Objective: To investigate the consequences of the thickness of ganglion cell layer (GCL) and visual field defect of non-functional pituitary adenoma with chiasm compression. Methods: A case control study. The study included 40 (80 eyes) non-functional pituitary adenoma patients in Peking Union Medical College Hospital from March 2015 to February 2017. Twenty patients (no visual field defect group, 40 eyes) of them were detected to be chiasm compressed or touched by the adenoma with no visual field defect detected, and the other 20 patients (visual field defect group, 40 eyes) were the sex-and-age matched pituitary adenoma patients with bitemporal heminopsia. This study also included 20 (control group, 40 eyes) sex-and-age matched healthy controls. The para-papillary retinal nerve fiber layer (RNFL) thickness in 6 quadrants including nasal, temporal, nasal superior, temporal superior, nasal inferior and temporal inferior as well as the macular GCL thickness and ganglion cell-inner plexiform layer (GCIPL) thickness in 4 quadrants including nasal superior, nasal inferior, temporal superior and temporal inferior were measured. The non-parametric test was used to compare the RNFL, GCL and GCIPL thickness among the three groups. Results: The mean age among the three groups was (46±10) years and the difference among the three groups was not significant (P=0.88). The sex ratio of the three groups was 9∶11 (male∶female) and the difference among the three groups was not significant. The mean axial length among the three groups was (23.22±0.90) mm and the difference among the three groups was not significant (P=0.51). The thickness of para-papillary RNFL of temporal superior, temporal, nasal superior, nasal, nasal inferior quadrants and whole circumference was significantly thinner in the visual field defect group than the control group [(129.88±28.64) μm, (63.63±26.84) μm, (88.08±32.16) μm, (50.68±19.99) μm, (92.48±25.06) μm, and (85.00±20.65) μm vs. (141.10±18.95) μm, (79.12±16.78) μm, (113.68±21.28) μm, (69.67±14.23) μm, (117.80±31.32) μm, and (102.80±9.68) μm, t=2.26, 3.06, 4.14, 4.84, 4.25, 4.88, all P<0.05]. In the nasal quadrant, the para-papillary RNFL of the no visual field defect group was significantly thinner compared with the control group [(61.45±9.83) μm vs. (69.67±14.23) μm, t=2.97, P<0.05]. The total GCL thickness was (30.48±5.42) μm in the visual field defect group, (31.35±2.77) μm in the no visual field defect group, thinner than that in the control group [(33.32±2.92) μm, t=2.92, 3.62; both P<0.05]. The total GCIPL thickness showed no significant difference among the three groups (P=0.07). In the superior and inferior temporal quadrants, the GCL and GCIPL thickness showed no significant difference among the three groups (all P>0.05). In the superior and inferior nasal quadrants, the GCL thickness was (29.41±5.97) μm, and (28.47±5.13) μm in the visual field defect group, (31.15±3.27) μm and (30.61±2.96) μm in the no visual field defect group, and (34.23±3.16) μm and (32.97±2.78) μm in the control group. The GCL thickness in the nasal quadrant was thinner in the visual field defect group (t=4.45, 4.82)and the no visual field defect group(t=4.23, 3.63) than in the control group (all P<0.01). However, no significant difference in GCL thickness was detected between the visual field defect group and the no visual field defect group (both P>0.05). In the superior and inferior nasal quadrants, the GCIPL thickness was (54.06±10.50) μm and (51.77±9.18) μm in the visual field defect group, (58.03±4.00) μm and (56.23±5.37) μm in the no visual field defect group, and (62.26±7.11) μm and (59.39±6.64) μm in the control group. The GCIPL thickness was thinner in the nasal quadrant in the visual field defect group than in the control group (t=3.95, 4.20, both P<0.01). Only in the Superior nasal quadrant, the GCIPL was significantly thinner in the no visual field defect group than the control group (t=3.25, P<0.01). Conclusion: The optic GCL may get thinner in pituitary nonfunctional adenoma with chiasm compression patients without the RNFL layer thinning and visual field defect. (Chin J Ophthalmol, 2019, 55: 186-194).
目的： 研究伴有视交叉压迫的垂体无功能腺瘤患者视网膜神经节细胞层（GCL）厚度变化与视野缺损发生的相关性。 方法： 病例对照研究。纳入2015年3月至2017年2月就诊于北京协和医院的垂体无功能腺瘤患者40例（80只眼），其中包括垂体无功能腺瘤导致视交叉受压但视野无明显异常者（视野正常组）20例（40只眼）以及与其性别、年龄匹配的典型垂体无功能腺瘤压迫视交叉导致双颞侧偏盲患者（颞侧偏盲组）20例（40只眼）；同时纳入性别、年龄匹配的健康对照（健康对照组）20名（40只眼）。采用频域相干光层析成像术测量各组视盘旁鼻侧、颞侧、鼻上、颞上、鼻下、颞下6个象限视网膜神经纤维层（RNFL）厚度，黄斑区鼻上、鼻下、颞上、颞下4个象限GCL及神经节细胞-内丛状层（GCIPL）厚度。采用非参数检验对3组受试者RNFL、GCL及GCIPL厚度进行比较。 结果： 3组受试者年龄（46±10）岁，男女比例为9∶11，眼轴长度为（23.22±0.90）mm，3组间差异均无统计学意义（均P>0.05）。颞侧偏盲组视盘旁颞上、颞侧、鼻上、鼻侧、鼻下5个象限及全周RNFL厚度均低于健康对照组[（129.88±28.64）μm、（63.63±26.84）μm、（88.08±32.16）μm、（50.68±19.99）μm、（92.48±25.06）μm、（85.00±20.65）μm分别与（141.10±18.95）μm、（79.12±16.78）μm、（113.68±21.28）μm、（69.67±14.23）μm、（117.80±31.32）μm、（102.80±9.68）μm比较]，差异均有统计学意义（Dunnet t=2.26、3.06、4.14、4.84、4.25、4.88，均P<0.05），视野正常组仅鼻侧象限RNFL厚度低于健康对照组，差异有统计学意义[（61.45±9.83）μm与（69.67±14.23）μm比较，Dunnet t=2.97，P<0.05]。颞侧偏盲组及视野正常组患者黄斑区GCL总体厚度分别为（30.48±5.42）μm、（31.35±2.77）μm，均低于健康对照组[（33.32±2.92）μm，Dunnet t=2.92、3.62，均P<0.05]；而3组间黄斑区GCIPL总体厚度差异无统计学意义（P=0.07）。在后极部颞侧（颞上及颞下象限）3组间GCL厚度及GCIPL厚度差异均无统计学意义（均P>0.05）；在后极部鼻侧（鼻上及鼻下）象限，颞侧偏盲组GCL厚度分别为（29.41±5.97）μm、（28.47±5.13）μm，视野正常组GCL厚度分别为（31.15±3.27）μm、（30.61±2.96）μm，健康对照组GCL厚度分别为（34.23±3.16）μm、（32.97±2.78）μm，颞侧偏盲组（t=4.45、4.82）及视野正常组（t=4.23、3.63）患者鼻侧2个象限GCL层厚度均低于健康对照组（均P<0.01）；但视野正常组与颞侧偏盲组差异无统计学意义（均P>0.05）；在后极部鼻侧（鼻上及鼻下）象限，颞侧偏盲组GCIPL厚度分别为（54.06±10.50）μm、（51.77±9.18）μm，视野正常组GCIPL厚度分别为（58.03±4.00）μm、（56.23±5.37）μm，健康对照组GCIPL厚度分别为（62.26±7.11）μm、（59.39±6.64）μm，颞侧偏盲组鼻侧2个象限GCIPL厚度均低于健康对照组（t=3.95、4.20，均P<0.01），视野正常组仅鼻上象限GCIPL厚度低于健康对照组（t=3.25，P<0.01）。 结论： 垂体无功能腺瘤患者发生视交叉压迫但未造成视野缺损时，尚未出现视盘旁RNFL变薄，但可能已经出现黄斑鼻侧GCL变薄。（中华眼科杂志，2019，55：186-194）.
Keywords: Adenoma; Hemianopsia; Nerve fibers; Pituitary neoplasms; Retinal ganglion cells.