Objective: To assess the ability of non-invasive ultrasonic cardiac output monitor (USCOM) combined with passive leg raising (PLR) test to predict volume responsiveness in septic shock children with spontaneous respiration.
Methods: Prospective and observational cohort study was performed in 40 septic shock children with spontaneous breathing admitted to pediatric intensive care unit (PICU) of Chongqing Medical University Shenzhen Children's Hospital from March 2011 to June 2013. The hemodynamic parameters including stroke volume (SV), cardiac output (CO) and systemic vascular resistance index (SVRI) were measured non-invasively by USCOM device before and after PLR and volume expansion (VE) test. And invasive mean arterial pressure (MAP) and central venous pressure (CVP) were monitored continuously. Based on the responsiveness of volume expansion [children were considered to be responders to volume expansion if SV increased (ΔSVVE)≥15%], all the children were divided into responders and non-responders. The roles of PLR in predicting volume responsiveness were evaluated by receiver operating characteristic curve (ROC curve).
Results: A total of 43 PLR and VE tests in 40 children were evaluated and resulting in 25 responders and 18 non-responders. There was no significant difference between two groups in the clinical data and hemodynamics indicators at incipient stage. After PLR and VE, the SV was increased compared with that at supine position in both responder group and non-responder group. The ΔSV after PLR (ΔSVPLR) and ΔSVVE in responder group were significantly higher than those in non-responder group [(14.95±3.44)% vs. (8.48±3.49)%, t=6.048, P=0.000; (18.28±2.84)% vs. (6.57±3.83)%, t=11.530, P=0.000]. Correlation analysis showed that there was the positive correlation between ΔSVPLR and ΔSVVE (r=0.649, P=0.000), but CVP increased (ΔCVPPLR) were unrelated with ΔSVVE (r=0.217, P=0.162). The area under the ROC curve (AUC) of ΔSVPLR and ΔCVPPLR for PLR predicting volume responsiveness and 95% confidence interval (95%CI) were 0.900±0.046 (95%CI 0.809-0.991, P=0.000) and 0.561±0.090 (95%CI 0.385-0.737, P=0.498). The ΔSVPLR≥12.25% was found to predict volume responsiveness with a sensitivity of 80.0% and specificity of 88.9%, the sensitivity and specificity of ΔCVPPLR≥15.48% were 76.0% and 38.9%, respectively. The capability of ΔSVPLR to predict volume responsiveness was better than ΔCVPPLR.
Conclusions: ΔSVPLR measured by USCOM can predict the volume responsiveness in septic shock children patients with spontaneous respiration and it is reliable to guide fluid therapy.