甲状腺良恶性结节的多模态超声研究

发布时间：2018-07-28 16:23

【摘要】：第一部分不同大小的甲状腺结节超声造影定性和定量分析目的:探讨不同大小的甲状腺结节的超声造影定性和定量分析价值。方法:178例患者共178个甲状腺结节术前行超声造影,所有结节均经手术病理证实(良性74个,恶性104个)。按结节最大径线分2组(≤1cm或1 cm),(1)定性评估结节造影的强化过程、程度、均匀性、有无环形增强、强化后的边界、形状和大小的超声造影特征。Fisher确切概率法比较良恶性结节造影特征的差异。ROC评价甲状腺超声造影的诊断价值。(2)定量分析绘制结节感兴趣区(ROI)和实质ROI的时间-强度曲线(TIC),记录各造影参数峰值强度(Peak)、达峰时间(TtoPk)及曲线下面积(Area)等,并计算结节ROI减去实质ROI的各超声造影参数△Peak、△Area等。t检验比较甲状腺良恶性结节造影参数的差异。结果:(1)CEUS示1cm和≤1cm组的甲状腺良恶性结节强化程度、过程、均匀性、强化后的形状差异均有统计学意义(P0.05),强化后的大小均无统计学意义(P0.001)。仅1cm甲状腺结节强化后的边界、强化完全、环形增强差异均有统计学意义(P0.05)。ROC示CEUS诊断1 cm和≤1 cm甲状腺良恶性结节最佳阈值均为3.5,曲线下面积分别为0.869(95%可信区间0.806～0.932)和0.864(95%可信区间0.717～1.000),1cm组CEUS的敏感性和特异性分别为75.8%、91.9%;≤1cm组CEUS的敏感性和特异性分别为89.5%、83.3%。(2)》1cm良性结节Peak、△Area、△Peak和△Grad大于恶性结节(P=0.001,P=0.012,P=0.001,P=0.004)。≤1cm 良性结节的 Area、Peak、△Area 和△Grad 大于恶性结节(P=0.001,P=0.003,P=0.003,P=0.008)。结论:对于不同大小的甲状腺结节,超声造影有助于其鉴别诊断。第二部分甲状腺癌的超声多模态诊断研究背景:高分辨率超声(Ultrasonography,US)是诊断甲状腺结节最敏感的方法,但良恶性鉴别较困难。超声新技术弹性超声(US elastography,USE)、超声造影(contrast-enhanced US,CEUS)目前也应用于临床。细针穿刺细胞学检查(fine needle aspiration cytology,FNAC)是术前诊断甲状腺良恶性结节价值最高的手段,但仍有20%～25%结节无法诊断。目的:探究US、USE、CEUS联合诊断,FNAC联合BRAF(V600E)基因突变诊断甲状腺癌的应用价值。方法:275例患者共320枚甲状腺结节同时行US、USE、CEUS和FNAC,经手术病理证实良性114例,恶性206例。其中有33例患者38枚甲状腺结节(良性4例,恶性34例)同时行穿刺标本BRAF基因检测。绘制US、USE、CEUS和联合评分ROC,计算并比较曲线下面积(the area under the ROC,AUC),选取合适截断点计算联合评分的诊断效能。并分析FNAC联合BRAF(V600E)基因突变对甲状腺癌的诊断效能。结果:联合评分的ROC的AUC(0.907,95%可信区间:0.871～0.942)均大于 TI-RADS 评分(0.763,95%可信区间:0.710～0.816)、弹性评分(0.745,95%可信区间:0.687～0.803)和造影评分(0.871,95%可信区间:0.828～0.913),前两者差异有统计学意义(Z=4.438、4.630、1.269;P0.01、P0.01、P0.05)。联合评分诊断敏感度、特异度和准确度为0.908、0.754和0.853。FNAC联合BRAF(V600E)基因突变诊断的敏感性、特异度和准确度均为1.000,高于FNAC(0.966、0.956 和 0.962)。结论:多模态超声影像能提高甲状腺癌的诊断效能,进一步筛选高危结节。FNAC是术前鉴别甲状腺良恶性结节的金标准,FNAC联合BRAF(V600E)基因检测可进一步提高准确性。
[Abstract]:The first part was qualitative and quantitative analysis of different sizes of thyroid nodules. Objective: to discuss the qualitative and quantitative analysis of different sizes of thyroid nodules. Methods: 178 patients with 178 thyroid nodules underwent ultrasound imaging before operation, all nodules were confirmed by hand pathology (74 benign, 104 malignant). The maximum diameter was divided into 2 groups (< < 1cm or 1 cm) and (1) qualitative assessment of the enhancement process of nodule contrast, degree, uniformity, non ring enhancement, enhanced boundary, shape and size of ultrasound contrast characteristics.Fisher accurate probability method compared the contrast between benign and malignant nodules by.ROC evaluation of the diagnostic value of thyroid sonography. (2) quantitative analysis The time intensity curve (TIC), the peak intensity (Peak), peak time (TtoPk) and the area under the curve (Area) were recorded in the nodular region of interest (ROI) and the parenchyma ROI, and the differences in the contrast parameters of the thyroid benign and malignant nodules were compared with the contrast-enhanced ultrasonography parameters of the nodular ROI minus the parenchyma ROI, Delta Area and other.T tests. Results: (1) C EUS showed that the degree of enhancement of thyroid benign and malignant nodules in 1cm and 1cm groups, process, uniformity, and the shape difference after intensification were statistically significant (P0.05), and the size after strengthening was not statistically significant (P0.001). Only 1cm thyroid nodules were strengthened, and the enhancement was complete, and the difference of ring shape was statistically significant (P0.05).ROC CEUS diagnosis 1 cm. The optimal threshold of thyroid benign and malignant nodules was 3.5, and the area under the curve was 0.869 (95% confidence interval 0.806 ~ 0.932) and 0.864 (95% confidence interval 0.717 to 1), and the sensitivity and specificity of CEUS in group 1cm were respectively 0.869, respectively, and the sensitivity and specificity of CEUS in group less 1cm were respectively 89.5%, 83.3%. (2) >1cm benign nodules Peak, and delta Area, respectively. Delta Peak and delta Grad are larger than malignant nodules (P=0.001, P=0.012, P=0.001, P=0.004). Peak, Delta Area and delta Grad are larger than malignant nodules (P=0.001, P=0.001, P=0.004). Conclusion: ultrasonography is helpful for differential diagnosis of thyroid nodules of different sizes. Ultrasound multimodality diagnosis in second parts of thyroid carcinoma Background: Ultrasonography (US) is the most sensitive method for the diagnosis of thyroid nodules, but it is difficult to differentiate between benign and malignant. US elastography (USE) and contrast-enhanced US (CEUS) are also used in clinical. Fine needle aspiration cytology (fine needle aspiration cytology) NAC) is the highest value of preoperative diagnosis of thyroid benign and malignant nodules, but there are still 20% to 25% nodules that cannot be diagnosed. Objective: To explore the value of US, USE, CEUS combined diagnosis, FNAC combined with BRAF (V600E) gene mutation in the diagnosis of thyroid cancer. Methods: 320 thyroid nodules in 275 patients underwent US, USE, CEUS and FNAC, confirmed by surgical pathology. There were 114 cases of benign and 206 malignant cases. Among them, 38 thyroid nodules (4 benign and 34 cases) were detected by BRAF gene. US, USE, CEUS and combined score ROC were plotted, and the area under the curve (the area under the ROC, AUC) was calculated and compared, and the joint scoring point was selected to calculate the diagnostic efficiency of the joint score. The efficacy of AF (V600E) gene mutation in the diagnosis of thyroid cancer. Results: the AUC (0.907,95% confidence interval: 0.871 ~ 0.942) of the combined score of ROC was greater than the TI-RADS score (0.763,95% confidence interval: 0.710 to 0.816), the elasticity score (0.745,95% confidence interval: 0.687 to 0.803) and the contrast score (0.871,95% confidence interval: 0.828 to 0.913), the difference between the former two Statistical significance (Z=4.438,4.630,1.269; P0.01, P0.01, P0.05). The sensitivity, specificity and accuracy of the combined score for diagnostic sensitivity, specificity and accuracy of 0.908,0.754 and 0.853.FNAC combined with BRAF (V600E) gene mutation were 1, higher than FNAC (0.966,0.956 and 0.962). Conclusion: multimodal ultrasound can improve thyroid cancer. The diagnostic efficiency and further screening of high risk nodules.FNAC are the gold criteria for preoperative identification of benign and malignant thyroid nodules. FNAC combined with BRAF (V600E) gene detection can further improve the accuracy.
【学位授予单位】：南京大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：R445.1;R581
，

本文编号：2150870