目的:本研究旨在利用体内共聚焦显微镜(IVCM)研究2型糖尿病(DM)患者的角膜缘变化,并探讨其眼部表现与全身状态的相关性。
方法:纳入55例2型DM患者和20例年龄匹配的对照。两组之间比较以下IVCM参数:Vogt栅栏(POV),角膜上皮厚度(CET),基底细胞密度(BCD),基底下神经丛,和树突状细胞密度。所有受试者都接受了血液和尿液采样以进行实验室分析,包括空腹血糖,糖化血红蛋白,总胆固醇(TC),甘油三酯(TG),低密度脂蛋白(LDL)胆固醇,高密度脂蛋白(HDL)胆固醇,C反应蛋白,尿白蛋白与肌酐比值,尿白蛋白,还有尿肌酐.检测IVCM参数与血液生物标志物之间的相关性。采用受试者工作特征曲线选择DM患者角膜干细胞损伤危险因素的临界值。
结果:与对照组相比,DM患者POV显著降低(上区,P=0.033;下区域,P=0.003;鼻区,P<0.001;时间区域,P<0.001),中心CET(44.8±3.6μmvs.51.9±3.6μm,P<0.001),中央角膜BCD(7415.5±563.2细胞/mm2vs.9177.9±977.8个细胞/mm2,P<0.001),和周围角膜BCD(6181.3±416.5细胞/mm2vs.8576.3±933.2细胞/mm2,P<0.001)。树突状细胞密度(41.0±33.7细胞/mm2vs.DM组24.6±7.8个细胞/mm2,P=0.001)明显高于DM组。IVCM参数与血液生物标志物之间存在以下弱相关性:中央角膜BCD与DM持续时间呈负相关(r=-0.3,P=0.024)。TC(r=-0.36,P=0.007),LDL(r=-0.39,P=0.004)。POV与TC(r=-0.34,P=0.011)、LDL(r=-0.31,P=0.022)呈负相关。HDL的截止值为1.215mmol/L,TG为1.59mmol/L,或4.75mmol/L的TC,以区分高风险患者和低风险的干细胞损伤。
结论:2型DM患者典型POV阳性率较低,BCD降低,CET,和基底下神经密度。与干细胞表型最相关的指标是DM持续时间,TC,LDL。糖尿病患者的脂质状态可能是发生角膜缘干细胞缺乏症的风险预测因子。需要更大样本量的进一步研究或基础研究来验证结果。
OBJECTIVE: This study aimed to investigate corneal limbus changes in patients with type 2 diabetes mellitus (DM) using in vivo confocal microscopy (IVCM) and explore the correlation between their ocular manifestations and systemic status.
METHODS: Fifty-five patients with type 2 DM and 20 age-matched controls were included. The following IVCM parameters were compared between the 2 groups: palisades of Vogt (POV), corneal epithelial thickness (CET), basal cell density (BCD), subbasal nerve plexus, and dendritic cell density. All subjects underwent blood and urine sampling for laboratory analysis, including fasting blood glucose, glycated hemoglobin, total cholesterol (TC), triglycerides (TG), low-density lipoprotein (LDL) cholesterol, high-density lipoprotein (HDL) cholesterol, C-reactive protein, urinary albumin-to-creatinine ratio, urine albumin, and urine creatinine. The correlations between IVCM parameters and blood biomarkers were detected. Receiver operating characteristic curve was used for selecting the cutoff value of risk factors for corneal stem cell injury in patients with DM.
RESULTS: Compared with controls, patients with DM displayed a significant reduction of POV (superior region, P = 0.033; inferior region, P = 0.003; nasal region, P < 0.001; temporal region, P < 0.001), central CET (44.8 ± 3.6 μm vs. 51.9 ± 3.6 μm, P < 0.001), central corneal BCD (7415.5 ± 563.2 cells/mm 2 vs. 9177.9 ± 977.8 cells/mm 2 , P < 0.001), and peripheral corneal BCD (6181.3 ± 416.5 cells/mm 2 vs. 8576.3 ± 933.2 cells/mm 2 , P < 0.001). Dendritic cell density (41.0 ± 33.7 cells/mm 2 vs. 24.6 ± 7.8 cells/mm 2 , P = 0.001) was significantly higher in the DM group. The following weak correlations were shown between IVCM parameters and blood biomarkers: central corneal BCD was negatively correlated with DM duration (r = -0.3, P = 0.024), TC (r = -0.36, P = 0.007), and LDL (r = -0.39, P = 0.004). The presence of POV in the superior region was negatively correlated with TC (r = -0.34, P = 0.011) and LDL (r = -0.31, P = 0.022). Cutoff values of 1.215 mmol/L for HDL, 1.59 mmol/L for TG, or 4.75 mmol/L for TC were established to distinguish patients with a high risk from a low risk for stem cell damage.
CONCLUSIONS: Patients with type 2 DM displayed a lower positive rate of typical POV and a decrease in BCD, CET, and subbasal nerve density. The most relevant indicators for stem cell phenotypes were DM duration, TC, and LDL. Lipid status in diabetic patients could be a predictor of risk for developing corneal limbal stem cell deficiency. Further studies with larger sample sizes or basic research are needed to verify the results.