Carotid artery ultrasound for assessing fluid responsiveness in patients undergoing mechanical ventilation with low tidal volume and preserved spontaneous breathing

Abstract

Objective This study aimed to investigate whether changes in carotid artery corrected flow time (ΔFTcbolus) and carotid artery peak flow velocity respiratory variation (ΔVpeakbolus) induced by the fluid challenge could reliably predict fluid responsiveness in mechanically ventilated patients with a tidal volume < 8 ml/kg PBW while preserving spontaneous breathing. Methods Carotid artery corrected flow time (FTc), carotid artery peak flow velocity respiratory variation (ΔVpeak), and hemodynamic data were measured before and after administration of 250 ml crystalloids. Fluid responsiveness was defined as a 10% or more increase in stroke volume index as assessed by noninvasive cardiac output monitoring (NICOM) after the fluid challenge. Results A total of 43 patients with acute circulatory failure were enrolled in this study. 43 patients underwent a total of 60 fluid challenges. ΔFTcbolus and ΔVpeakbolus showed a significant difference between the fluid responsiveness positive group (n = 35) and the fluid responsiveness negative group (n = 25). Spearman correlation test showed that ΔFTcbolus and ΔVpeakbolus with the relative increase in stroke volume index after fluid expansion (r = 0.5296, P < 0.0001; r = 0.3175, P = 0.0135). Multiple logistic regression analysis demonstrated that ΔFTcbolus and ΔVpeakbolus were significantly correlated with fluid responsiveness in patients with acute circulatory failure. The areas under the receiver operating characteristic curves (AUROC) of ΔFTcbolus and ΔVpeakbolus for predicting fluid responsiveness were 0.935 and 0.750, respectively. The optimal cut-off values of ΔFTcbolus and ΔVpeakbolus were 0.725 (sensitivity of 97.1%; specificity of 84%) and 4.21% (sensitivity of 65.7%; specificity of 80%), respectively. Conclusion In mechanically ventilated patients with a tidal volume < 8 ml/kg while preserving spontaneous breathing, ΔFTcbolus and ΔVpeakbolus could predict fluid responsiveness. The predictive performance of ΔFTcbolus was superior to ΔVpeakbolus.