RESUMO

Proper assessment of fluid responsiveness using accurate predictors is crucial to guide fluid therapy and avoid the serious adverse effects of fluid overload. The main objective of this study was to investigate the accuracy of respiratory variations in inferior vena cava diameter (∆IVC) to predict fluid responsiveness in mechanically ventilated children. This prospective single-center study included 32 children (median age and weight of 17 months and 10 kg, respectively) who received a fluid infusion of 10 ml kg-1 of crystalloid solutions over 10 min. ∆IVC and respiratory variation in aortic blood flow peak velocity (∆Vpeak) were determined over one controlled respiratory cycle before and after fluid loading. Thirteen (41%) participants were fluid-responders. ∆IVC, ∆Vpeak, stroke volume index, and cardiac index were found to be predictors of fluid responsiveness. However, the area under the ROC curve of ∆IVC was smaller when compared to ∆Vpeak (0.709 vs. 0.935, p < 0.012). The best cut-off values were 7.7% for ∆IVC (sensitivity, 69.2%; specificity 78.9%, positive predictive value, 69.2%; and negative predictive value, 78.9%) and 18.2% for ∆Vpeak (sensitivity, 84.6%; specificity, 89.5%; positive predictive value, 84.6%; negative predictive value, 89.5%). Changes in stroke volume were positively correlated with ∆IVC (ρ = 0.566, p < 0.001) and ∆Vpeak (ρ = 0.603, p < 0.001). A significant correlation was also found between changes in MAP and ∆Vpeak (ρ = 0.382; p = 0.031), but the same was not observed with ∆IVC (ρ = 0.011; p = 0.951). In conclusion, ∆IVC was found to have a moderate accuracy in predicting fluid responsiveness in mechanically ventilated children and is an inferior predictor when compared to ∆Vpeak.