Abstract P210: Measurement of Fluid Responsiveness Following Resuscitation From Prolonged Cardiac Arrest
Background: Fluid therapy after cardiac arrest is challenging as both hypovolemia and fluid overload may aggravate hemodynamic instability. Therefore, predicting fluid responsiveness is a major issue during post-resuscitation care. Variables such as pulse pressure variation (PPV) and stroke volume variation (SVV) have been shown to be superior to static filling pressures like central venous pressure (CVP) in predicting an increase of stroke volume (SV) following fluid loading (1). The aim of the present study was to evaluate the performance of PPV and SVV to predict fluid responsiveness in the post-resuscitation period.
Methods: After institutional animal board approval, 20 pigs were anesthetized and equipped with a thermistor-tipped arterial catheter (PiCCO plus, Pulsion Medical Systems, Munich, Germany) and a central venous line. Electrically-induced cardiac arrest of 8 minutes was followed by resuscitation according to the AHA-guidelines (2). PPV, SVV, CVP and SV obtained by trans-cardiopulmonal thermodilution were measured before and after a 5 ml/kg fluid bolus at baseline, after 30 minutes and four hours following return of spontaneous circulation (ROSC). Animals were classified as responders when SV increased >15% after fluid load, otherwise as non-responders. Performance of variables was analyzed using receiver-operator characteristics (ROC) analysis.
Results: 15 animals were successfully resuscitated. SV significantly decreased after ROSC (p<0.01). At baseline, PPV and SVV but not CVP indicated an increase in SV following fluid loading with an area under the curve of 0.93 (p<0.01), 0.82 (p0.05) respectively. 30 minutes after ROSC, all variables failed to predict fluid responsiveness. Four hours after ROSC only PPV enabled prediction of fluid responsiveness with an area under the curve of 0.76 (p<0.05).
Conclusion: During the initial post-resuscitation period all variables failed to predict fluid responsiveness. After hemodynamic stabilization, however, the dynamic variable PPV was able to predict increase of stroke volume after fluid challenge. Failure of SVV derived from pulse contour analysis to predict fluid responsiveness may be rooted in the more complex calculation algorithm underlying SV determination.