Stroke Volume Variation for Prediction of Fluid Responsiveness in Patients Undergoing Gastrointestinal Surgery

BACKGROUND: Stroke volume variation (SVV) has been shown to be a reliable predictor of fluid responsiveness. However, the predictive role of SVV measured by FloTrac/Vigileo system in prediction of fluid responsiveness was unproven in patients undergoing ventilation with low tidal volume. METHODS: Fifty patients undergoing elective gastrointestinal surgery were randomly divided into two groups: Group C [n(1)=20, tidal volume (V(t)) = 8 ml/kg, frequency (F) = 12/min] and Group L [n(2)=30, V(t)= 6 ml/kg, F=16/min]. After anesthesia induction, 6% hydroxyethyl starch130/0.4 solution (7 ml/kg) was intravenously transfused. Besides standard haemodynamic monitoring, SVV, cardiac output, cardiac index (CI), stroke volume (SV), stroke volume index (SVI), systemic vascular resistance (SVR) and systemic vascular resistance index (SVRI) were determined with the FloTrac/Vigileo system before and after fluid loading. RESULTS: After fluid loading, the MAP, CVP, SVI and CI increased significantly, whereas the SVV and SVR decreased markedly in both groups. SVI was significantly correlated to the SVV, CVP but not the HR, MAP and SVR. SVI was significantly correlated to the SVV before fluid loading (Group C: r = 0.909; Group L: r = 0.758) but not the HR, MAP, CVP and SVR before fluid loading. The largest area under the ROC curve (AUC) was found for SVV (Group C, 0.852; Group L, 0.814), and the AUC for other preloading indices in two groups ranged from 0.324 to 0.460. CONCLUSION: SVV measured by FloTrac/Vigileo system can predict fluid responsiveness in patients undergoing ventilation with low tidal volumes during gastrointestinal surgery.