Departments of Thoracic Surgery.
Department of Medicine and Surgery, Section of Radiology, University of Parma, Parma, Italy.
J Thorac Imaging. 2023 Jul 1;38(4):W52-W63. doi: 10.1097/RTI.0000000000000698. Epub 2023 Jan 20.
To assess automated coronary artery calcium (CAC) and quantitative emphysema (percentage of low attenuation areas [%LAA]) for predicting mortality and lung cancer (LC) incidence in LC screening. To explore correlations between %LAA, CAC, and forced expiratory value in 1 second (FEV 1 ) and the discriminative ability of %LAA for airflow obstruction.
Baseline low-dose computed tomography scans of the BioMILD trial were analyzed using an artificial intelligence software. Univariate and multivariate analyses were performed to estimate the predictive value of %LAA and CAC. Harrell C -statistic and time-dependent area under the curve (AUC) were reported for 3 nested models (Model survey : age, sex, pack-years; Model survey-LDCT : Model survey plus %LAA plus CAC; Model final : Model survey-LDCT plus selected confounders). The correlations between %LAA, CAC, and FEV 1 and the discriminative ability of %LAA for airflow obstruction were tested using the Pearson correlation coefficient and AUC-receiver operating characteristic curve, respectively.
A total of 4098 volunteers were enrolled. %LAA and CAC independently predicted 6-year all-cause (Model final hazard ratio [HR], 1.14 per %LAA interquartile range [IQR] increase [95% CI, 1.05-1.23], 2.13 for CAC ≥400 [95% CI, 1.36-3.28]), noncancer (Model final HR, 1.25 per %LAA IQR increase [95% CI, 1.11-1.37], 3.22 for CAC ≥400 [95%CI, 1.62-6.39]), and cardiovascular (Model final HR, 1.25 per %LAA IQR increase [95% CI, 1.00-1.46], 4.66 for CAC ≥400, [95% CI, 1.80-12.58]) mortality, with an increase in concordance probability in Model survey-LDCT compared with Model survey ( P <0.05). No significant association with LC incidence was found after adjustments. Both biomarkers negatively correlated with FEV 1 ( P <0.01). %LAA identified airflow obstruction with a moderate discriminative ability (AUC, 0.738).
Automated CAC and %LAA added prognostic information to age, sex, and pack-years for predicting mortality but not LC incidence in an LC screening setting. Both biomarkers negatively correlated with FEV 1 , with %LAA enabling the identification of airflow obstruction with moderate discriminative ability.
评估自动化冠状动脉钙（CAC）和定量肺气肿（低衰减区百分比 [%LAA]）在肺癌（LC）筛查中对死亡率和肺癌（LC）发病率的预测作用。探讨 %LAA、CAC 与 1 秒用力呼气量（FEV 1 ）之间的相关性，以及 %LAA 对气流阻塞的鉴别能力。
使用人工智能软件对 BioMILD 试验的基线低剂量计算机断层扫描进行分析。进行单变量和多变量分析以估计 %LAA 和 CAC 的预测价值。报告了 3 个嵌套模型（模型调查：年龄、性别、吸烟指数；模型调查-LDCT：模型调查加%LAA 加 CAC；模型最终：模型调查-LDCT 加选定混杂因素）的 Harrell C-统计量和时间依赖性曲线下面积（AUC）。使用 Pearson 相关系数和 AUC-受试者工作特征曲线分别测试 %LAA、CAC 与 FEV 1 之间的相关性以及 %LAA 对气流阻塞的鉴别能力。
共纳入 4098 名志愿者。%LAA 和 CAC 独立预测了 6 年全因（模型最终风险比[HR]，每增加 1%LAA 四分位间距[IQR]增加[95%CI，1.05-1.23]，CAC≥400[95%CI，1.36-3.28]）、非癌症（模型最终 HR，每增加 1%LAA IQR 增加[95%CI，1.11-1.37]，CAC≥400[95%CI，1.62-6.39]）和心血管（模型最终 HR，每增加 1%LAA IQR 增加[95%CI，1.00-1.46]，CAC≥400[95%CI，1.80-12.58]）死亡率，与模型调查相比，模型调查-LDCT 的一致性概率增加（P<0.05）。调整后与 LC 发病率无显著相关性。两种生物标志物均与 FEV 1 呈负相关（P<0.01）。%LAA 对气流阻塞有中等鉴别能力（AUC，0.738）。
在 LC 筛查环境中，自动化 CAC 和 %LAA 增加了年龄、性别和吸烟指数的预后信息，可预测死亡率，但不能预测 LC 发病率。两种生物标志物均与 FEV 1 呈负相关，%LAA 对气流阻塞有中等鉴别能力。
