Emergency department point-of-care ultrasonography improves time to pericardiocentesis for clinically significant effusions
Clinical and Experimental Emergency Medicine September 2017 - Pubmed Link
Take Home Points
1. This study was limited but did not provide definitive evidence that point of care ultrasound improves time to intervention or mortality in patients with significant pericardial effusion.
2. Because classic signs and symptoms are often not present, maintain a low threshold to perform cardiac ultrasound to evaluate for pericardial effusion and tamponade.
Background
Point of care ultrasound thrives on identifying an actionable diagnosis in sick patients. That’s why cardiac tamponade is a great example of how POCUS can be useful. Your physical exam and history are not as accurate as you would like so being able to use ultrasound can help get the patient to the life-saving procedure they need. That’s the basis of this study. The authors wanted to see if using point of care ultrasound to identify significant pericardial effusions early would change patient centered outcomes such as time to intervention, length of stay, or mortality.
Questions
In patients with cardiac tamponade or a large pericardial effusion, does using POCUS in the emergency department affect time to procedure and outcomes?
Population
Single center in Israel
Data collected October 2007 to February 2012
Inclusion:
Presented to the emergency department and had hospital discharge diagnosis of cardiac tamponade or pericardial effusion
Clinically significant effusion. Defined as effusion that would lead to referral for pericardiocentesis or operative repair
Exclusion:
- Trauma
Design
Retrospective cohort study, single center academic ED
Patient were identified from a hospital database (searching using an ICD code) who had a large pericardial effusion or tamponade. These were then split into two groups - POCUS (those who received a point of care ultrasound in the emergency department) and non-POCUS (those who had not).
They recorded information from chart including demographics, medical background, ECG, chest xray, ultrasound, official echocardiogram, patient outcomes, type of procedures, analysis of pericardial fluid, length of hospital stay, status at hospital discharge.
Door to pericariocentesis time recorded from a cardiology database system.
Outcomes:
Multiple Primary Outcomes (sad face): Door-to-pericardiocentesis time, Amount of pericardiocentesis fluid, and Length of hospital stay
Secondary Outcome: Mortality
Intervention
The intervention was receiving a point of care ultrasound in the emergency department.
Who did the ultrasounds?
6 physicians - 5 attendings and 1 fellow
- Trained in emergency ultrasound according to curriculum of the Israeli Association of Emergency Medicine
The Scan
Phased array probe
Standard 4 view focused cardiac ultrasound
5 Minute Sono - Pericardial Effusion
5 Minute Sono - Pericardiocentesis
Results
Patients
Over 4.5 years, there were 500,000 ED visits and 1000 POCUS studies.
73 total patients with significant pericardial effusions
18 of these patients were diagnosed with POCUS
55 were diagnosed by other means
Patient characteristics:
POCUS group had lower systolic BP (117.6 vs 134.8)
POCUS group had more syncope (3 vs 0)
POCUS had less infiltrates or effusions on CXR (17.6% vs 71.6%)
The majority of CXRs in both groups showed cardiomegaly (70.6% POCUS, 86.4% non-POCUS).
Malignancy was the most common cause of effusion in both groups
82.6% underwent a pericardiocentesis or surgery
Primary Outcomes
Door to Pericardiocentesis → No statistically significant difference
POCUS: 11.3 hours (± 13.8)
Non-POCUS: 70.2 hours (± 126.2)
P = 0.055
Amount of Fluid Drained → No statistically significant difference
POCUS: 660.6 mLs (± 339.9)
Non-POCUS: 825.8 mLs (± 409.2)
P = 0.139
Length of Hospital Stay → No statistically significant difference
POCUS: 5.0 days (±0.6)
Non-POCUS: 7.0 days (±0.9)
P = 0.222
Secondary Outcomes
Death → No statistically significant difference
POCUS: 1 death (5.6%)
Non-POCUS: 4 deaths (7.3%)
P = 0.802
Limitations
Study performed in Israel which may have different patient population and practice environment, limiting external validity.
Retrospective design is not ideal. Many patients could have been missed. Although RCT may be best evidence, it would be unethical to randomize a patient with possible tamponade to not get an immediate ultrasound. One possibility would be prospective study of patients with possible tamponade who all get POCUS compared to matched controls.
Population may have been too small to draw conclusions from. Small number of POCUS studies, only 18 in this study.
No strict criteria for “significant” effusion. Do all effusions need pericardiocentesis? No. The decision to perform a procedure is multi-factorial and can be somewhat complex, this limits the conclusions drawn from these findings.
Not very sick population - very few deaths, very few hypotensive. Ultrasound is likely going to make the most difference in the most critically ill where time matters most.
Selection bias - those who got POCUS studies likely were thought to need them. You look because you think you might find something.
Discussion
The main question is - when to perform a scan for a pericardial effusion? Tamponade is a diagnosis you can't miss. Unfortunately, as this study shows, clinical diagnosis is difficult. Many of these patients were not hypotensive, less than half presented with dyspnea, and many even had an infiltrate on their chest xray! Another cool tamponade study confirmed that not many of these patients will present with Beck's triad or obvious clinical signs pointing towards this diagnosis. That is why many suggest performing a cardiac ultrasound in undifferentiated dyspnea, undifferentiated hypotension, and cardiac arrest. You can certainly find other important causes for the presentation as well. Some people might say that you shouldn't apply a test for a relatively rare disease to population with a common complaint (such as dyspnea). I think the risk of the study (taking time and resources) is low compared to the potential gains of finding a patient who needs a life-saving intervention. What do you think? When should we be doing cardiac ultrasounds for suspected pericardial effusion or tamponade?
Take Home Points
1. This study was limited but did not provide definitive evidence that point of care ultrasound improves time to intervention or mortality in patients with significant pericardial effusion.
2. Because classic signs and symptoms are often not present, maintain a low threshold to perform cardiac ultrasound to evaluate for pericardial effusion and tamponade.
Our score
