自主呼吸试验前后下腔静脉变异度对呼吸衰竭合并心功能不全患者撤机失败的预测价值

Predictive value of inferior vena cava variability for weaning failure before and after spontaneous breathing test in patients with respiratory failure and cardiac insufficiency

摘要:

目的 评价自主呼吸试验(SBT)前后下腔静脉变异度(△DIVC)对呼吸衰竭合并心功能不全患者机械通气后撤机失败的预测价值。方法 选取2016年11月至2018年2月在徐州市中心医院ICU进行机械通气的呼吸衰竭合并心功能不全患者120例。患者30 min SBT成功后拔管,48 h内不需要再次气管插管和无创呼吸机辅助为撤机成功组(n=62)；30 min SBT失败或者SBT成功后拔管,但48 h内需要再次插管或无创呼吸机辅助为撤机失败组(n=58)。分别在SBT前和30 min后记录并比较2组患者的临床特征及△DIVC。采用SPSS 17.0软件统计分析。根据数据类型,组间比较采用独立样本t检验、Mann-Whitney U检验或χ²检验。采用多因素logistic回归法分析影响撤机失败的独立危险因素。利用受试者工作特征(ROC)曲线评估△DIVC对撤机失败的预测价值。结果 2组患者在慢性呼吸疾病、急性生理学和慢性健康状况评价Ⅱ(APACHE Ⅱ)评分、第1次插管到拔管的持续时间和血红蛋白水平方面比较差异有统计学意义(P<0.05)。与撤机成功组比较,撤机失败组患者在SBT前及30 min后二氧化碳分压和N端脑钠肽前体显著升高,左室射血分数(LVEF)显著降低,差异有统计学意义(P<0.05)；在SBT后30 min,撤机失败组患者△DIVC较撤机成功组显著升高,差异亦有统计学意义(P<0.05)。多变量logistic回归分析显示,LVEF(OR=1.204,95%CI 1.133~1.381; P=0.015)和SBT后30 min的△DIVC(OR=1.450,95%CI 1.102~2.026; P=0.009)是患者撤机失败的独立危险因素。SBT后30 min的△DIVC预测患者撤机失败的ROC曲线下面积为0.905,最佳截断点0.27,灵敏度为82.4%,特异度为94.4%。LVEF预测患者撤机失败的曲线下面积为0.806,最佳截断点为42.0%,灵敏度为54.5%,特异度为86.8%。结论 SBT 30 min后的△DIVC对于呼吸衰竭合并心功能不全患者撤机失败具有较高的预测价值。

Abstract:

Objective To evaluate the predictive value of inferior vena cava variability (△DIVC) for weaning failure before and after spontaneous breathing test (SBT) after mechanical ventilation in patients with respiratory failure and cardiac insufficiency. Methods From November 2016 to February 2018,0 patients with respiratory failure and cardiac insufficiency were included, who underwent mechanical ventilation in the ICU of Xuzhou Central Hospital of Jiangsu Province. The patients were divided into successful withdrawal (SW) group (n=62) who did not need tracheal intubation and non-invasive ventilator within 48 h after successful 30-min SBT and failed-withdrawal (FW) group (n=58) who needed reintubation or non-invasive ventilator 48 h after successful or failed 30-min SBT. Clinical characteristics and △DIVC of the two groups were recorded and compared before SBT and 30 min after SBT. SPSS statistics 17.0 was used for analysis. According to the data type, independent sample t test, Mann-Whitney U test or χ² test was used for comparison between groups. [JP+1]Multivariate logistic regression was used to analyze the risk factors for weaning failure, and the predictive value of △DIVC on weaning failure was analyzed using receiver operating characteristic(ROC)curve. Results There were significant differences between the 2 groups (P<0.05) in chronic respiratory disease, acute physiology and chronic health evaluation Ⅱ(APACHE Ⅱ), duration from the first intubation to extubation, and hemoglobin levels. The partial pressure of carbon dioxide and N-terminal pro-brain natriuretic peptide were significantly higher and left ventricular ejection fraction (LVEF) significantly lower in FW group than in SW group before and 30 min after SBT (P<0.05). At 30 min after SBT, the △DIVC in the FW group was significantly higher than that in the SW group (P<0.05). Multivariate logistic regression showed that LVEF (OR=1.204,5%CI 1.133-1.381; P=0.015) and △DIVC at 30 min after SBT (OR=1.450,5%CI 1.102-2.026; P=0.009) were independent risk factors for weaning failure. The area under ROC curve (AUC) for △DIVC was 0.905, and the best cut-off point was 0.27,2.4% for sensitivity and 94.4% for specificity. AUC for LVEF was 0.806, and the best cut-off point was 42.0%, 54.5% for sensitivity and 86.8% for specificity. Conclusion △DIVC at 30 min after SBT has a high predictive value for weaning failure in the patients with respiratory failure and cardiac disease.