Background and goals: Functional hemodynamic tests can reduce the risk of complications related to fluid challenge. The transient increase of positive end expiratory pressure (PEEP test) is a promising bedside approach to assess the response of fluid resuscitation, however requiring further validation in different clinical situations. The goal of our study was to assess whether PEEP test predicts fluid responsiveness in patients with septic shock. 
Methods: Seventeen adult patients with septic shock were enrolled into an ongoing prospective observational study. All the patients received vasopressor support with norepinephrine in a dose of 1.37 (1.6–4.34) mcg/kg/min and/or epinephrine 0.52 (0.03–1.07) mcg/kg/min, pressure-controlled ventilation with tidal volume of 8 mL/kg and standardised PEEP 8 cm H2O, and sedation with continuous infusion of propofol. The PEEP test consisted of a transient increase of PEEP from 8 to 20 cm H2O during 5 min. Thereafter, all patients received fluid challenge (7 ml/kg of crystalloids during 10 min). Systolic arterial pressure (SAP), mean arterial pressure (MAP), pulse pressure variations (PPV), thermodilution cardiac index (CI), and CI determined with pulse contour analysis (PCCI) were measured using PiCCO2 monitor (Pulsion Medical System, Germany) before, during, and after PEEP test, as well as two minutes after the fluid load. The patients with an increase in CI by ≥ 15 % after fluid challenge were defined as fluid responders. The statistical analysis was performed using Mann–Whitney U-test, Wilcoxon test and ROC analysis. Data are presented as median (25th–75th percentiles). A p value < 0.05 was regarded as statistically significant.
Results: In both responders and non-responders, PEEP test resulted in a significant decrease in SAP within the first 30 seconds only (p < 0.05), while in the responders the decrement of SAP persisted until the end of the test. To 3 min after the start of PEEP test, SAP reduced by 18 (4–24) mm Hg in responders, but increased by 11 (-7–15) mm Hg in non-responders (p = 0.03).  The decrease of SAP exceeding 14 mm Hg identified fluid responders with area under the curve (AUC) 0.85, sensitivity 58% and specificity 100% (p = 0.03). During PEEP-test, PPV, MAP, and PCCI decreased without significant differences between the responders and non-responders. ROC analysis revealed that the decrease in CI ≥ 0.5 L/min/m2 from the first to the fifth minute of the test predicted responders to fluid load with AUC 0.89, sensitivity 80 % and specificity 100%  (p = 0.015).
Conclusion: In septic shock, the PEEP test can be used to predict fluid responsiveness but requires thorough hemodynamic monitoring. 

07 / Positive end expiratory pressure test can predict fluid responsiveness in patients with septic shock