新冠肺炎患者出院1个月后呼吸功能障碍预测列线图的构建与验证

Development and Validation of a Nomogram for Prediction of Respiratory Dysfunction One Month after Discharge in COVID-19 Patients

目的利用单中心研究队列建立和验证新型冠状病毒肺炎(新冠肺炎)患者出院1个月后的呼吸功能障碍预测列线图。方法前瞻性分析96例新冠肺炎患者的临床资料,对缺失的数据进行多重插补。数据降维、预测变量筛选采用LASSO(the least absolute shrinkage and selection operator)回归模型,采用多元Logistic回归分析建立预测模型,并制作列线图。从区分度、校准度与内部验证方面对列线图的性能进行评估。结果在LASSO回归模型中从8个潜在预测变量中选择了6个非零系数的预测变量,包括年龄、性别、就医延迟天数(DMT)、临床分型、吸烟史和基础疾病,该模型具有良好的区分性,一致性指数(C-index)为0.78(95%CI:0.69~0.87)。结论研究建立的新冠肺炎患者出院1个月后呼吸功能障碍预测列线图具有较好的区分度和校准度。该列线图的建立有助于医务人员提高对新冠肺炎患者出院1个月后呼吸功能障碍程度的预测,从而及时调整治疗方案,提高疗效,降低呼吸功能障碍的风险。

Objective To develop and validate a nomogram for the prediction of respiratory dysfunction 1 month after discharge in patients with COVID-19 using a single-center study cohort.Methods Clinical data of 96 patients with COVID-19 were prospectively analyzed.Multiple imputation was used to deal with the missing data.The least absolute shrinkage and selection operator(LASSO)regression model was used for data dimension reduction and predictive factor selection.The predictive model was established by multivariable logistic regression analysis,and the selected risk factors were incorporated to develop a nomogram.The performance of the nomogram was assessed with respect to its calibration,discrimination and internal validation.Results In the LASSO regression model,6 predictors with non-zero coefficients were selected from the 8 potential predictors,including age,gender,delayed medical treatment,clinical classification,smoking history and underlying disease.The model showed good discrimination,with a C-index of 0.78(95%CI:0.69-0.87).Conclusion The nomogram developed for the prediction of respiratory dysfunction 1 month after discharge in patients with COVID-19 has good discrimination and calibration.It can help medical staff to improve their prediction of the degree of respiratory dysfunction in COVID-19 patients 1 month after discharge and thereby to timely adjust treatment regiments for enhancing efficacy and reducing the risk of respiratory dysfunction.