结缔组织病并发间质性肺疾病风险的列线图预测模型构建

Construction of a nomogram prediction model for the risk of connective tissue disease with interstitial lung disease

目的分析结缔组织病(CTD)并发间质性肺疾病(ILD)的风险,构建CTD患者并发ILD风险的预测模型,为此类患者的预防及早期干预提供依据。方法选取2018年1月至2024年1月在该院住院的208例患者作为研究对象,根据是否并发ILD将患者分为CTD组(82例)和CTD-ILD组(126例)。收集所有患者的人口学信息、实验室检查结果及其他相关数据。根据多因素Logistic回归分析结果构建预测CTD患者并发ILD风险的列线图模型,采用受试者工作特征(ROC)曲线分析列线图模型的预测效能。结果CTD-ILD组中性粒细胞计数、单核细胞计数、白细胞计数、白细胞介素-6水平、纤维蛋白原水平、铁蛋白水平、D-二聚体水平、中性粒细胞计数/(淋巴细胞计数×血小板计数)、AISI(AISI为中性粒细胞计数×血小板计数×单核细胞计数/淋巴细胞计数)、(白细胞计数-中性粒细胞计数)/淋巴细胞计数水平均高于CTD组,年龄大于CTD组,25羟维生素D、清蛋白水平及男性占比均低于CTD组,差异均有统计学意义(P<0.05)。多因素Logistic回归分析结果显示,年龄、D-二聚体、铁蛋白、AISI水平升高是CTD患者并发ILD的独立危险因素(P<0.05)。基于多因素Logistic回归分析的4个危险因素构建CTD患者并发ILD风险预测线性模型,ROC曲线下面积为0.831(95%CI:0.777~0.886),灵敏度为67.5%,特异度为89.0%,预测模型判别良好。标定曲线与理想曲线重叠,临床决策曲线显示该模型具有一定的临床应用价值。结论CTD患者并发ILD的独立影响因素为年龄、D-二聚体、铁蛋白、AISI,并且上述指标是CTD患者并发ILD的预测因子,上述指标所构建的预测模型可为预测CTD患者并发ILD的危险概率提供一种新的、方便的、快速的方法,可指导临床早期干预。

Objective To analyze the risk of connective tissue disease(CTD)complicated with interstitial lung disease(ILD),and construct a prediction model for the risk of ILD in CTD patients,so as to provide a basis for the prevention and early intervention of these patients.Methods A total of 208 patients hospitalized in this hospital from January 2018 to January 2024 were selected as the research objects.According to whether ILD occurred,the patients were divided into CTD group(82 cases)and CTD-ILD group(126 cases).Demographic information,laboratory test results and other relevant data of all patients were collected.According to the results of multivariate Logistic regression analysis,a nomogram model was constructed to predict the risk of ILD in CTD patients.The receiver operating characteristic(ROC)curve was used to analyze the predictive efficiency of the nomogram model.Results CTD-ILD group of neutrophil count,mononuclear cell count,white blood cell count,level of interleukin-6,fibrinogen level,ferritin level,D-dimer level,neutrophil count/(lymphocyte count×platelet count),AISI(AISI was neutrophil count×platelet count×monocyte count/lymphocyte count)and(white blood cell count-neutrophil count)/lymphocyte count in the CTD group were higher than those in the CTD group,the age were older than CTD group,while the levels of 25-hydroxyvitamin D,albumin and the proportion of males in the CTD group were lower than those in the CTD group,and the differences were statistically significant(P<0.05).Multivariate Logistic regression analysis showed that age,elevated levels of D-dimer,ferritin and AISI were independent risk factors for ILD in CTD patients(P<0.05).The area under the ROC curve was 0.831(95%CI:0.777-0.886),the sensitivity was 67.5%,and the specificity was 89.0%.The prediction model showed good discrimination.The calibration curve overlapped with the ideal curve,and the clinical decision curve showed that the model had certain clinical application value.Conclusion The independent influencing factors of ILD in patients with CTD are age,D-dimer,ferritin and AISI,and the above indicators are the predictors of ILD in patients with CTD.The prediction model constructed by the above indicators can provide a new,convenient and rapid method for predicting the risk probability of ILD in patients with CTD,which can guide early clinical intervention.