Department of Surgery, MacKay Memorial Hospital and Mackay Medical College, Taipei, Taiwan; Department of Pharmacology, School of Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan.
Department of Surgery, MacKay Memorial Hospital and Mackay Medical College, Taipei, Taiwan.
Eur J Surg Oncol. 2020 Oct;46(10 Pt A):1814-1819. doi: 10.1016/j.ejso.2020.06.044. Epub 2020 Jul 20.
Lymphovascular invasion (LVI) is associated with disease recurrence and compromised survival in patients with thyroid cancer. Nonetheless, LVI is not identifiable on preoperative ultrasound or cytologic assessment. We aimed to explore the clinicopathological features associated with LVI.
We conducted a retrospective review of our prospectively maintained database from 2009 to 2018. Multivariate analyses were performed to determine the associations between clinicopathological parameters and LVI. Generalized additive models were used to examine the nonlinear relationship between continuous variables and LVI.
A total of 795 patients were included in the analysis, and 174 (22%) had LVI. Patients' age (odds ratio [OR] = 0.982), tumor size (OR = 1.466), clinical lymphadenopathy (OR = 6.975), and advanced extrathyroidal extension (OR = 2.938) were independently associated with LVI. In the subset analysis of 198 patients with available genetic information, tumor size (OR = 1.599), clinical lymph node metastasis (OR = 3.657), and TERT promoter mutation (OR = 4.726) were predictive of LVI. Among 573 patients who had no clinical lymphadenopathy or advanced extrathyroidal extension, tumor size was the only predictor of LVI. Tumor size >1.5 cm had an increased risk of LVI based on the generalized additive model plot and receiver operating characteristic curve analysis.
Tumor size is positively associated with the risk of LVI in papillary thyroid cancer. To avoid delayed treatment in patients with LVI, a tumor size of 1.5 cm may be considered as the safe upper limit for active surveillance.
淋巴管浸润(LVI)与甲状腺癌患者的疾病复发和生存预后受损有关。然而,术前超声或细胞学评估无法识别 LVI。我们旨在探讨与 LVI 相关的临床病理特征。
我们对 2009 年至 2018 年前瞻性维护的数据库进行了回顾性分析。采用多变量分析确定临床病理参数与 LVI 之间的关联。广义加性模型用于检查连续变量与 LVI 之间的非线性关系。
共纳入 795 例患者,其中 174 例(22%)有 LVI。患者年龄(比值比[OR] = 0.982)、肿瘤大小(OR = 1.466)、临床淋巴结转移(OR = 6.975)和晚期甲状腺外侵犯(OR = 2.938)与 LVI 独立相关。在 198 例有可用遗传信息的患者亚组分析中,肿瘤大小(OR = 1.599)、临床淋巴结转移(OR = 3.657)和 TERT 启动子突变(OR = 4.726)预测 LVI。在 573 例无临床淋巴结转移或晚期甲状腺外侵犯的患者中,肿瘤大小是 LVI 的唯一预测因素。基于广义加性模型图和受试者工作特征曲线分析,肿瘤大小>1.5cm 时 LVI 的风险增加。
肿瘤大小与甲状腺乳头状癌 LVI 的风险呈正相关。为避免 LVI 患者治疗延误,肿瘤大小 1.5cm 可作为主动监测的安全上限。