欢迎来到《四川大学学报(医学版)》

糖尿病视网膜病变微血管改变定量评估及其与血视网膜屏障破坏的关联研究

Quantitative Assessment of Microvascular Changes in Diabetic Retinopathy and Their Association with Blood-Retinal Barrier Impairment

  • 摘要:
    目的 基于光学相干断层扫描血管成像(optical coherence tomography angiography, OCTA)技术定量评估糖尿病视网膜病变(diabetes retinopathy, DR)患者的视网膜微血管改变,并探讨其与血视网膜屏障(blood-retinal barrier, BRB)破坏的关联。
    方法 纳入本院收治的208例2型糖尿病合并DR患者,行OCTA检查获取视网膜微血管参数;同步检测血清血管内皮生长因子(vascular endothelial growth factor, VEGF)和细胞间黏附因子1(intercellular adhesion molecule 1, ICAM-1)水平。分析OCTA参数与血清标志物的相关性,采用Logistic回归分析危险因素,并通过受试者工作特征曲线(receiver operating characteristic curve, ROC)曲线评估诊断效能。
    结果 208例DR患者浅层毛细血管密度(superficial capillary density, SCP-D)和深层毛细血管密度(deep capillary density, DCP-D)分别为42.67%±4.35%、47.89%±5.02%;中心凹无血管区(foveal avascular zone, FAZ)面积、周长、圆形指数均值分别为(0.38±0.10)mm2、(2.04±0.28)mm、0.72±0.08;无灌注区面积均值(1.87±0.45)mm2。其中121例(58.17%)SCP-D异常(<45%),114例(56.25%)DCP-D异常(<50%),88例(42.31%)FAZ面积异常,77例(37.02%)FAZ周长异常,69例(33.17%)FAZ圆形指数异常;142例(68.27%)无灌注区面积异常。FAZ面积与VEGF(r=0.559, 95% CI: 0.457~0.661)、ICAM-1(r=0.411, 95% CI: 0.289~0.533)呈正相关;FAZ圆形指数、SCP-D、DCP-D与VEGF及ICAM-1均呈负相关(P<0.05);无灌注区面积与二者呈正相关。Logistic回归显示,糖尿病病程(OR=1.159, 95% CI: 1.060~1.267)及VEGF(OR=1.013, 95% CI: 1.005~1.022)是发生严重视网膜微血管改变的独立危险因素(P<0.05)。4个OCTA评估指标中无灌注区面积的预测价值最高,AUC为0.879(95% CI:0.820~0.938)。
    结论 DR患者的OCTA评估指标水平与BRB相关标志物密切相关,无灌注区面积对于DR患者发生严重视网膜微血管改变的预测价值最高。

     

    Abstract:
    Objective To quantitatively evaluate retinal microvascular changes in patients with diabetic retinopathy (DR) using optical coherence tomography angiography (OCTA), and to explore their association with blood-retinal barrier (BRB) disruption.
    Methods A total of 208 patients with type 2 diabetes and DR underwent OCTA to obtain microvascular parameters. Serum vascular endothelial growth factor (VEGF) and intercellular adhesion molecule 1 (ICAM-1) levels were measured. Correlations were analyzed, risk factors were identified using logistic regression, and diagnostic efficacy was evaluated with ROC curves.
    Results The superficial capillary density (SCP-D) and deep capillary density (DCP-D) of the 208 DR patients were (42.67 ± 4.35)% and (47.89 ± 5.02)%, respectively. The mean values for the area, perimeter, and circularity index of the foveal avascular zone (FAZ) were (0.38 ± 0.10) mm², (2.04 ± 0.28) mm, and 0.72 ± 0.08, respectively. The mean area of the non-perfusion zone was (1.87 ± 0.45) mm². Among these patients, 121 (58.17%) cases had abnormal SCP-D (< 45%), 114 (56.25%) cases had abnormal DCP-D (< 50%), 88 (42.31%) cases had abnormal FAZ area, 77 (37.02%) cases had abnormal FAZ perimeter, 69 (33.17%) cases had abnormal FAZ circularity index, and 142 (68.27%) cases had abnormal non-perfusion zone area. The FAZ area was positively correlated with VEGF (r = 0.559, 95% CI: 0.457-0.661) and ICAM-1 (r = 0.411, 95% CI: 0.289–0.533). The FAZ circularity index, SCP-D, and DCP-D were negatively correlated with VEGF and ICAM-1 (P < 0.05). The area of the non-perfusion zone was positively correlated with both. Logistic regression showed that the duration of diabetes (OR = 1.159, 95% CI: 1.060-1.267) and VEGF (OR = 1.013, 95% CI: 1.005-1.022) were independent risk factors for severe retinal microvascular changes (P < 0.05). Among the four OCTA assessment indicators, the area of the non-perfusion zone had the highest predictive value, with an AUC of 0.879 (95% CI: 0.820–0.938).
    Conclusion The OCTA assessment indicators in DR patients are closely related to BRB-related markers. The area of the non-perfusion zone has the highest predictive value for severe retinal microvascular changes in DR patients.

     

/

返回文章
返回