Résumé
Objective To evaluate the feasibility of using noninvasive ultrasonic cardiac output monitor USCOM velocity time integral (VTI) as the observation index of PLR.Methods This prospective study recruited 36 septic shock or acute pancreatitis patients from October 2014 to October 2016 in the resuscitation room and EICU of Peking Union Medical College Hospital.The change of VTI and plus pressure before and after PLR (⊿VTIplr and ⊿pp),and the change of VTI and stroke volume before and after 500 mL of volume expansion (⊿VTIve and ⊿SV) were recorded.Fluid response positive was defined as stroke volume increase more than 15％ after volume expansion.Results ⊿VTIplr was positively correlated with ⊿SV (Spearman correlation coefficient r=0.888,P＜0.01).The predicting value of⊿VTIve,⊿VTIplr and ⊿PP in fluid response were as follows:the sensitivity of ⊿VTIve in ＞15％ was 94.7％,the specificity was 94.1％,area under the ROC curve was 0.989;the sensitivity of⊿ VTIplr in ＞12％ was 84.2％,the specificity was 88.2％,area under the ROC curve was 0.916;and the sensitivity of⊿ PP in ＞10.5％ was 78.9％,the specificity was 88.2％,the area under the ROC curve was 0.870.Conclusions ⊿ VTIplr measured by USCOM before and after the PLR is a sensitive and specific index.It is better than the classic index ⊿ PP.⊿ VTIplr measured by USCOM is completely noninvasive,which has very good application prospect in the emergency department.