医学
内科学
背景(考古学)
甲状腺癌
多元分析
接收机工作特性
回顾性队列研究
甲状腺
烧蚀
甲状腺切除术
肿瘤科
胃肠病学
古生物学
生物
作者
Tian Tian,Yangmengyuan Xu,Xinyue Zhang,Bin Liu
标识
DOI:10.1210/clinem/dgab445
摘要
Abstract Context The risk of persistent and recurrent disease in patients with differentiated thyroid cancer (DTC) is a continuum that ranges from very low to very high, even within the 3 primary risk categories. It is important to identify independent clinicopathological parameters to accurately predict clinical outcomes. Objective To examine the association between pre-ablation stimulated thyroglobulin (ps-Tg) and persistent and recurrent disease in DTC patients and investigate whether incorporation of ps-Tg could provide a more individualized estimate of clinical outcomes. Design, Setting, and Participants Medical records of 2524 DTC patients who underwent total thyroidectomy and radioiodine ablation between 2006 and 2018 were retrospectively reviewed. Main Outcome Measure Ps-Tg was measured under thyroid hormone withdrawal before remnant ablation. Association of ps-Tg and clinical outcomes. Results In multivariate analysis, age, American Thyroid Association (ATA) risk stratification, distant metastasis, ps-Tg, and cumulative administered activities were the independent predictive factors for persistent/recurrent disease. Receiver operating characteristic analysis identified ps-Tg cutoff (≤10.1 ng/mL) to predict disease-free status with a negative predictive value of 95%, and validated for all ATA categories. Integration of ps-Tg into ATA risk categories indicated that the presence of ps-Tg ≤ 10.1 ng/mL was associated with a significantly decreased chance of having persistent/recurrent disease in intermediate- and high-risk patients (9.9% to 4.1% in intermediate-risk patients, and 33.1% to 8.5% in high-risk patients). Conclusion The ps-Tg (≤10.1 ng/mL) was a key predictor of clinical outcomes in DTC patients. Its incorporation as a variable in the ATA risk stratification system could more accurately predict clinical outcomes.
科研通智能强力驱动
Strongly Powered by AbleSci AI