文献速递胸部CT

肺气肿的视觉与定量评估:瑞典心肺生物影像研究(SCAPIS)预试验中1111名参与者的病例对照研究

Visual and Quantitative Evaluation of Emphysema: A Case-Control Study of 1111 Participants in the Pilot Swedish CArdioPulmonary BioImage Study (SCAPIS).
PubMed 文献 · Acad Radiol · 2019-07-18

本研究基于SCAPIS预试验低剂量多排CT数据,纳入100例肺气肿受试者与100例匹配对照,评估两名胸科放射科医生视觉判读轻中度肺气肿的观察者间与观察者内一致性,并用商用软件定量测量LAV950、平均肺密度和全肺体积;结果显示视觉判读一致性良好(Krippendorff's α≥0.8),但定量分析无法可靠识别轻度肺气肿受试者。

要点

  • 低剂量多排CT视觉判读轻度肺气肿的观察者间与观察者内一致性良好,Krippendorff's α≥0.8。
  • 定量 densitometric 测量未能区分肺气肿与非肺气肿受试者,平均LAV950分别为8.3%与8.4%。
  • 定量分析无法可靠识别轻度肺气肿,限制了自动评估的应用。

核心结论

  • 低剂量多排CT视觉检测轻度肺气肿的观察者间与观察者内一致性良好。
  • 定量分析无法可靠识别轻度肺气肿受试者,妨碍了自动评估的使用。

研究设计与样本

  • 病例对照研究,纳入SCAPIS预试验电子病例报告表中报告的肺气肿受试者100例(男56例、女44例)及按性别、年龄、身高和体重匹配的无肺气肿对照100例。
  • 观察者间变异由两名胸科放射科医生对随机化检查进行评估;观察者内变异由三名放射科医生对各自原先评估的随机化检查重新评估。
  • 采用Krippendorff's α进行统计学评价;并用商用软件定量评估LAV950、平均肺密度和全肺体积。

关键影像表现/征象

  • 肺气肿视觉评分为轻度。
  • 肺气肿存在与否的观察者间和观察者内一致性Krippendorff's α均≥0.8。
  • 上叶肺气肿存在与否及范围的观察者间和观察者内一致性Krippendorff's α接近0.8。
  • 肺气肿组与非肺气肿组平均LAV950无差异,分别为8.3%与8.4%。

诊断标准与关键数值

  • 视觉判读一致性以Krippendorff's α评价,肺气肿存在与否的观察者间和观察者内α均≥0.8,上叶肺气肿存在及范围的α接近0.8。
  • 定量指标为%肺衰减值-950 HU(LAV950)、平均肺密度和全肺体积。

临床应用与阅片要点

  • 原文结论提示低剂量多排CT视觉判读可用于轻度肺气肿的一致性检测。
  • 原文结论提示定量分析无法可靠识别轻度肺气肿,妨碍自动评估的使用。

局限性

  • 原文未报告局限性。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

英文原文(摘要)· 点击展开

RATIONALE AND OBJECTIVES: Emphysema is a hallmark of chronic obstructive pulmonary disease. The primary aim of this study was to investigate inter- and intraobserver agreement of visual assessment of mild emphysema in low-dose multidetector computed tomography of subjects in the pilot SCAPIS in order to certify consistent detection of mild emphysema. The secondary aim was to investigate the performance of quantitative densitometric measurements in the cohort.

MATERIALS AND METHODS: Participants with emphysema (n = 100, 56 males and 44 females) reported in the electronic case report form of pilot SCAPIS and 100 matched controls (gender, age, height, and weight) without emphysema were included. To assess interobserver variability the randomized examinations were evaluated by two thoracic radiologists. For intraobserver variability three radiologists re-evaluated randomized examinations which they originally evaluated. The results were evaluated statistically by Krippendorff's α. The dataset was also assessed quantitively for % lung attenuation value -950 HU (LAV950), mean lung density and total lung volume by commercially available software.

RESULTS: Emphysema was visually scored as mild and Krippendorff's α was ≥0.8 for both the inter- and intraobserver agreement regarding presence of emphysema and approaching 0.8 regarding presence and extent of emphysema by location in the upper lobes. Mean LAV950 was not different between the emphysematous and the nonemphysematous participants; 8.3% and 8.4%, respectively.

CONCLUSION: The inter- and intraobserver agreement for visual detection of mild emphysema in low-dose multidetector computed tomography was good. Surprisingly, quantitative analysis could not reliably identify participants with mild emphysema, which hampers the use of automatic evaluation.

来源:Acad Radiol,查看原文。版权归原作者所有。

引用

  1. Vikgren J, Khalil M, Cederlund K, et al. Visual and Quantitative Evaluation of Emphysema: A Case-Control Study of 1111 Participants in the Pilot Swedish CArdioPulmonary BioImage Study (SCAPIS). Acad Radiol. 2019. PMID: 31326310. doi:10.1016/j.acra.2019.06.019

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。