SESSION TITLE: Imaging Posters
SESSION TYPE: Original Investigation Poster
PRESENTED ON: Wednesday, October 28, 2015 at 01:30 PM - 02:30 PM
PURPOSE: Predicting fluid responsiveness (FR) is an essential but challenging component of managing critically ill patients with shock. Point-of-care ultrasound (POCUS) trained physicians increasingly advocate for the use of inferior vena cava (IVC) parameters to guide fluid administration decisions despite a paucity of studies supporting this practice. We compared the diagnostic accuracy of IVC parameters to predict FR with the results of a passive leg raise (PLR) test measured by bioreactance.
METHODS: We performed a prospective investigation that included consecutive patients with shock newly admitted to our Medical Intensive Care Unit (MICU) while a POCUS trained critical care fellow was on duty. Upon enrollment, the admitting ICU physician performed a POCUS exam which assessed multiple IVC parameters; 1) maximal diameter (IVCmax), 2) minimum diameter (IVCmin) and 3) collapsibility index (IVCCI = (IVCmax - IVCmin/IVCmax). Immediately thereafter, FR was assessed using a bioreactance monitor (Cheetah Medical, Newton Center, MA). FR was defined as an increase of stroke volume index >10% after a PLR. The ability of IVC parameters to predict FR was assessed through receiver operating curves (ROC) analysis using bioreactance derived FR as the criterion standard.
RESULTS: 59 patients admitted during the study period met inclusion criteria and had a fellow performed POCUS. 42 of these patients underwent PLR testing and were thus included in the analysis. The patients received an average of 2,220 ml of fluid and 52% required vasopressors. 33% of patients were deemed fluid responders by PLR testing. Area under the ROC curve for IVCCI was .59 (CI .40 - .76), for IVCmax .34 (CI .17, .50) and IVCmin = .36 (CI .20, .53).
CONCLUSIONS: IVC parameters are poor predictors of FR in critically ill shock patients.
CLINICAL IMPLICATIONS: Physicians should limit the use of POCUS derived IVC parameters to guide fluid administration decisions.
DISCLOSURE: The following authors have nothing to disclose: Daniel Fein, Christopher Jordan, Samuel Acquah, Pierre Kory
No Product/Research Disclosure Information