Haug Logan P, Dahiya Nirvikar, Young Scott W, Patel Maitray D
Department of Radiology, Mayo Clinic Arizona, Phoenix, AZ, USA.
J Ultrasound Med. 2023 Feb;42(2):409-415. doi: 10.1002/jum.16034. Epub 2022 Jun 7.
We evaluated the performance of ACR TI-RADS when points for lobulated margins are applied only when the margins meet a quantified measure of margin microlobulation and not applied when nodules only demonstrate macrolobulation.
We retrospectively reviewed ultrasound and pathology records (May 01, 2018 to July 31, 2020) to find all thyroid nodules at one institution characterized as having lobulated margins using the ACR TI-RADS lexicon and subsequently undergoing fine needle aspiration (FNA). Nodule margins were evaluated to note the presence or absence of microlobulation, quantitatively defined as a protrusion with a base <2.5 mm in length. The impact to detection of malignant nodules and avoidance of benign FNA when margin points for lobulation were added only when microlobulated was analyzed.
58 of 516 thyroid nodules undergoing US-guided FNA were classified as lobulated, comprising the study population. 21 (36.2%) had microlobulated margins, with 12 of the 21 (57.1%) being malignant. Comparatively, of the 37 nodules showing only macrolobulated margins without microlobulation, only 2 (5.4%) were malignant (P < .0001). For 53 nodules ≥10 mm, 15 (28.3%) benign nodules would not have met size criteria for FNA had points for margins not been applied when only showing macrolobulation, whereas all 10 malignant nodules would still have been sampled.
Adding two points to the ACR TI-RADS score for lobulated thyroid nodules should only apply when microlobulations are present.
我们评估了美国放射学会（ACR）甲状腺影像报告和数据系统（TI-RADS）的性能，即仅当边缘符合边缘微叶状的量化标准时才应用叶状边缘的评分，而当结节仅表现为大叶状时则不应用该评分。
我们回顾性分析了超声和病理记录（2018年5月1日至2020年7月31日），以找出在一家机构中使用ACR TI-RADS术语表被描述为具有叶状边缘并随后接受细针穿刺活检（FNA）的所有甲状腺结节。评估结节边缘以记录微叶状的存在与否，微叶状被定量定义为基部长度<2.5毫米的突出。分析了仅在存在微叶状时添加叶状边缘评分对恶性结节检测和避免良性FNA的影响。
516个接受超声引导下FNA的甲状腺结节中有58个被分类为叶状，构成研究人群。21个（36.2%）有微叶状边缘，其中21个中的12个（57.1%）为恶性。相比之下，在37个仅显示大叶状边缘而无微叶状的结节中，只有2个（5.4%）为恶性（P <.0001）。对于53个≥10毫米的结节，如果仅显示大叶状时不应用边缘评分，15个（28.3%）良性结节将不符合FNA的大小标准，而所有10个恶性结节仍将被取样。
对于叶状甲状腺结节，仅当存在微叶状时才应在ACR TI-RADS评分中增加两分。
