Reliability of Passive Leg Raising, Stroke Volume Variation and Pulse Pressure Variation to Predict Fluid Responsiveness During Weaning From Mechanical Ventilation After Cardiac Surgery: A Prospective, Observational Study.

OBJECTIVE: During assisted ventilation and spontaneous breathing, functional haemodynamic parameters, including stroke volume variation (SVV) and pulse pressure variation (PPV), are of limited value to predict fluid responsiveness, and the passive leg raising (PLR) manoeuvre has been advocated as a surrogate method. We aimed to study the predictive value of SVV, PPV and PLR for fluid responsiveness during weaning from mechanical ventilation after cardiac surgery.
METHODS: Haemodynamic variables and fluid responsiveness were assessed in 34 patients. Upon arrival at the intensive care unit, measurements were performed during continuous mandatory ventilation (CMV) and spontaneous breathing with pressure support (PSV) and after extubation (SPONT). The prediction of a positive fluid responsiveness (defined as stroke volume increase >15% after fluid administration) was tested by calculating the specific receiver operating characteristic (ROC) curves.
RESULTS: A significant increase in stroke volumes was observed during CMV, PSV and SPONT after fluid administration. There were 19 fluid responders (55.9%) during CMV, with 22 (64.7%) and 13 (40.6%) during PSV and SPONT, respectively. The predictive value for a positive fluid responsiveness (area under the ROC curve) for SVV was 0.88, 0.70 and 0.56; was 0.83, 0.69 and 0.48 for PPV; was 0.72, 0.74 and 0.70 for PLR during CMV, PSV and SPONT, respectively.
CONCLUSION: During mechanical ventilation, adequate prediction of fluid responsiveness using SVV and PPV was observed. However, during spontaneous breathing, the reliability of SVV and PPV was poor. In this period, PLR as a surrogate was able to predict fluid responsiveness better than SVV or PPV but was less reliable than previously reported.