Endocrinology Unit, Azienda Ospedaliero-Universitaria Careggi, Florence, Italy.
J Endocrinol Invest. 2012 Sep;35(8):720-4. doi: 10.3275/7978. Epub 2011 Oct 3.
Cytological examination of fine needle aspirates (FNA) is the standard procedure for discriminating potentially malignant thyroid nodules to be referred to surgery. In a fraction of cases, ultrasound (US) examination could provide information theoretically sufficient to avoid FNA, when typical US features suggesting malignancies are lacking.
The aim of this study was to construct a simple US score predicting malignant nodules so as to reduce the number of unnecessary FNA.
In a series of 1632 consecutive patients undergoing US-guided FNA (1812 nodules), echostructure, echogenicity, margins, halo, microcalcification, and vascularization were assessed.
At multivariate analysis, the following parameters showed a strong predictive value for positive cytology (Thy 4 and Thy 5, suspicious and diagnostic for malignancy, respectively, according to the Thyroid British Association): solid echostructure, irregular margins and hypoechogenicity [adjusted odd ratio (OR) 5.13 (1.58-16.66), 3.03 (1.70-5.39), 2.05 (1.17-3.57), respectively]. A 10-point Thyroid Risk Ultrasound Score (TRUS) was constructed on the basis of the adjusted OR. A TRUS≥6 identified malignant nodules with sensitivity and specificity of 73% and 65%, respectively. Among the patients with follicular lesions (Thy 3) and final diagnosis of carcinoma, about 65% had a TRUS≥6.0.
The sensitivity of TRUS, although higher than that of other scores, could still be insufficient for the identification of patients who could avoid FNA in routine clinical practice, whereas its predictive value for Thy 3 lesions deserves further investigations.
细针抽吸细胞学检查（FNA）是鉴别可能恶性甲状腺结节以进行手术的标准程序。在少数情况下，当缺乏提示恶性的典型超声特征时，超声（US）检查可以提供理论上足够的信息来避免 FNA。
本研究旨在构建一种简单的 US 评分预测恶性结节，以减少不必要的 FNA 数量。
在连续 1632 例接受 US 引导下 FNA（1812 个结节）的患者系列中，评估了回声结构、回声强度、边缘、晕环、微钙化和血管化。
在多变量分析中，以下参数对细胞学阳性具有很强的预测价值（根据甲状腺英国协会的 Thy 4 和 Thy 5，可疑和诊断为恶性）：实性回声结构、不规则边缘和低回声 [校正优势比（OR）分别为 5.13（1.58-16.66）、3.03（1.70-5.39）、2.05（1.17-3.57）]。基于校正 OR，构建了 10 分甲状腺风险超声评分（TRUS）。TRUS≥6 可识别恶性结节，其敏感性和特异性分别为 73%和 65%。在滤泡性病变（Thy 3）和最终诊断为癌的患者中，约 65%的患者 TRUS≥6.0。
TRUS 的敏感性虽然高于其他评分，但在常规临床实践中仍不足以识别可以避免 FNA 的患者，而其对 Thy 3 病变的预测价值值得进一步研究。
