．8±14．3）μm，脉络膜厚度为（139．6±60．3）μm；正常对照者视乳头全周RNFL厚度为（106．5±7．8）μm，脉络膜厚度为（138．2±56．7）μm；患者与对照者全周RNFL厚度差异有统计学意义（t=16．087，P〈0．05），但脉络膜厚度差异无统计学意义（t=-0．107，P〉0．05）。POAG患者其余各象限RNFL厚度也均明显薄于正常人（t=6．975～13．454，P〈0．05），而脉络膜厚度的差异无统计学意义（t=-0．474～0．367，P〉0．05）。POAG患者视乳头全周RNFL平均厚度与视野平均缺损之间呈负相关性（rs=-0．511，P〈0．05），而脉络膜平均厚度与视野平均缺损之间无相关性（rs=-0．285，P〉0．05）。结论POAG患者的视乳头周围脉络膜厚度与正常人并无不同，且与RNFL厚度、视野平均缺损之间无相关性。因此，不支持将视乳头周围脉络膜厚度作为诊断POAG和病情评估的指标。

Objective To compare perlpapillary choroidal thickness （CT） between primary open angle glaucoma（POAG） and normal subjects using enhanced depth imaging optical coherence tomography （EDI-OCT）. Methods This research was a cross-sectional study. A 360-degree 3.4 mm diameter peripapillary circle scan was performed for retinal nerve fiber layer （RNFL） assessment by EDI-OCT in one eye of 40 POAG patients and 41 age- and sex-matched normal subjects. The observer used the manual segmentation function to delineate the posterior edge of the retinal pigment epithelium and the sclerochoroidal interface. The RNFL thickness algorithm function was used to automatically generate the CT in correspondingsectors. The intra- and inter-observer reproducibility was evaluated by intraclass correlation coefficient （ICC）. Repeated measures ANOVA and LSD-t test were used to analyze differences in CT by peripapillary location. A Pearson correlation was calculated for variation in CT relative to age and RNFL. Spearman rank correlation was calculated to test the strength of the relationship between global mean RNFL and visual field mean deviation （MD） and between CT and MD. Independent-samples t-test was conducted to compare mean RNFL and CT between POAG and normal subjects. Results The intra- and interobserver ICC values ranged from 0. 993 to 0. 999 and 0. 993 to 0. 998 respectively. In both groups, the CT showed a negative correlation with age at each peripapillary location （ r = - 0. 509 to - 0. 608, - 0. 549 to - 0. 668 ; P = 0. 000 to 0. 001 ） but no significant correlation with RNFL for any location （r = -0. 186 to 0. 273, -0. 090 to 0. 132; P = 0. 092 to 0. 869）. The peripapillary CT was thickest superotemporally and superonasally, then nasally and temporally, and was thinnest inferotemporally and inferonasally （P =0. 000 to 0. 036）. Compared to normal subjects, eyes with POAG had significantly thinner RNFL at all locations （ t = 6. 975 to 16. 087, P = 0. 000）, but CT measurements did not differ between groups for any location （ t = - 0. 474 to 0. 367, P = 0. 637 to 01978）. In POAG patients, the global mean RNFL （rs = - 0. 511, P = 0. 001 ） but not CT （rs = -0. 285, P = 0. 079 ） was found to significantly correlate with visual field MD. Conclusions The peripapillary CT in POAG neither differ from that in normal subjects nor correlate with RNFL and visual field index, which does not support using it as a clinical parameter in glaucoma diagnosis or management.