(767) Comparison of Stroke Volume Between Echocardiography and FloTrac During Passive Leg Raise

Introduction: Acute circulatory failure is common in children admitted to the pediatric intensive care unit (PICU), however optimal resuscitation remains controversial. Accurate measurements of fluid responsiveness and dynamic variables to assess hemodynamics are lacking. We investigated if changes in stroke volume (SV) obtained with FloTrac™ were similar to changes in SV with transthoracic echocardiography (TTE) during passive leg raise (PLR).

Methods: Patients 5 years and older with an arterial line placed for routine care were eligible. The FloTrac™ transducer was placed for continuous hemodynamic monitoring. TTE was used to measure SV by calculating the velocity time integral (VTI) of aortic blood flow. Patients were considered to be fluid responsive if SV increased >10%. Baseline data were collected with the patient semi-recumbent and then repeated during PLR. Patient demographics, respiratory support and vasoactive use were recorded. Concordance between percentages changes in VTI and Flotrac™ SV from baseline to PLR was analyzed using Pearson correlation and Bland Altman analysis. The accuracy, sensitivity, and specificity of a 10% change threshold in Flotrac™ SV from baseline to PLR to detect a greater or less than 10% change in VTI from baseline to PLR was calculated.

Results: 18 participants completed the study. 8 (42.1%) were female. Mean (SD) age was 11.8 (4.5) years. Primary diagnoses included craniotomy (n=5 [26.3%]), septic shock (n=4 [21.0%]), posterior spinal fusion (n=3 [15.8%]), and respiratory failure (n=3 [15.8%]). 7 (36.8%) were mechanically ventilated and 6 (31.6%) received vasoactives. There was no correlation (r = -0.04, P = 0.88) between percentage change in Flotrac™ SV from baseline to PLR and percentage change in VTI from baseline to PLR. Systematic bias (average of Flotrac percent change – VTI percentage change) was -3.2 percentage points with wide limits of agreement (lower limit -42.8 percentage points; upper limit 36.4 percentage points). A 10% increase threshold in Flotrac™ SV from baseline to PLR was 63.2% accurate in detecting 10% increase in VTI from baseline to PLR. Sensitivity with this threshold was 16.7%, specificity was 84.6%.

Conclusions: FloTrac™ changes in SV during PLR cannot substitute for TTE measurements to predict fluid responsiveness in PICU patients.