ICU中妊娠合并肺动脉高压孕产妇死亡因素分析

Analysis on risk factors for mortality of pregnant women with pulmonary hypertension in ICU

目的分析孕产妇合并肺动脉高压(PH)进入ICU的第一次记录的生命体征、实验室检查,探究与妊娠合并PH入ICU住院死亡相关的危险因素。方法采用回顾性分析的方法,选择2011年1月1日至2020年3月1日在西京医院进入ICU接受治疗的113例合并PH孕产妇作为研究对象,按照住院期间是否发生死亡事件将患者分为存活组和死亡组。通过病例管理系统提取患者的个人基本信息、入住ICU的第一次生命体征和实验室检查结果,以及住院期间患者的结局情况。对进入ICU的第一次实验室检查进行logistic回归分析,筛选出合并PH孕产妇住院死亡相关危险因素,采用受试者工作特征(ROC)曲线分析危险因素联合诊断ICU中合并PH孕产妇预后的效能。结果113例孕产妇合并PH住进ICU进行治疗,排除1例在ICU中完全缺失实验室检查的患者,最终112例纳入分析,住院期间死亡10例(病死率8.9%)。与存活组相比,死亡组母亲年龄小[(24.30±4.32)岁vs(27.27±5.47)岁],怀孕次数少[(1.30±0.48)次vs(2.03±1.17)次],肺动脉压值高[(111.80±32.91)mmHg vs(82.19±33.78)mmHg],入ICU时APACHEⅡ评分高[(15.43±7.39)分vs(8.71±5.21)分]。将入住ICU后的第一次记录的生命体征、血气、血常规、肝功、肾功变量进行logistic回归分析,多因素logistic回归显示入住ICU后首次记录的血氧饱和度(SpO_(2))、K^(+)和谷草转氨酶是合并PH孕产妇住院死亡的独立危险因素。结合AIC值,模型最优原则,最终将SpO_(2)、K^(+)、Ca^(2+)、血小板、谷草转氨酶组合构建logistic回归模型。ROC曲线分析显示,5个危险因素联合诊断住院死亡的曲线下面积为0.97。结论入住ICU后第一次记录的SpO_(2)、K^(+)和谷草转氨酶是合并PH孕产妇住院死亡的独立危险因素,结合血小板和Ca^(2+)联合诊断入住ICU的合并PH孕产妇住院死亡效能更好。

Objective To analyze the first recorded vital signs and basic laboratory workup of pregnant women with pulmonary hypertension(PH)entering ICU,and to explore the risk factors related to the death of pregnant women with PH in ICU.Methods Using a retrospective analysis method,113 pregnant women with PH admitted to ICU in Xijing Hospital from January 1,2011 to March 1,2020 were selected as the research objects.The patients were divided into survival group and death group according to whether death occurred during hospitalization.We extracted the basic personal information,the results of first vital signs and laboratory workup in ICU,and the outcome of the patients during hospitalization from the case management system.Logistic regression analysis was performed on the first laboratory workup data in ICU to find out the risk factors related to in-hospital mortality of pregnant women with PH.Receiver operator characteristic(ROC)curve was used to analyze the efficacy of the combined diagnosis of the risk factors for the prognosis of the pregnant women with PH in ICU.Results In 113 pregnant women with PH admitted to ICU,1 patient without laboratory workup in ICU was excluded,and finally 112 cases were included in the analysis.Ten patients died during hospitalization(mortality 8.9%).Compared with the survival group,the death group was younger[(24.30±4.32)years vs(27.27±5.47)years],and had fewer gravidity times(1.30±0.48 vs 2.03±1.17),higher pulmonary arterial pressure[(111.80±32.91)mmHg vs(82.19±33.78)mmHg],and higher APACHE II scores(15.43±7.39 vs 8.71±5.21)when entering ICU.Multivariate logistic regression analysis of the vital signs,blood gas,blood routine examination,liver function,and renal function variables recorded for the first time after admission to ICU showed that the first recorded SpO_(2),potassium ions and aspartate aminotransferase were independent risk factors for in-hospital mortality of pregnant women with PH.Based on AIC value and the principle of model optimization,the logistic regression model was constructed by the combination of SpO_(2),potassium ion,serum calcium,platelets and aspartate aminotransferase.ROC curve analysis showed that the area under the curve for the combination of five risk factors in diagnosis of in-hospital mortality was 0.97.Conclusion SpO_(2),potassium ion and aspartate aminotransferase recorded for the first time after ICU admission were independent risk factors for in-hospital death of pregnant women with PH,and combined with platelet and serum calcium,they had good efficacy in the diagnosis of in-hospital death of pregnant women with PH admitted to ICU.