无视网膜病变2型糖尿病患者黄斑区血管密度及其影响因素

Analysis of vascular density and its risk factors in fundus imaging of patients with type 2 diabetes but without retinopathy

目的分析无视网膜病变2型糖尿病患者的黄斑区血管密度及其影响因素。方法入选140例无视网膜病变的2型糖尿病患者,应用计算机辅助软件,采用基于方向跟踪和像素卷积描述的黄斑血管分割法自动计数患者黄斑区的血管密度。根据眼底血管密度中位数分为低血管密度组和高血管密度组,比较两组患者间一般资料及生化指标的差异。另外入选了40例健康体检者为健康对照组,同时在140例患者中选择年龄、性别、身高、体质量、血压等指标均匹配的40例作为糖尿病组,比较两组患者的黄斑区血管密度水平。结果低血管密度组患者的年龄为(58.47±10.39)岁,收缩压为(135.23±12.67)mmHg(1 kPa=7.5 mmHg),均较高血管密度组(50.86±13.86)岁、(129.73±10.41)mmHg高,差异均有统计学意义(均为P<0.05);低血管密度组患者的体质量为(74.00±11.05)kg,红细胞计数为(4.48±0.49)×10^12个·L^-1,红细胞压积为40.12%±4.24%,血红蛋白含量为(135.31±16.34)g·L^-1,均较高血管密度组(78.48±15.16)kg、(4.66±0.49)×10^12个·L^-1、42.10%±4.00%、(142.71±15.59)g·L^-1减少,差异均有统计学意义(均为P<0.05)。Logistics回归分析结果显示,收缩压和红细胞计数是黄斑区血管密度的独立影响因素,且与血管密度均呈负相关(r=-0.077、-4.062,均为P<0.05)。与糖尿病组(0.07554±0.02000)相比,健康对照组眼底血管密度(0.10049±0.03000)明显较高,差异具有统计学意义(P<0.001)。结论利用计算机辅助软件可以自动快速识别黄斑区微循环障碍的患者。黄斑区血管密度与收缩压及红细胞计数有关。改善高血压及缺氧导致的红细胞计数减少可能有利于改善黄斑区血管密度,从而预防糖尿病视网膜病变。

Objective To analyze the fundus vascular density and its influencing factors in patients with type 2 diabetes but without retinopathy.Methods A total of 140 type 2 diabetes patients with non-diabetic retinopathy were enrolled.Computer-assisted software was used to automatically count the vascular density of the macular area with macular vascular segmentation based on direction tracking and pixel convolution in patients.Based on the median vascular density of the fundus,all patients were divided into low vascular density(LVD)group and high vascular density(HVD)group,and the differences of all general and biochemical indicators were compared between the two groups.In addition,40 healthy subjects were selected for control group.At the same time,40 diabetic patients with matching indexes such as age,gender,height,body mass and blood pressure were selected as diabetic group,and the density level of fundus vessels was compared between the two groups.Results Compared with HVD group,the age(50.86±13.86)years vs.(58.47±10.39)years and systolic blood pressure(SBP)(129.73±10.41)mmHg(1 kPa=7.5 mmHg)vs.(135.23±12.67)mmHg in LVD group were higher(both P<0.05).The body mass of the patients in the LVD group was(74.0±11.05)kg,the red blood cell count was(4.48±0.49)×10^12 cells·L^-1,and the hematocrit was 40.12%±4.24%,the hemoglobin content was(135.31±16.34)g·L^-1,all of which were higher than those in the HVD group(78.48±15.16)kg,(4.66±0.49)×10^12 cells·L^-1,42.10%±4.00%,(142.71±15.59)g·L^-1,and the difference was statistically significant(all P<0.05).Among the factors related to vascular density mentioned above,SBP and red blood cell count were independent risk factors for vascular density,and both were negatively correlated with fundus vascular density(r=-0.077,-4.062,both P<0.05).The fundus vascular density of the control group(0.10049±0.03000)was higher than that of the diabetic group(0.07554±0.02000)under the premise of balanced basic information between the two groups,and the difference was statistically significant(P<0.001).Conclusion Computer-assisted software is an easy,rapid and automate way to identify patients with microcirculatory disorder in the macula.The vascular density in the macula is associated with increased hypertension and red blood cell count.Improving red blood cell and hypertension may be beneficial in improving blood vessel density in the macular area and thus helpful in preventing diabetic retinopathy.