Department of Ultrasound, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, China.
Front Endocrinol (Lausanne). 2024 Mar 20;15:1350123. doi: 10.3389/fendo.2024.1350123. eCollection 2024.
There is currently insufficient data to validate adult-based US risk stratification systems (RSSs) for the identification of malignant thyroid nodules in a pediatric population.
From October 2016 and May 2023, 173 thyroid nodules of pediatric patients (age ≤ 18 years) with definitive pathology results and ultrasound (US) examination within 1 month before surgery or fine-needle aspiration (FNA) biopsy in our institution were enrolled in this study. The clinical and US characteristics of these nodules were retrospectively reviewed and categorized according to the ACR-TIRADS, C-TIRADS, and ATA guidelines. The diagnostic performance of US-based FNA criteria (original and simulating) of the three guidelines in thyroid cancer detection was estimated.
The three RSSs had similar AUC according to the categories(0.849-0.852, all > 0.05). When combined with the original FNA criteria of the three RSSs to manage the nodules, the FNA rate of ACR-TIRADS and C-TIRADS were significantly less than ATA guidelines (53.18% vs. 64.63%, < 0.05, and 52.60% vs. 64.63%, < 0.05). The missed malignancy rate (MMR) and unnecessary FNA rate (UFR) of ATA guidelines (50.00%, 35.85%) was highest among the three RSSs, followed by the C-TIRADS (37.80%, 19.57%) and the ACR-TIRADS (37.04%, 19.57%). When nodules < 1 cm with the highest category in each RSS biopsied, that is when using the simulating FNA thresholds, the MMR was reduced overall (all < 0.001), without a change in the UFR (all > 0.05). All the three RSSs showed a substantial improvement in accuracy and malignant detection rate (all < 0.05).
The ACR-TIRADS, C-TIRADS, and ATA guidelines showed high missed malignancy rates when using their original recommended FNA criteria. When nodules < 1 cm with the highest category in each RSS biopsied, the missed malignancy rate of each RSS was decreased. Decreasing the FNA thresholds for highly suspicious malignant nodules may therefore be an effective means of managing malignant thyroid nodules in pediatric patients.
目前,尚无足够的数据来验证适用于儿科人群的基于美国成年人的风险分层系统(RSS),以识别恶性甲状腺结节。
本研究回顾性分析了 2016 年 10 月至 2023 年 5 月期间在我院接受手术或细针抽吸(FNA)活检前 1 个月内有明确病理结果和超声(US)检查的 173 例儿科患者(≤ 18 岁)的甲状腺结节的临床和 US 特征。这些结节根据 ACR-TIRADS、C-TIRADS 和 ATA 指南进行分类。评估了三种指南的基于 US 的 FNA 标准(原始和模拟)在甲状腺癌检测中的诊断性能。
根据分类,三种 RSS 的 AUC 相似(0.849-0.852,均>0.05)。当结合三种 RSS 的原始 FNA 标准来管理结节时,ACR-TIRADS 和 C-TIRADS 的 FNA 率明显低于 ATA 指南(53.18% vs. 64.63%,<0.05,52.60% vs. 64.63%,<0.05)。三种 RSS 中,ATA 指南的恶性漏诊率(MMR)和不必要的 FNA 率(UFR)最高(50.00%,35.85%),其次是 C-TIRADS(37.80%,19.57%)和 ACR-TIRADS(37.04%,19.57%)。当每个 RSS 中最高分类的结节<1cm 进行活检时,即使用模拟 FNA 阈值,整体 MMR 降低(均<0.001),UFR 无变化(均>0.05)。三种 RSS 的准确性和恶性检出率均显著提高(均<0.05)。
当使用其原始推荐的 FNA 标准时,ACR-TIRADS、C-TIRADS 和 ATA 指南的恶性漏诊率较高。当每个 RSS 中最高分类的结节<1cm 进行活检时,每个 RSS 的恶性漏诊率降低。降低高度疑似恶性结节的 FNA 阈值可能是管理儿科患者恶性甲状腺结节的有效手段。