医学
淋巴结
倾向得分匹配
阶段(地层学)
子群分析
回顾性队列研究
生存分析
多元分析
内科学
外科
置信区间
生物
古生物学
作者
Song Xu,Xiongfei Li,Fan Ren,Jinling He,Shikang Zhao,Yanye Wang,Dian Ren,Shuai Zhu,Xi Lei,Gang Chen,Jun Chen
出处
期刊:Annals of Surgery
[Ovid Technologies (Wolters Kluwer)]
日期:2020-11-13
卷期号:276 (6): e991-e999
被引量:14
标识
DOI:10.1097/sla.0000000000004593
摘要
Objective: This study aimed to determine the optimal surgical procedure for early-stage pulmonary carcinoids (PCs). Background: PCs, comprising typical carcinoids (TCs) and atypical carcinoids (ACs), are rare low-grade malignant tumors. We determine the optimal surgical management for early-stage PCs using data from the Surveillance, Epidemiology, and End Results registry. Methods: Clinical and survival data of patients with early-stage PC tumors with a diameter ≤3 cm were retrieved. The Kaplan-Meier method and logrank tests were used to assess the differences in overall survival (OS). Subgroup analyses were also performed. To reduce the inherent bias of retrospective studies, two propensity score matching (PSM) analysis with (PSM2) or without (PSM1) consideration of lymph node assessment were performed. Results: In total, 2934 patients with PCs, including 2741 (93.42%) with TCs and 193 (6.58%) with ACs, were recruited. After PSM1 analysis, TC patients in the lobectomy group had a significantly better OS than those in the sublobar resection group ( P = 0.0067), which is more remarkable for patients with a tumor diameter of 2 cm <T ≤ 3 cm ( P = 0.0345) and those aged <70 years ( P = 0.0032). However, survival benefits were not found after PSM2 analysis which balanced lymph node assessment. In multivariate cox analysis, age <70 years, female, TC histology and adequate lymph node assessment were associated with better OS. Conclusions: Sublobar resection may not significantly compromise the longterm oncological outcomes in early-stage PCs ≤3 cm in size if lymph node assessment is performed adequately. Further validation in large randomized clinical trials is warranted.
科研通智能强力驱动
Strongly Powered by AbleSci AI