Background: Previous studies in adult patients and animal models have demonstrated that pulse pressure variation (PPV) and stroke volume variation (SVV) can be used to predict the response to fluid administration. Currently, little information is available on the performance of these variables in infants and neonates. The aim of our study was to assess whether PPV and SVV can predict fluid responsiveness in an animal model and to investigate the influence of different tidal volumes applied.
Methods: PPV and SVV were monitored by pulse contour analysis in 19 anaesthetized and paralysed piglets during ventilation with tidal volumes (V(T)) of 5, 10 and 15 ml/kg both before and after fluid loading with 25 ml/kg of hydroxy-ethyl starch 6% (HES). Cardiac output was measured by pulmonary artery thermodilution and a positive response to HES infusion was defined as >/=20% increase in the stroke volume index (SVI).
Results: Before HES infusion, PPV and SVV were significantly greater during ventilation with a V(T) of 10 and 15 ml/kg than during ventilation with a V(T) of 5 ml/kg (P<0.05). After HES infusion, only ventilation with V(T) 15 ml/kg resulted in a significant increase in PPV and SVV. As assessed by receiver operating characteristic curve analysis, SVV during ventilation with V(T) 10 ml/kg was the best predictor of a positive response to fluid loading (AUC=0.87).
Conclusions: In this paediatric animal model, we found that SVV during ventilation with 10 ml/kg was a sensitive and specific predictor of the response to fluid loading.