Furuya-Kanamori Luis, Bell Katy J L, Clark Justin, Glasziou Paul, Doi Suhail A R
Luis Furuya-Kanamori and Suhail A.R. Doi, Research School of Population Health, Australian National University; Suhail A.R. Doi, Canberra Hospital, Canberra, Australian Capital Territory; Katy J.L. Bell, Justin Clark, and Paul Glasziou, Centre for Research in Evidence Based Practice, Bond University, Gold Coast; Suhail A.R. Doi, School of Agricultural, Computational and Environmental Sciences, University of Southern Queensland, Toowoomba, Queensland; Katy J.L. Bell, School of Public Health, University of Sydney; Suhail A.R. Doi, Prince of Wales Hospital, Sydney, New South Wales, Australia; and Suhail A.R. Doi, College of Medicine, Qatar University, Doha, Qatar.
J Clin Oncol. 2016 Oct 20;34(30):3672-3679. doi: 10.1200/JCO.2016.67.7419.
Differentiated thyroid cancer (DTC) incidence has been reported to have increased three- to 15-fold in the past few decades. It is unclear whether this represents overdiagnosis or a true increase in incidence. Therefore, the current study aimed to estimate the prevalence of incidental DTC in published autopsy series and determine whether this prevalence has been increasing over time.
PubMed, Embase, and Web of Science were searched from inception to December 2015 for relevant studies. Two authors searched for all autopsy studies that had included patients with no known history of thyroid pathology and reported the prevalence of incidental DTC (iDTC). Two authors independently extracted the data, and discrepancies were resolved by another author. The pooled prevalence of iDTC was assessed using a fixed-effects meta-analysis model with robust error variance. The time effect was studied using an inverse-variance weighted logit-linear regression model with robust error variance and a time variable.
Thirty-five studies, conducted between 1949 and 2007, met the inclusion criteria and contributed 42 data sets and 12,834 autopsies. The prevalence of iDTC among the partial and whole examination subgroups was 4.1% (95% CI, 3.0% to 5.4%) and 11.2% (95% CI, 6.7% to 16.1%), respectively. Once the intensiveness of thyroid examination was accounted for in the regression model, the prevalence odds ratio stabilized from 1970 onward, and no time effect was observed.
The current study confirms that iDTC is common, but the observed increasing incidence is not mirrored by prevalence within autopsy studies and, therefore, is unlikely to reflect a true population-level increase in tumorigenesis. This strongly suggests that the current increasing incidence of iDTC most likely reflects diagnostic detection increasing over time.
据报道，在过去几十年中，分化型甲状腺癌（DTC）的发病率增加了3至15倍。目前尚不清楚这是代表过度诊断还是发病率的真正增加。因此，本研究旨在估计已发表尸检系列中偶然发现的DTC的患病率，并确定该患病率是否随时间增加。
检索PubMed、Embase和Web of Science数据库，从建库至2015年12月，查找相关研究。两位作者检索了所有包含无甲状腺疾病病史患者的尸检研究，并报告了偶然发现的DTC（iDTC）的患病率。两位作者独立提取数据，分歧由另一位作者解决。使用具有稳健误差方差的固定效应荟萃分析模型评估iDTC的合并患病率。使用具有稳健误差方差和时间变量的逆方差加权logit线性回归模型研究时间效应。
1949年至2007年间进行的35项研究符合纳入标准，提供了42个数据集和12,834例尸检。部分检查亚组和全部检查亚组中iDTC的患病率分别为4.1%（95%CI，3.0%至5.4%）和11.2%（95%CI，6.7%至16.1%）。一旦在回归模型中考虑了甲状腺检查的强度，患病率比值比从1970年起趋于稳定，未观察到时间效应。
本研究证实iDTC很常见，但尸检研究中的患病率并未反映出观察到的发病率增加，因此不太可能反映肿瘤发生在人群水平上的真正增加。这强烈表明，目前iDTC发病率的增加很可能反映了随着时间推移诊断检测的增加。
