摘要
目的:探讨直肠肿瘤的多模态超声参数术前评估与术后病理结果的差异.方法:选取2020年1月-2022年4月在我院治疗的直肠占位病变患者100例,术前接受经直肠超声(ERUS)、剪切波弹性成像(SWE)和超声造影(CEUS)检查,分析直肠恶性病灶(79例)与良性病灶(21例)的ERUS、SWE和CEUS参数差异,同时用ERUS、SWE和CEUS参数构建Logistic回归方程预测直肠恶性病灶的价值.结果:恶性病灶最大径、血流峰值流速和阻力指数高于良性病灶[(4.20±1.00)cm、(25.12±3.81)cm/s和(0.80±0.14)vs(3.18±0.98)cm、(16.60±4.11)cm/s和(0.66±0.12)],差异有统计学意义(P＜0.05);恶性病灶内部回声不均匀、血流信号丰富、形态不规则和边缘模糊高于良性病灶[72.15％、75.95％、64.56％和72.15％vs 33.33％、14.29％、23.81％和47.62％],差异有统计学意义(P＜0.05).恶性病灶平均杨氏模量(Emean)、最大杨氏模量(Emax)和最小杨氏模量(Emin)高于良性病灶[(81.19±15.53)kPa、(110.20±23.32)kPa 和(61.18±15.02)kPa vs(51.14±11.40)kPa、(70.05±21.05)kPa 和(36.60±11.68)kPa],差异有统计学意义(P＜0.05).恶性病灶上升时间(RT)、达峰时间(TTP)和平均渡越时间(mTT)短于良性病灶[(6.22±1.34)s、(8.78±1.82)s 和(14.65±2.34)s vs(9.10±1.32)s、(12.21±2.01)s 和(18.82±2.51)s],差异有统计学意义(P＜0.05),造影剂呈"快进快出"比例高于良性病灶[70.89％vs 28.57％],差异有统计学意义(P＜0.05)a EURS、SWE、CEUS参数构建Logistic回归方程,该方程预测直肠恶性病灶的ROC曲线下面积为0.892(95％CI:0.828～0.956),灵敏度和特异度分别为83.50％和76.20％.结论:构建多模态超声参数在鉴别直肠恶性病灶方面有较好的应用价值,值得进一步研究.
Abstract
Objective To explore the differences between preoperative evaluation of multimodal ultrasound parameters and postoperative pathological outcomes of rectal tumors.Methods 100 patients with rectal space occupying lesions who were treated in our hospital from January 2020 to April 2022 were selected,and received transrectal ultrasound(ERUS),shear wave elastography(SWE),and contrast-enhanced ultrasound(CEUS).The difference of ERUS,SWE,and CEUS parameters between malignant and benign rectal lesions were analyzed.At the same time,the value of ERUS,SWE,and CEUS parameters to construct a Logistic regression equation to predict rectal malignant lesions were analyzed.Results The maximum diameter,peak blood flow velocity,and resistance index of malignant lesions were higher than those of benign lesions(4.20±1.00)cm,(25.12±3.81)cm/s and(0.80±0.14)vs(3.18±0.98)cm,(16.60±4.11)cm/s and(0.66±0.12),respectively),and the difference was statistically significant(P＜0.05).Malignant lesions have uneven internal echoes,rich blood flow signals,irregular morphology,and blurred edges,which are higher than benign lesions[proportions of 72.15％,75.95％,64.56％and 72.15％vs 33.33％,14.29％,23.81％and 47.62％,respectively],and the differences were statistically significant and significantly higher than benign lesions(P＜0.05).The average Young's modulus(Emean),maximum Young's modulus(Emax),and minimum Young's modulus(Emin)of malignant lesions were higher than those of benign lesions[(81.19±15.53)kPa,(110.20±23.32)kPa and(61.18±15.02)kPa vs(51.14±11.40)kPa,(70.05±21.05)kPa and(36.60±1 1.68)kPa,respectively)],with statistically significant differences compared to benign lesions(P＜0.05).The rise time(RT),peak time(TTP),and mean transition time(mTT)of malignant lesions were shorter than those of benign lesions[(6.22±1.34)s,(8.78±1.82)s and(14.65±2.34)s vs(9.10±1.32)s,(12.21±2.01)s and(18.82±2.51)s,respectively)],and the differences were statistically significant and significantly shorter than benign lesions(P＜0.05).The contrast agent showed a"fast in and fast out"pattern compared to benign lesions[(70.89％vs 28.57％)],The difference was statistically significant higher than that of benign lesions(P＜0.05).The parameters of EURS,SWE and CEUS were used to construct a Logistic regression equation,which predicted the area under the ROC curve of rectal malignant lesions to be 0.892(95％CI:0.828～0.956),and the sensitivity and specificity were 83.50％and 76.20％respectively.Conclusion The construction of multimodal ultrasound parameters has good application value in the differential prediction of rectal malignant lesions,and it is worthy of further study.
基金项目
河北省廊坊市科学技术研究与发展计划(2022013121)
维普
万方数据