摘要
目的:探讨结缔组织病患者并发肺间质病变(interstitial lung disease,ILD)的危险因素,基于计算机断层扫描(computer tomography,CT)肺容积参数及血清学指标建立结缔组织病患者并发ILD的列线图模型。方法:选取2020年1月至2023年8月医院收治的235例结缔组织病患者作为模型集,选取2023年9月至2024年11月医院收治的100例结缔组织病患者作为验证集。收集CT参数(间质纤维化肺容积ILDV、全肺总容积WL、正常肺组织容积NL)及血清学指标(透明质酸HA、涎液化糖链抗原KL-6)。采用LASSO回归筛选预测因素,Logistic回归分析危险因素,构建列线图模型并进行内部与外部验证。结果:ILD发生率为33.19%(78/235)。Logistic回归显示,ILDV升高、WL降低、NL降低、HA升高、KL-6升高为独立危险因素(均P<0.05)。模型集与验证集的ROC曲线下面积分别为0.970(95%CI:0.952~0.988)和0.988(95%CI:0.974~1.000),最佳截断值敏感度、特异度达96.20%、89.20%。结论:本研究首次联合CT参数与血清学指标构建CTD-ILD列线图模型,经验证效能优异,可为早期筛查提供非侵入性工具。
Objective:To explore the risk factors for interstitial lung disease(ILD)in patients with connective tissue diseases and to establish a nomogram model for predicting ILD in patients with connective tissue diseases based on computer tomography(CT)lung volume parameters and serological indicators.Methods:A total of 235 patients with connective tissue disease admitted to the hospital from January 2020 to August 2023 were selected as the model set,and 100 patients with connective tissue disease admitted to the hospital from September 2023 to November 2024 were selected as the validation set.CT parameters(interstitial fibrosis lung volume,ILDV;whole lung volume,WL;normal lung tissue volume,NL)and serological parameters(hyaluronic acid,HA;krebs von den lungen-6,KL-6)were collected.LASSO regression was used to screen the predictors,Logistic regression was used to analyze the risk factors,and a nomogram model was constructed for internal and external validation.Results:The incidence of ILD was 33.19%(78/235).Logistic regression showed that increased ILDV,decreased WL,decreased NL,increased HA and increased KL-6 were independent risk factors(all P<0.05).The area under ROC curve of the model set and the validation set were 0.970(95%CI:0.952~0.988)and 0.988(95%CI:0.974~1.000),respectively,and the sensitivity/specificity of the best cut-off value was 96.20%/89.20%.Conclusion:In this study,for the first time,a CTD-ILD nomogram model was constructed by combining CT parameters and serological indexes,which proved excellent efficacy and could provide a non-invasive tool for early screening.
作者
袁鑫
刘连城
李玲
YUAN Xin;LIU Liancheng;LI Ling(Department of Medical Imaging,Taizhou Hospital of Traditional Chinese Medicine,Nanjing University of Chinese Medicine,Jiangsu 225300,China)
出处
《影像科学与光化学》
2026年第2期157-166,共10页
Imaging Science and Photochemistry
关键词
计算机断层扫描
血清学指标
结缔组织病
肺间质病变
列线图模型
computed tomography
serological indicators
connective tissue diseases
interstitial lung disease
nomogram model
