Background: Prediction of the response of the left ventricular stroke volume to fluid administration remains an unsolved clinical problem. We compared the predictive performance of various haemodynamic parameters in the perioperative period in patients undergoing coronary artery bypass surgery. These parameters included static indicators of cardiac preload and functional parameters, derived from the arterial pressure waveform analysis. These included the systolic pressure variation (SPV) and its delta down component (dDown), pulse pressure variation (PPV), stroke volume variation (SVV), and a new parameter, termed the respiratory systolic variation test (RSVT), which is a measure of the slope of the lowest systolic pressure values during a standardized manoeuvre consisting of three successive incremental pressure-controlled breaths.
Methods: Eighteen patients were included into this prospective observational study. Seventy volume loading steps (VLS), each consisting of 250 ml of colloid administration were performed before surgery and after the closure of the chest. The response to each VLS was considered as a positive (increase in stroke volume more than 15%) or non-response. Receiver operating characteristic curves were plotted for each parameter to evaluate its predictive value.
Results: All functional parameters predicted fluid responsiveness better than the intrathoracic blood volume and the left ventricular end-diastolic area. Parameters with the best predictive ability were the RSVT and PPV.
Conclusions: Functional haemodynamic parameters are superior to static indicators of cardiac preload in predicting the response to fluid administration. The RSVT and PPV were the most accurate predictors of fluid responsiveness, although only the RSVT is independent of the settings of mechanical ventilation.