河北医科大学学报 ›› 2021, Vol. 42 ›› Issue (9): 1032-1036.doi: 10.3969/j.issn.1007-3205.2021.09.008

• • 上一篇    下一篇

2型糖尿病胫神经病变超声特征的ROC曲线分析

  

  1. 1.华北医疗健康集团峰峰总医院超声科，河北 邯郸 056200；2.河北省优抚医院功能科，河北 石家庄 050051；
    3.河北省优抚医院放射科，河北 石家庄 050051
  • 出版日期:2021-09-25 发布日期:2021-09-28
  • 作者简介:于冬（1982-），男，河北张家口人，华北医疗健康集团峰峰总医院副主任医师，医学学士，从事医学超声诊断研究。
  • 基金资助:
    邯郸市科技研发计划（19422083051ZC）

The ROC curve analysis of the ultrasound characteristics of tibial neuropathy in patients with type 2 diabetes mellitus

  1. 1.Department of Ultrasound, Fengfeng General Hospital of North China Health-Care Industry Group， 
    Hebei Province, Handan 056200, China; 2.Department of Function, the Special Care Hospital of 
    Hebei Province, Shijiazhuang 050051, China; 3.Department of Radiology, the Special 
    Care Hospital of Hebei Province, Shijiazhuang 050051, China
  • Online:2021-09-25 Published:2021-09-28

HTML

PDF (PC)

286

摘要： 目的 比较2型糖尿病胫神经病变超声形态学各项指标，寻找2型糖尿病胫神经病变的高诊断价值指标。
方法 回顾性分析进行下肢血管及胫神经超声检查的2型糖尿病患者60例，以性别、年龄、病程等因素选择纳入标准及排除标准选择病例，以密歇根神经筛查量表（Michigan Neuropathy Screening Instrument,MNSI）体格检查2分作为诊断截断值分为两组（有神经病变组和无神经病变组）。比较两组胫神经宽度、厚度、横截面积、“筛孔样”结构以及是否有能量多普勒信号。
结果 两组胫神经宽度、“筛孔样”结构是否清晰显示及能量多普勒信号是否显示差异无统计学意义（P＞0.05）。胫神经的厚度、横截面积差异有统计学意义（P＜0.05）。胫神经厚度（AUC=0.826）的敏感度84.21%，特异度73.81%，阳性预测值69.60%，阴性预测值82.40%；横截面积(AUC=0.921)的敏感度94.74%，特异度85.71%，阳性预测值81.80%，阴性预测值94.40%；胫神经厚度与横截面积ROC的AUC差异有统计学意义（P=0.047）。
结论 2型糖尿病胫神经病变的超声各特征中胫神经厚度诊断价值中等，选择cut-off值为3.62 mm可得到较为理想的诊断效能；横截面积诊断价值较高，选择cut-off值为21.0 mm2可得到较为理想的诊断效能。


关键词: 糖尿病, 2型, 糖尿病神经病变, 超声检查

Abstract: Objective To compare the ultrasonic morphological indexes of tibial neuropathy in patients with type 2 diabetes mellitus(T2DM), so as to seek the indexes with high diagnostic value in tibial neuropathy in patients with T2DM. 
Methods Sixty patients with T2DM who underwent lower extremity vascular ultrasound examination and tibial nerve ultrasound examination were reviewed. The patients were included or excluded based on various factors such as gender, age, course of disease, and so on. Subjects were divided into two groups(DPN group and non-DNP group) with 2 points as the diagnostic cut-off value by the Michigan Neuropathy Screening Instrument(MNSI) table. Two groups of data were compared, including width, thickness, cross-sectional area, mesh-like structure, and presence or absence of Doppler energy signal. 
Results There were no statistically significant differences in tibial nerve width, presence of clear mesh-like structure and presence of Doppler energy signal between two groups(P＞0.05). There were statistically significant differences in tibial nerve thickness and cross-sectional area between two groups(P＜0.05). The sensitivity and specificity of tibial nerve thickness(AUC=0.826) were 84.21% and 73.81% respectively, and positive and negative predictive values were 69.60% and 82.40% respectively. The sensitivity and specificity of cross-sectional area(AUC=0.921) were 94.74% and 85.71% respectively, and positive and negative predictive values were 81.80% and 94.40% respectively. The difference in ROC AUC between tibial nerve thickness and cross-sectional area was statistically significant(P=0.047). 
Conclusion The diagnostic value of tibial nerve thickness is medium in all ultrasonic features of tibial neuropathy in patients with T2DM, and when the cut-off value is 3.62 mm, the ideal efficacy can be obtained. The diagnostic value of cross-sectional area is relatively high, and when the cut-off value is 21.0 mm2, the ideal efficacy can be obtained.


Key words: diabetes mellitus, type 2, diabetic neuropathies, ultrasonography

版权所有 © 《河北医科大学学报》编辑部
本系统由北京玛格泰克科技发展有限公司设计开发