老年慢性阻塞性肺疾病急性加重期合并肺血栓栓塞症风险的Nomogram预测模型的构建

Construction of Nomogram prediction model for risk of pulmonary thromboembolism in elderly patients with acute exacerbation of chronic obstructive pulmonary disease

目的分析老年慢性阻塞性肺疾病急性加重期(AECOPD)发生肺血栓栓塞症(PTE)的危险因素,并构建Nomogram风险预警模型。方法回顾性分析2019年1月至2021年6月新疆维吾尔自治区人民医院收治的265例老年AECOPD患者,根据AECOPD患者是否发生PTE分为PTE组和非PTE组。收集两组一般资料、实验室指标、基础疾病等数据,分析老年AECOPD患者发生PTE的危险因素;根据危险因素建立老年AECOPD患者合并PTE风险的Nomogram预测模型并验证其预测效能。结果265例老年AECOPD患者中,PTE发生率为20.38%。两组性别、体重指数、饮酒史、抽烟史、高血压、糖尿病、脑血管疾病、冠心病、心房颤动、血红蛋白、白细胞计数比较,差异无统计学意义(P>0.05);两组年龄、静脉血栓史、下肢肿胀、下肢疼痛、发热时间、卧床时间、D-二聚体(D-D)、动脉血二氧化碳分压(PaCO_(2))、动脉血氧分压(PaO_(2))比较,差异有统计学意义(P<0.05)。年龄>69岁(OR=3.205)、静脉血栓史(OR=4.035)、下肢肿胀(OR=3.149)、下肢疼痛(OR=8.193)、发热时间≥3 d(OR=2.323)、卧床时间≥7 d(OR=3.451)、D-D≥200μg/L(OR=2.426)、PaCO_(2)<60 mmHg(OR=3.786)、PaO_(2)<39 mmHg(OR=2.860)为老年AECOPD患者发生PTE的独立危险因素(P<0.05)。方程为Y=-4.780+1.165×年龄+1.395×静脉血栓史+1.147×下肢肿胀+2.103×下肢疼痛+0.843×发热时间+1.239×卧床时间+0.886×D-D+1.331×PaCO_(2)+1.051×PaO_(2)。老年AECOPD患者发生PTE风险的Nomogram预测模型显示,模型一致性指数(C-index)为0.862;模型受试者操作特征曲线的曲线下面积为0.851。结论年龄、静脉血栓史、下肢肿胀、下肢疼痛、发热时间、卧床时间、D-D、PaCO_(2)、PaO_(2)与老年AECOPD患者发生PTE密切相关,基于上述危险因素所构建的Nomogram预测模型可有效预警老年AECOPD患者PTE的发生风险。

Objective To analyze the risk factors of pulmonary thromboembolism(PTE)in elderly patients with acute exacerbation of chronic obstructive pulmonary disease(AECOPD),and construct a Nomogram risk warning model.Methods A retrospective analysis of 265 elderly patients with AECOPD admitted to Xinjiang Uygur Autonomous Region People’s Hospital from January 2019 to June 2021 were conducted,they were divided into PTE group and non-PTE group according to whether PTE occurred.General data,laboratory indicators,and underlying diseases of two groups were collected,and the risk factors of PTE in elderly AECOPD patients were analyzed;the Nomogram risk model to predict the risk of PTE in elderly patients with AECOPD based on the risk factors and its predictive efficacy was verified.Results Among 265 elderly patients with AECOPD,the incidence of PTE was 20.38%.There were no significant differences in gender,body mass index,drinking history,smoking history,hypertension,diabetes mel litus,cerebrovascular disease,coronary heart disease,atrial fibrillation,hemoglobin,and white blood cell count between two groups(P>0.05);there were statistically significant differences in age,history of venous thrombosis,lower limb swelling,lower limb pain,fever time,bed time,D-dimer(D-D),arterial partial pressure of carbon dioxide(PaCO_(2)),and arterial partial pressure of oxygen(PaO_(2))between two groups(P<0.05).Age>69 years old(OR=3.205),history of venous thrombosis(OR=4.035),lower limb swelling(OR=3.149),lower limb pain(OR=8.193),fever time≥3 d(OR=2.323),bed time≥7 d(OR=3.451),D-D≥200μg/L(OR=2.426),PaCO_(2)<60 mmHg(OR=3.786),and PaO_(2)<39 mmHg(OR=2.860)were independent risk factors for PTE in elderly patients with AECOPD(P<0.05).The equation was Y=-4.780+1.165×age+1.395×history of venous thrombosis+1.147×lower limb swelling+2.103×lower limb pain+0.843×fever time+1.239×bed time+0.886×D-D+1.331×PaCO_(2)+1.051×PaO_(2).The Nomogram prediction model to predict the risk of PTE in elderly patients with AECOPD showed that the model consistency index was 0.862;the area under the curve of receiver operating characteristic curve of the model was 0.851.Conclusion Age,history of venous thrombosis,lower extremity swelling,lower extremity pain,fever time,bed time,D-D,PaCO_(2),and PaO_(2) are closely related to the occurrence of PTE in elderly AECOPD patients.The Nomogram prediction model based on these risk factors can effectively predict the risk of PTE in elderly patients with AECOPD.