(Last Updated On: June 24, 2019)

Does Point-of-Care Ultrasonography Improve Clinical Outcomes in Emergency Department Patients With Undifferentiated Hypotension? An International Randomized Controlled Trial From the SHoC-ED Investigators

Atkinson et al. Annals of Emergency Medicine. October 2018. [paper]

Why I Chose This Study

In my brief time as a physician I have blindly accepted point-of-care ultrasonography (POCUS) as a useful tool in the resuscitation of sick patients. So, when I saw an article that questioned its dogmatic use in diagnosing and treating early undifferentiated nontraumatic hypotension, I had to read on.


It has been shown that use of POCUS in certain resuscitation situations (such as traumatic shock or volume status after early resuscitation) can lead to improved patient centered outcomes. However, there is a paucity of data regarding patient-centered outcomes for its use in early, nontraumatic, undifferentiated hypotension.

Research Question

What is the “effect of a standardized point-of-care ultrasonography protocol on clinical outcomes for selected patients presenting to the ED with undifferentiated hypotension?”

Study Design

International (North America and South Africa), multicenter, randomized controlled trial of patients who presented to an emergency department (ED) with undifferentiated nontraumatic shock. All the centers were academic sites staffed by accredited emergency medicine physicians with active POCUS programs.

Scans were performed by physicians who had ‘demonstrated training and competency in point-of-care ultrasonography’ as determined by PIs at the respective sites. Protocolized scan consisted of cardiac, lung, IVC, abdominal aorta, abdominal, and pelvic views. Specific questions included evaluating the ‘pump’ (left/right size and contractility as well as pericardial fluid), ‘tank’ (IVC collapsibility and pleural/peritoneal/pelvic fluid), and ‘pipes’ (evidence of aortic disease or central signs of thromboembolism).

After reviewing inclusion/exclusion criteria and consent, patients were randomized to control v. POCUS intervention. In the intervention group, physicians performed their normal initial clinical assessment and then completed the protocolized scans within the first 60 minutes of the patient visit. In both groups, a repeat clinical assessment was performed at 60 minutes.


Inclusion Criteria: Adult patients (19 yo) with sustained SBP <100 or a shock index >1.0 and SBP <120 identified at triage.

Exclusion Criteria:

  • Pregnancy
  • CPR/defibrillation/emergency pacing/placement of VAD before screening
  • Significant trauma in the last 24 hours
  • 12 lead ECG diagnostic of AMI
  • Evident clear mechanism of cause for hypotension/shock that was immediately identifiable by treating physician (i.e. UGI hemorrhage)
  • Previously known diagnosis (i.e. transfer from OSH)
  • Vagal episode
  • Non-pathologic hypotension (i.e. young and healthy)


Primary Outcome: survival to 30 days or hospital discharge.

Secondary Outcomes: volume of IV fluid administered in ED, use of inotropes, rate of CT scanning, hospital and ICU admission rates, and lengths of stay.


Categorical data is presented as percentages and binomial confidence intervals, and medians and interquartile ranges for continuous data. Results are presented in terms of differences in proportion or median between experimental and control groups, along with a binomial confidence interval for the observed differences.


There was no difference in any primary or secondary outcome and enrollment was stopped at 270 patients due to futility.

Hocus Pocus © Disney


While it is interesting that this study showed no patient-centered benefit with the use of standardized ultrasound protocols for patients presenting to the ED with undifferentiated nontraumatic shock or hypotension, I’m not sure it’s a huge surprise. As one can see from the most common final diagnoses (sepsis, severe dehydration, abdominal ‘inflammation,’ and LV failure/AMI) the majority get better with crystalloids +/- antibiotics or, for LV failure/AMI, are often easily diagnosed clinically and with an ECG. The few conditions where POCUS may make a difference (cardiac tamponade, tension pneumothorax, aortic dissection/bleeding) are too rare to affect the outcome in this underpowered study.

Currently, my practice pattern in undifferentiated hypotension is to start with my clinical assessment. If the patient has a rip-roaring fever with hypotension, or they have been vomiting for 3 days straight, I feel pretty comfortable with an empiric initial crystalloid bolus as my resuscitation move of choice. However, if my clinical assessment elucidates a diagnosis that can be made with an US probe, for instance they complain of abdominal pain, they have no breath sounds on one side, terrible JVD, or pleuritic chest pain, I’m going to grab the ultrasound. Alternatively, if they do not get better despite my resuscitative efforts, I’m going to use the ultrasound to guide my further resuscitation. And, I feel justified doing it this way. I don’t think the results of this paper are going to change my current practice.

Gabe Hoffman is an IM/EM resident, class of 2019. He will be starting fellowship in critical care in July at the University of Pittsburgh.

EM/IM Sessions are reviewed in journal club style by the current attendings and residents, as well as alumni of the UIC IM/EM program prior to publication. This post was specifically reviewed by David Snow, MD, former APD of the EM program, current PD at Loyola. Elspeth Pearce, MD, Editor.