改良型被动直腿抬高试验对出血性休克稳定期患者液体反应性的预测价值

Modified Passive Leg Raising Maneuver in Fluid Responsiveness:A Study of Predictive Values in Patients with Hemorrhagic Shock in Stable Phase

目的探讨改良型被动直腿抬高试验(PLRm)对出血性休克稳定期患者液体反应性的预测价值。方法选取2011年5月—2013年5月中国人民武装警察部队后勤学院附属医院收入消化内科后转入重症医学科的出血性休克稳定期患者30例。首先进行被动抬腿试验(PLR),记录PLR前后的血流动力学指标变化〔心率(HR)、平均动脉压(MAP)、心脏指数(CI)、心排出量(CO)、每搏输出量(SV)、中心静脉压(CVP)等〕。PLR后进行液体负荷试验,在30 min内给予患者250 ml羟乙基淀粉,记录液体负荷试验前后的血流动力学指标,液体负荷试验后SV的变化(△SV)≥15%定义为液体反应性阳性(作为液体反应性阳性组),反之为液体反应性阴性(作为液体反应性阴性组)。采用ROC曲线分析各血流动力学指标预测液体反应性的准确性(DA),以曲线下面积(AUC)及95%可信区间(CI)表示;各指标AUC的比较采用非参数检验;计算各指标预测液体反应性的敏感度、特异度、阳性预测值(PPV)、阴性预测值(NPV)、DA,指标阈值的选取遵循约登指数最大化的原则。结果经液体负荷试验后,液体反应性阳性组17例,液体反应性阴性组13例。两组的性别构成、年龄、体质指数、多巴胺用量、HR、舒张压(DBP)、MAP、CVP、血红蛋白(Hb)比较,差异均无统计学意义(P>0.05)。液体反应性阳性组的收缩压(SBP)低于液体反应性阴性组,而PLRm阳性标准高于液体反应性阴性组(P<0.05)。PLRm阳性标准与△SV呈正相关(r=0.673,P<0.05);患者在步骤一时的SBP与△SV呈负相关(r=-0.450,P=0.013);而患者在步骤一时的DBP、MAP、HR及CVP均与△SV无直线相关(r=-0.063、-0.292、0.343、0.061,P>0.05)。PLRm阳性标准预测液体反应性的AUC为(0.96±0.03),约登指数最大为0.805,PLRm阳性标准预测值为≥8.7%,95%CI为(0.82,1.00),与AUC=0.5比较,差异有统计学意义(P<0.05)。步骤一时的SBP和HR的AUC与AUC=0.5比较,差异有统计学意义(P<0.05);而步骤一时DBP、MAP、CVP的AUC与AUC=0.5比较,差异无统计学意义(P>0.05)。以步骤二与步骤一各指标的变化(△SBP、△DBP、△MAP、△HR、△CVP)来预测患者液体反应性,只有△SBP和△HR的AUC与AUC=0.5比较,差异有统计学意义(P<0.05)。PLRm阳性标准的AUC与SBP、HR、△SBP和△HR的AUC比较,差异有统计学意义(Z=2.24、2.70、2.31、1.99,P<0.05);约登指数最大为0.457、0.416、0.398、0.647,所对应的临界值分别为SBP≤122 mm Hg(1 mm Hg=0.133 kPa)、HR≥96次/min、△SBP>1.5%和△HR≤-9.28%。以PLRm≥8.7%为阳性标准值,预测液体反应性的敏感度为88.2%(15/17),特异度为92.3%(12/13),PPV为93.8%(15/16),NPV为85.7%(12/14),DA为90.0%(27/30)。SBP≤122 mm Hg、HR≥96次/min、△SBP>1.5%和△HR≤-9.28%为临界值,预测液体反应性的DA分别为73.3%(22/30)、70.0%(21/30)、70.0%(21/30)和80.0%(24/30)。结论对于出血性休克的患者,PLRm预测液体反应性的准确性高,应用其判断患者的液体反应性阳性与否,然后根据预测结果决定是否进行液体治疗,是可行性较高、简单易行、易于推广的临床治疗策略。

Objective To explore the predictive values of modified passive leg raising maneuver（PLRm） for fluid re-sponsiveness in patients with hemorrhagic shock（HS） in stable phase. Methods PLRm was carried out in 30 HS patients in the＆amp;nbsp;Affiliated Hospital of Logistics University of Chinese Peopleˊs Armed Police Force from May 2011 to May 2013. The changes of hemodynamic indexes were recorded before and after PLRm including heart rate（HR）, mean arterial pressure（MAP）, cardiac index（CI）, cardiac output（CO）, stroke volume（SV）, central venous pressure（CVP）, etc. After PLRm,liquid load test（LLT） was performed and the patients were given 250 ml hydroxyethyl starch within 30 min. Hemodynamic indexes and SV chan-ges recorded before and after LLT. △SV ≥15% was defined as positive liquid reactivity（group A）,the contrary as negative（group B）. Receiver operating characteristic curve（ROC） was used to analyze the accuracy of hemodynamic indexes predicting fluid responsiveness,expressed by the area under the curve（AUC）and 95% confidence intervals（CI）. Non-parametric test was used to compare the AUC of the indexes;the sensitivity,specificity,positive predictive value（PPV）, negative predictive value（NPV） and diagnostic DA calculated;indicator threshold selected according to Youden index maximization principle. Re-sUlts After LLT,there were 17 cases in group A,13 in group B. There was no significant difference in gender,age,BMI, dosage of dopamine,HR,DBP,MAP,CVP,Hb between 2 groups（P 〉0. 05）. SBP was lower,PLRm positive standard higher in group A than in group B（P〈0. 05）. PLRm positive standard was positively correlated with △SV（r=0. 673,P〈0. 05）;SBP negatively with△SV at Step 1（r= -0. 450,P=0. 013）,but DBP,MAP,HR,CVP not correlated with△SV （r= -0. 063,-0. 292,0. 343,0. 061,P＆gt;0. 05）. When AUC of PLRm positive standard predicting fluid responsiveness was（0. 96 ＆#177; 0. 03）,the maximum Youden index was 0. 805,PLRm standard prediction was ≥8. 7%,95%CI =（0. 82, 1. 00）,differing from that when AUC was 0. 5（P〈0. 05）. AUC of SBP,HR at Step 1 was different from AUC=0. 5（P〈0. 05）,AUC of DBP,MAP,CVP not from AUC=0. 05 at Step 1（P 〉0. 05）. Only AUC of △SBP,△HR differed from AUC=0. 5 when using the changes of the indicators（△SBP,△DBP、△MAP,△HR,△CVP）to predict fluid responsive-ness at Steps 2,1（P〈0. 05）. There was difference between AUC of PLRm positive standard and that of SBP,HR,△SBP,△HR（Z=2. 24、2. 70,2. 31,1. 99,P〈0. 05）;the maximum Youden indexes were 0. 457,0. 416,0. 398,0. 647,and their dividing values were SBP≤122 mm Hg（1 mm Hg =0. 133 kPa）,HR≥96 times/min,△SBP ＆gt;1. 5%,△HR≤ -9. 28%. Taking PLRm≥8. 7% as positive standard value,the sensitiveness was 93. 8%（15/17）,specificity 92. 3%（12/13）,PPV 93. 8%（15/16）,NPV 85. 7%（12/14）,DA 90. 0%（27/30）. Taking SBP≤122 mm Hg,HR≥96 次/min,△SBP〉1. 5%,△HR≤ -9. 28% as dividing values,DA was 73. 3%（22/30）,70. 0%（21/30）,70. 0%（21/30）, 80. 0%（24/30）. ConclUsion For HS patients,the accuracy of PLRm predicting fluid responsiveness is higher. Using PLRm to judge positive fluid responsiveness and to decide whether to perform fluid therapy is feasible,simple and practicable.