Impact of Point-of-Care Ultrasound in the Emergency Department on Care Processes and Outcomes in Critically Ill Nontraumatic Patients.

Abstract

Retrospective cohort study of critically ill emergency department patients in two academic emergency departments. All emergency department patients admitted to the intensive care unit or that die in the emergency department were entered prospectively into a registry. Two academic emergency departments. All adult (> 18 years old) non-trauma patients with hemodynamic instability [shock index (heart rate/systolic blood pressure) > 0.6] between November 1, 2013-October 31, 2016, were included. Cohorts were assigned as follows: no POCUS (cohort 1), POCUS prior to a key intervention (cohort 2), and POCUS after a key intervention (cohort 3). A key intervention was either a fluid bolus or vasoactive drug initiation. Multivariable logistic regression was used to evaluate the association between POCUS use and the primary outcome of in-hospital mortality. We conducted several sensitivity analyses including propensity score matching and inverse-probability-weighted regression-adjustment along with multiple imputation to account for non-random assignment of POCUS as well as bias due to missing data. Of the 7,734 eligible patients, 2,293 patients were excluded. The remaining 5,441 patients were included in the analysis: 4165 in Cohort 1, 614 in Cohort 2, and 662 in Cohort 3. Mortality was 22%, 29%, and 26%, respectively (p < 0.001). POCUS prior to an intervention was associated with an adjusted odds ratio for death of 1.41 (95% CI, 1.12-1.76) compared to no POCUS. The sensitivity analyses showed an absolute increased mortality of +0.05 (95% CI, 0.02-0.09) for cohort 2 compared to 1. POCUS use prior to interventions appears to be associated with care delays and increased in-hospital mortality compared to critically ill patients with no POCUS. Further explorations of the impact of POCUS in the emergency department appear warranted.