被动抬腿试验预测液体复苏后心功能不全患者容量反应性的价值

Value of passive leg raising test in predicting volume responsiveness in patients with cardiac dysfunction after fluid resuscitation

目的探讨两种体位下被动抬腿试验(PLR)对液体复苏后心功能不全患者容量反应性的预测价值。方法采用前瞻性、观察性队列研究方法,选取2016年1月至2017年3月在梅州市人民医院ICU住院的52例液体复苏后并心功能不全的重症患者为研究对象。对患者依次行平卧位PLR、半卧位PLR和补液试验(VE),用心阻抗无创血流动力学监测系统监测整个过程,记录基线、PLR和VE后心排血量(CO)、心率(HR)以及平均动脉压(MAP)等数据。根据VE后CO变化率(ΔCOVE)是否≥10%将患者分为有反应组和无反应组。分别对PLR和VE后ΔCO、ΔHR、ΔMAP进行相关性分析,采用ROC曲线评价两种体位下PLR对容量反应性的预测价值。结果 52例患者中有反应组27例,无反应组25例。PLR和VE时有反应组的ΔCO均明显大于无反应组[平卧位PLR:(11.17±3.36)%vs(4.14±7.17)%,t=4.469,P<0.01;半卧位PLR:(9.31±3.14)%vs(-4.14±10.55)%,t=6.147,P<0.01;VE:(13.79±2.35)%vs(0.84±7.16)%,t=8.620,P<0.01,且ΔCOPLR平卧、ΔCOPLR半卧均与ΔCOVE呈正相关(平卧位r=0.737,半卧位r=0.903,均P<0.01)。△COPLR平卧、△COPLR半卧预测患者容量反应性的ROC曲线下面积(AUC)分别为0.845、0.978,△COPLR平卧预测容量反应性最佳临界值、灵敏度和特异度分别为8.7%、85.2%、76.0%,△COPLR半卧预测容量反应性最佳临界值、灵敏度和特异度分别为5.9%、92.6%、96.0%。结论两种体位下PLR均能准确预测液体复苏后心功能不全患者的容量反应性,且半卧位PLR优于平卧位PLR。

Objective To investigate the value of passive leg raising（PLR） test in two different postures in predicting volume responsiveness in patients with cardiac dysfunction after fluid resuscitation. Methods A prospective observational cohort study was conducted in ICU of Meizhou People＇s Hospital from January 2016 to March 2017. Fiftytwo severe patients with cardiac dysfunction after fluid resuscitation were selected as the research subjects. A supine PLR test, a half-recumbent PLR trial, a final volume expansion test（VE） were conducted in-order for every patient. Impedance under monitoring of impedance cardiography throughout the process, the baseline data, cardiac output（CO）,heart rate（HR）, and mean arterial pressure（MAP） after PLR and VE were recorded. The patients were divided into reactive group and no response group according to whether the added value of CO（ΔCO） after VE was no less than 10%.The correlations of ΔCO orΔHR orΔMAP after PLR and VE were respectively analyzed. The receiver operating characteristic curve（ROC） was used to evaluate the predictive value of PLR in different postures on volume responsiveness in severe patients. Results Of the 52 patients, 27 were in reactive group, and the rest were in no response group. ΔCO of the patients in the reactive group were significantly higher than that in the no response group, either after PLR or VE,（11.17±3.36）% vs（4.14±7.17）% for supine PLR, t=4.469, P0.01;（9.31±3.14）% vs（-4.14±10.55）% for half-recumbent PLR, t=6.147, P0.01;（13.79 ± 2.35）% vs（0.84 ± 7.16）% for VE, t=8.620, P0.01. ΔCO supine PLR, ΔCO half-recumbent PLR were positively correlated with ΔCOVE（r=0.737 for supine, r=0.903 for half-recumbent, P0.01）. The area under ROC curve（AUC） of ΔCO for supine PLR and ΔCO for half-recumbent PLR in predicting volume responsiveness of patients was 0.845, 0.978, respectively. The optimal diagnostic cutoff point, sensitivity and specificity for ΔCO of supine PLR were 8.7%, 85.2%, 76.0%,respectively. The optimal diagnostic cutoff point, sensitivity and specificity forΔCO of half-recumbent PLR were 5.9%, 92.6%, 96.0%,respectively. Conclusion Both PLRs can be used to accurately predict the volume responsiveness in patients with cardiac dysfunction after fluid resuscitation, and the evaluation value of half-recumbent PLR is better than that of supine PLR.