Danielle Parker, DO | EM PGY3
The Idea: To determine if the use of physician-performed immediate vs. delayed goal-directed ultrasound protocol in the management of non-traumatic, symptomatic, hypotensive ER patients led to increased diagnosis accuracy
The Study: This was a randomized, control trial performed between July 2002 and September 2003 in the emergency department of Carolinas Medical Center which is an urban tertiary care center with over 100,000 patient visits per year. 184 patients were randomized into two groups: group 1 received immediate (time 0) goal-directed ultrasound in addition to standard care (history, physical, lab work, x ray); group 2 received a delayed goal-directed ultrasound 15 mins after receiving standard care.
Inclusion criteria included: nontrauma emergency department patients age of 17 or older; initial vitals consistent with shock (systolic blood pressure < 100 mmHg systolic or shock index >1) and agreement of two independent observers for at least one sign and one symptom of shock (inadequate tissue perfusion). Once a patient was deemed to be eligible for the study, attending ER physicians and third year ER residents received a random numbered, sealed envelope that contained the randomization assignment (group 1 (immediate US) or 2 (delayed US)) and data collection sheets. Both groups of patients received standard history and physical, lab work, blood gases and xrays. Group 1 received an immediate ultrasound (at time 0) which included the following views: subcostal to assess for RV diastolic collapse, IVC to assess intravascular volume status, parasternal long axis to assess LV function and pericardial effusion, apical 4 chamber to compare ventricle size, hepatorenal recess to assess for free intraperitoneal fluid, pelvis to assess for intraperitoneal fluid and aorta to assess for AAA. Goal-directed US was performed by either a board certified ER attending or a third year ER resident. At time 15 mins examining physicians in both groups completed a data sheet with their differential diagnoses based on the information they had acquired. Group 2 participants then received the same goal-directed US after 15 minutes (time = 15). At 30 mins both groups completed the data sheets again.
The Findings:
Outcomes included the number of viable diagnoses at 15 mins and the rank of their likelihood of occurrence at both 15 and 30 mins. At 15 minutes, group 1 had a median of 4 differential diagnoses while group 2 had a median of 9 (median difference= 5; 95% CI, 4 - 6; Mann-Whitney U test, p < .0001). At 30 minutes group 1 still had a median of 4 diagnoses while group 2 had a revised median of 3 diagnoses (Mann-Whitney U test, p < .4463). At time 15 mins physicians in group one indicated the correct diagnosis in 80% (95% CI, 70 – 87%) of their patients while group 2 was only 50% (95% CI, 40-60%). There was no difference in mortality between patients in group 1 and 2.
The Takeaway:
It was determined that the incorporation of goal-directed ultrasound in the evaluation of nontraumatic, symptomatic, undifferentiated hypotension helped physicians determine fewer viable diagnoses and causes. Goal-directed ultrasound results in a more accurate impression of patients and final diagnosis.