Slope analysis for the prediction of fluid responsiveness by a stepwise PEEP elevation recruitment maneuver in mechanically ventilated patients

OBJECTIVE: Assessment of fluid responsiveness is problematic in intensive care unit patients. Lung recruitment maneuvers (LRM) can be used as a functional test to predict fluid responsiveness. We propose a new test to predict fluid responsiveness in mechanically ventilated patients by analyzing the variations in central venous pressure (CVP) and systemic arterial parameters during a prolonged sigh breath LRM without the use of a cardiac output measuring device. DESIGN: Prospective observational cohort study. SETTING: Intensive Care Unit, Saint-Etienne University Central Hospital. PATIENTS: Patients under mechanical ventilation, equipped with invasive arterial blood pressure, CVP, pulse contour analysis (PICCO™), requiring volume expansion, with no right ventricular dysfunction. Interventions. None. MEASUREMENTS AND MAIN RESULTS: CVP, systemic arterial parameters and stroke volume (SV) were recorded during prolonged LRM followed by a 500 mL fluid expansion to asses fluid responsiveness. 25 patients were screened and 18 patients analyzed. 9 patients were responders to volume expansion and 9 were not. Evaluation of hemodynamic parameters suggested the use of a linear regression model. Slopes for systolic arterial pressure, pulse pressure (PP), CVP and SV were all significantly different between responders and non-responders during the pressure increase phase of LRM (STEP-UP) (p = 0.022, p = 0.014, p = 0.006 and p = 0.038, respectively). PP and CVP slopes during STEP-UP were strongly predictive of fluid responsiveness with an AUC of 0.926 (95% CI, 0.78 to 1.00), sensitivity = 100%, specificity = 89% and an AUC = 0.901 (95% CI, 0.76 to 1.00), sensibility = 78%, specificity = 100%, respectively. Combining sensitivity of PP and specificity of CVP, prediction of fluid responsiveness can be achieved with 100% sensitivity and 100% specificity (AUC = 0.96; 95% CI, 0.90 to 1.00). One patient showed inconclusive values using the grey zone approach (5.5%). CONCLUSIONS: In patients under mechanical ventilation with no right heart dysfunction, the association of PP and CVP slope analysis during a prolonged sigh breath LRM seems to offer a very promising method for prediction of fluid responsiveness without the use and associated cost of a cardiac output measurement device. TRIAL REGISTRATION: NCT04304521, IRBN902018/CHUSTE. Registered 11 March 2020, Fluid responsiveness predicted by a stepwise PEEP elevation recruitment maneuver in mechanically ventilated patients (STEP-PEEP)