高频超声与声脉冲辐射力成像诊断2型糖尿病胫神经损害的价值对比

Comparison of High-Frequency Ultrasound and Acoustic Radiation Force Impulse Imaging in Diagnosis of Tibial Nerve Damage in Type 2 Diabetes Mellitus

目的对比分析高频超声与声脉冲辐射力成像(acoustic radiation force impulse imaging,ARFI)应用于诊断2型糖尿病周围神经病变中胫神经损害的临床价值。方法选取2022年6月—2023年6月就诊于青岛大学附属威海市中心医院的168例2型糖尿病患者及随机选取的60例无糖尿病病史、无周围神经病变病史和临床症状的健康志愿者作为研究对象。从168例2型糖尿病患者中选取60例神经电生理检查(neuroelectrophysiological examination,NEE)结果阳性并伴有周围神经病变临床体征者作为糖尿病周围神经病(diabetic peripheral neuropathy,DPN)组;选取60例2型糖尿病患者,要求NEE结果阴性并且无周围神经病变临床体征,作为无糖尿病周围神经病(non-diabetic peripheral neuropathy,NDPN)组;排除48例不符合入组条件2型糖尿病患者。对照组选取同期60例健康志愿者。选取内踝上方3~5 cm处胫神经行高频二维灰阶成像与ARFI,二维测量胫神经前后径、左右径、横截面积(cross-sectional area,CSA),运用声触诊组织成像定量技术(virtual touch tissue imaging and quantification,VTIQ)测量剪切波速度(shear wave velocity,SWV),对比观察各项数据应用于诊断DPN的诊断价值。绘制受试者工作特征曲线(receiver operating characteristic curve,ROC curve),计算曲线下面积,对比常规超声与ARFI应用于定量诊断2型糖尿病周围神经病的价值。结果DPN组、NDPN组、对照组胫神经左右径[左侧:(4.53±0.59)mm、(4.16±0.37)mm、(3.90±0.34)mm;右侧:(4.45±0.58)mm、(4.19±0.41)mm、(3.99±0.37)mm]和CSA[左侧:(22.27±3.90)mm^(2)、(19.00±3.18)mm^(2)、(17.37±3.48)mm^(2);右侧:(21.80±3.27)mm^(2)、(19.10±3.34)mm^(2)、(17.57±3.71)mm^(2)]比较,差异有统计学意义(P<0.05)。DPN组胫神经SWV[左侧:(2.47±0.27)m/s;右侧:(2.48±0.27)m/s]高于NDPN组[左侧:(2.23±0.11)m/s;右侧:(2.25±0.10)m/s]及对照组[左侧:(1.99±0.07)m/s;右侧(1.99±0.06)m/s],NDPN组高于对照组,差异有统计学意义(P<0.05)。根据测量结果分别绘制ROC曲线,胫神经前后径敏感度为0.567,特异度为0.617,曲线下面积为0.583(P=0.133);胫神经左右径敏感度为0.583,特异度为0.733,曲线下面积为0.671(P=0.024);胫神经CSA敏感度为0.717,特异度为0.683,曲线下面积为0.731(P<0.001);ARFI胫神经SWV敏感度为0.700,特异度为0.733,曲线下面积为0.783(P<0.001)。结论高频二维灰阶成像及ARFI对2型糖尿病DPN均具有诊断价值,ARFI意义更大。

Objective To compare and analyze the clinical value of high-frequency ultrasound and acoustic radiation force impulse imaging(ARFI)in the diagnosis of tibial nerve damage in type 2 diabetic peripheral neuropathy.Methods A total of 168 patients with type 2 diabetes who were treated in Weihai Central Hospital Affiliated to Qingdao University from June 2022 to June 2023,and 60 healthy volunteers who had no history of diabetes,peripheral neuropathy and clinical symptoms were randomly selected as the research objects.Neuro-electrophysiological examination(NEE)was performed in 168 patients with type 2 diabetes,and 60 patients with positive NEE results and clinical signs of peripheral neuropathy were selected as diabetic peripheral neuropathy(DPN)group.And 60 patients with type 2 diabetes were selected,who were required to have negative NEE results and no clinical signs of peripheral neuropathy,and served as non-diabetic peripheral neuropathy(NDPN)group.And 48 patients with type 2 diabetes who did not meet the conditions for inclusion were excluded.The control group selected 60 healthy volunteers in the same period.The tibial nerve 3-5 cm above the medial malleolus was selected for high-frequency two-dimensional gray-scale imaging and ARFI,and the anteroposterior diameter,left-right diameter and cross-sectional area(CSA)of the tibial nerve were measured in two dimensions.The shear wave velocity(SWV)was measured by virtual touch tissue imaging and quantification(VTIQ),and the diagnostic value of each data in diagnosing DPN was compared.Receiver operating characteristic curve(ROC curve)was drawn,and the area under the curve was calculated,and the value of conventional ultrasound and ARFI in quantitative diagnosis of type 2 diabetic peripheral neuropathy was compared.Results Among DPN group,NDPN group and control group,the left and right diameter of tibial nerve[left side:(4.53±0.59)mm,(4.16±0.37)mm,(3.90±0.34)mm;right side:(4.45±0.58)mm,(4.19±0.41)mm,(3.99±0.37)mm],CSA[right side:(22.27±3.90)mm^(2),(19.00±3.18)mm^(2),(17.37±3.48)mm^(2);right side:(21.80±3.27)mm^(2),(19.10±3.34)mm^(2),(17.57±3.71)mm^(2)]were compared,and the differences were statistically significant(P<0.05).Tibial nerve SWV[left side:(2.47±0.27)m/s;right side:(2.48±0.27)m/s]in DPN group was higher than that in NDPN group[left side:(2.23±0.11)m/s;right side:(2.25±0.10)m/s]and control group[left side:(1.99±0.07)m/s;right side:(1.99±0.06)m/s],NDPN group was higher than control group,the differences were statistically significant(P<0.05).According to the measured results,ROC curves were plotted respectively.The sensitivity of anterior and posterior diameter of tibial nerve was 0.567,the specificity was 0.617,and the area under the curve was 0.583(P=0.133).The sensitivity of tibial nerve was 0.583,the specificity was 0.733,and the area under the curve was 0.671(P=0.024).The CSA sensitivity of tibial nerve was 0.717,the specificity was 0.683,and the area under the curve was 0.731(P<0.001).ARFI tibial nerve SWV sensitivity was 0.700,specificity 0.733,and area under the curve 0.783(P<0.001).Conclusion Both high-frequency twodimensional gray-scale imaging and ARFI imaging have diagnostic value for DPN in type 2 diabetes mellitus,and ARFI imaging is more significant.