甲状腺滤泡癌与滤泡腺瘤的超声特征分析
Sonographic Characteristics and Differentiation of Follicular Thyroid Carcinoma and Follicular Adenoma

目的：探讨甲状腺滤泡癌(Follicular thyroid carcinoma, FTC)与滤泡腺瘤(Follicular adenoma, FA)的超声特征并筛选FTC的独立危险因素。方法：回顾性分析2018年1月至2020年12月于我院行甲状腺切除术并经病理证实为FTC (87例)和FA (209例)患者的临床及声像图资料，比较FTC与FA超声特征方面的差异，采用logistic回归分析筛选FTC的独立危险因素并构建多因素诊断模型，绘制ROC曲线并比较曲线下面积(AUC)，比较各超声特征及多因素诊断模型诊断FTC的灵敏度、特异度、阳性预测值(PPV)、阴性预测值(NPV)及准确率。结果：FTC与FA在性别(χ² = 0.311)、年龄(t = 1.618)及最大径(t = 1.457)方面差异无统计学意义(P > 0.05)。在超声特征方面，FTC与FA在回声、形态、边界、囊性变、声晕、钙化及血流方面差异存在统计学意义(χ² = 11.459, 61.218, 3.922, 23.169, 78.191, 30.605, 56.582, P < 0.05)；而在位置、回声性质、数量方面差异无统计学意义(χ² = 4.824, 3.277, 0.143, P > 0.05)。多因素logistic回归分析表明形态不规则、无囊性变、无细晕、粗钙化与血流丰富是FTC的独立危险因素(OR = 83.093, 0.373, 0.076, 6.098, 7.162, P < 0.05)，多因素诊断模型ROC曲线下面积为0.913，对FTC的整体诊断效能优于各超声特征。结论：利用超声特征可对FTC与FA进行鉴别诊断，以超声特征建立的多因素诊断模型对于FTC有较好的诊断效能及临床应用价值。
Objective: To explore the sonographic characteristics of follicular thyroid carcinoma (FTC) and follicular adenoma (FA) and identify the independent risk factors of FTC. Methods: The clinical and sonographic characteristics of 87 patients with FTC and 209 patients with FA from January 2018 to December 2020 were retrospectively analyzed. All of them underwent thyroidectomy in our hospital and were confirmed by pathology. The sonographic characteristics were analyzed. Multivariate logistic regression analysis was used to identify the independent risk factors of FTC, then build a multifactor diagnostic model. ROC curve was drawn and the area under the curve (AUC) was compared. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and accuracy of PTC were compared. Results: There were no significant differences in gender (χ2 = 0.311), age (t = 1.618) and largest diameter (t = 1.457) between the two groups (P > 0.05). In terms of sonographic characteristics, there were significant differences in echo, shape, margin, cystic changes, halo, calcification and vascularity (χ² = 11.459, 61.218, 3.922, 23.169, 78.191, 30.605, 56.582, P < 0.05). However, there were no significant differences in location, echo nature and number of nodules (χ² = 4.824, 3.277, 0.143, P > 0.05). Multivariate logistic regression analysis showed that irregular shape, absence of cystic changes and thin halo, coarse calcification and abundant vascularity were independent risk factors for FTC (OR = 83.093, 0.373, 0.076, 6.098, 7.162, P < 0.05). The area under the ROC curve of the multivariate diagnostic model was 0.913. The overall diagnostic efficiency of FTC was better than each sonographic characteristic. Conclusion: