列线图在老年COPD患者发生肺动脉高压风险预测中的应用

The application of nomogram in risk prediction of pulmonary hypertension in elderly patients with COPD

目的建立列线图模型个体化预测老年慢性阻塞性肺病(COPD)患者发生肺动脉高压(PH)的高危风险因素。方法选取2015年1月至2019年1月湖州市第三人民医院收治的COPD老年患者420例,其中发生PH的患者80例作为发生PH组,未发生PH的患者340例作为未发生PH组。回顾性分析患者的临床资料,采用Logistic回归分析法筛选出老年COPD患者发生PH的独立风险因素。利用R软件对筛选出的独立风险因素建立列线图预测模型,并对模型的预测性及准确度进行内部验证。结果通过Logistic二元回归分析患者的风险因素数据可知,居住环境(OR=2.861,95%CI:1.575~5.198)、吸烟史(OR=3.034,95%CI:1.318~6.984)、C-反应蛋白(CRP)(OR=3.028,95%CI:1.599~5.733)、白细胞(WBC)计数(OR=6.156,95%CI:2.915~12.997)、第1秒用力呼气容积比率(FEV_(1)%)(OR=3.443,95%CI:1.891~6.270)、二氧化碳分压(PaCO_(2))(OR=2.746,95%CI:1.513~4.984)为老年COPD患者发生PH的独立危险因素。基于上述6项风险因素建立预测老年COPD患者发生PH的列线图模型,校准曲线验证结果显示预测值与实测值曲线重合度较高,表明该预测模型准确度优良;同时由自举法内部验证,结果显示一致性指数(C-index)高达0.830(95%CI:0.734~0.924),提示该模型精准度和区分度较好。结论基于多项老年COPD患者发生PH的危险因素建立的列线图模型具有较高的预测价值,医护人员可以根据危险因素在早期进行有效防范与干预,进而减少患者发生PH,改善患者预后。

Objective To predict the high-risk factors of pulmonary hypertension(PH)in elderly patients with chronic obstructive pulmonary disease(COPD)by means of individual nomogram models.Methods 420 elderly patients with COPD treated in Huzhou Third People's Hospital from January 2015 to January 2019 were selected,including 80 patients with PH as the PH group and 340 patients without PH as the PH group.The clinical data of patients were analyzed retrospectively,and independent risk factors for PH in elderly patients with COPD were screened out by Logistic regression analysis.R software was used to establish the nomograph prediction model for the screened independent risk factors,and the prediction and accuracy of the model were verified internally.Results The risk factors of the patients were analyzed by Logistic binary regression,including living environment(OR=2.861,95%CI:1.575~5.198),smoking history(OR=3.034,95%CI:1.318~6.984),C-reactive protein(CRP)(OR=3.028,95%CI:1.599~5.733)and white blood cell(WBC)count(OR=6.156,95%CI:2.915~12.997),forced expiratory volume ratio at the first second(FEV1%)(OR=3.443,95%CI:1.891~6.270),and partial pressure of carbon dioxide(PaCO_(2))(OR=2.746,95%CI:1.513~4.984)were independent risk factors for PH in elderly patients with COPD.Based on the above 6 risk factors,a mapper model was established to predict PH of elderly COPD patients.The calibration results showed that the predicted value was closely aligned with the measured value,indicating that the predicted model was of good accuracy.At the same time,the internal validation of bootstrap method showed that the C-index was as high as 0.830(95%CI:0.734~0.924),indicating that the model had better accuracy and differentiation.Conclusion The nomogram model established in this study based on multiple risk factors for PH in elderly COPD patients has highly prediction value.In clinical treatment,medical staff can effectively prevent and intervene in the early stage according to the risk factors,so as to reduce the incidence of PH and improve the prognosis of patients.