增强深部成像相干光断层扫描术观察原发性开角型青光眼视乳头筛板的形态学变化

Analysis of morphologic changes of lamina cribrosa in primary open angle glaucoma using enhanced depth imaging optical coherence tomography

目的评价增强深部成像的相干光断层扫描（EDI-OCT）测量POAG患者视乳头筛板厚度、观察其形态学变化的临床价值。方法病例对照研究。选取2013年8月至12月在北京协和医院眼科确诊的POAG患者52例（52只眼）的临床资料，同期眼科门诊体检健康者50例（50只眼）作为对照组。使用EDI-OCT对其视乳头进行连续水平和垂直方向扫描，分析视乳头筛板局部缺损的数目、部位及其与视网膜神经纤维层（RNFL）缺损部位的关系。选取经过视乳头中心及其与视乳头上下缘连线中点的3个水平截面，利用ImageJ软件测量各截面中的筛板厚度；信度分析采用组内相关系数（ICC）。POAG患者与健康人视乳头筛板厚度比较，采用成组设计t检验；视乳头各部位筛板厚度比较，采用重复测量的方差分析和LSD-t检验；视乳头筛板平均厚度与视乳头全周RNFL平均厚度的相关性分析，采用Pearson相关分析法；筛板平均厚度与视野平均缺损（非正态分布）的相关性分析，采用Spearman秩相关分析法。结果52只POAG患眼中，有20只眼（38．5％）可见26个视乳头筛板局部缺损，其中13个局部缺损（50．0％）位于颞下，6个缺损（23．1％）位于下方，4个（15．4％）位于颞上，3个（11．5％）位于上方；25个缺损（96．2％）同时存在相应象限RNFL缺损。而健康眼未见筛板局部缺损。测量视乳头筛板厚度的信度分析，检测者内ICC值为0．90-0．95，检测者间ICC值为0．84．0．90。POAG患者视乳头中央及上、下方中周部筛板厚度分别为（196．7±24．0）μm、（180．7±26．3）μm和（184．1±23．5）μm；健康人分别为（225．2±22．7）μm、（203．2±21．0）μm和（202．6±16．5）μm；两组视乳头中央筛板厚度均厚于上、下方中周部（P〈0．01）。POAG患者视乳头筛板平均厚度为（187．2±19．4）μm，健康对照者为（210．3±17．4）μm，患者较对照者明显变薄（t=5．06，P〈0．01）。POAG患者其余各部位筛板厚度也均明显薄于健康人（t=3．52．4．96，P〈0．01）。POAG患者视乳头全周RNFL平均厚度为（65．7±13．9）μm，筛板平均厚度与其之间呈正相关（r=0．45，P〈0．01），而与视野平均缺损之间呈负相关（rs=-0．58，P〈0．01）。结论EDI-OCT可显示POAG患者的视乳头筛板厚度较健康人变薄，且与POAG病情存在相关性；部分患者筛板存在局部缺损。EDI-OCT是评估POAG视乳头筛板活体形态学变化的一种有效手段。

Objective To analyze the morphologic changes of the lamina cribrosa （LC） in primary open angle glaucoma （POAG） patients using enhanced depth imaging optical coherence tomography （EDI- OCT）. Methods This was a case-control study. Serial horizontal and vertical B-scans of the optic nerve head were obtained from one eye of 52 POAG patients and 50 age- and sex-matched healthy subjects by EDI- OCT. The number and location of focal LC defects as well as their relationship with retinal nerve fiber layer （RNFL） defects were investigated in glaucomatous eyes and compared with healthy controls. The LC thickness was measured at the center of mid-superior, central, and mid-inferior horizontal B-scans. The intra- and interobserver reproducibility were evaluated by intraclass correlation coefficient （ICC）. A repeated measure ANOVA was used to analyze regional differences of the LC thickness. Independent t-test was conducted to compare LC thickness between POAG and healthy controls. A Pearson correlation was calculated for variation in LC thickness relative to global mean RNFL thickness. Spearman rank correlation was calculated to test the strength of the relationship between LC thickness and visual field mean deviation （MD）. Results Twenty-six focal LC defects were found in 20 eyes （38.5%） with glaucoma versus none in the healthy controls. There were 13 （50.0%） focal LC defects in the inferotemporal sector, 6 （23.1%） defects in the inferior sector, 4 （15.4%） in the superotemporal sector, and 3 （11.5%） in the superior sector. Twenty- five of 26 focal LC defects （96.2%） corresponded to the RNFL defects. The intra- and interobserver ICCs of LC thickness measurement ranged from 0.90 to 0.95 and 0.84 to 0.90, respectively. The thicknesses of the superior midperipheral, central, and inferior midperipheral LC in POAG eyes were 180.7±26.3, 196.7±24.0, and 184.1 ±23.5 μm, respectively. In healthy controls, the laminar thicknesses were 203.2±21.0, 225.2±22.7, and 202.6±16.5 μm, respectively. In both groups, the midperipheral LC was significantly thinner than the central LC （P〈0.01）. The mean LC thickness was （187.2± 19.4）p.m in POAG eyes. It was significantly thinner than that in healthy controls （210.3±17.4 μm） （t=5.06, P〈0.01）. The superior midperipheral, central, and inferior midperipheral regions of the LC were also significantly thinner in POAG than in controls （t=3.52 to 4.96, P〈0.01）. In POAG patients, mean LC thickness was found to significantly correlate with both global mean RNFL thickness （r=0.45, P〈0.01） and visual field MD （r,=-0.58, P〈0.01）. Conclusions The LC thickness was thinner in POAG than in healthy eyes, and correlated with the severity of POAG. Focal LC defects occurred frequently in POAG eyes, and might be associated with RNFL defect. EDI-OCT is a useful tool for the evaluation of LC deformation in glaucoma.