• 
• 

Three Part Question

In [patients in cardiac arrest with cardiac standstill], [does any variability exists] [in the interpretation of POCUS amongst physicians]?

Clinical Scenario

On-going CPR on a 50 year-old male is occurring in your ED. He remains pulseless after 20 minutes and you doubt about the cardiac activity on the POCUS examination. You wonder what your other colleague would say.

Search Strategy

Search performed on January 3rd 2018

A. MEDLINE using the PubMed interface (1900 � December 2017)
1- Ultrasound [Mesh term] : 409 587 articles
2- �point of care Ultrasound� [ti/ab] : 711 articles
3- �sonograph* [ti/ab] : 50 141 articles
4- �cardiac arrest� [Mesh term] : 41 698 articles
5- �cardiac standstill� [ti/ab] : 179 articles
6- 1 OR 2 OR 3 : 422 228 articles
7- 4 OR 5 : 41 814 articles
8- 6 AND 7 : 1392 articles
9- Emergency [Mesh term] : 37 876 articles
10 - Emergency [ti/ab] : 211 355 articles
11 - 9 OR 10 : 229 683 articles
12 � 8 AND 11 : 123 articles

B. Embase using the Elsevier interface (1966� December 2017)
1- Ultrasound [Emtree] : 149 982 articles
2- �bedside ultrasound� [ti/ab] : 755 articles
3- �heart arrest� [Emtree] : 77 775 articles
4- �cardiac arrest� OR �cardiac standstill� [ti/ab] : 35 705 articles
5- 1 OR 2 : 150 348 articles
6- 3 OR 4 : 84 157 articles
7- Emergency ward [Emtree] : 103 398 articles
8- Emergency [ti/ab] : 257 949 articles
9- 7 OR 8 : 284 261 articles
10 � 5 AND 6 AND 9 : 105 articles

C. The website www.clinicaltrials.gov was searched for ongoing trials on the subject. No trials were found.

D. No BestBETs or critical appraisals were found on this specific topic. However, one Best BET (Dallon 2011) was published and measured the prognostic accuracy of bedside ultrasound in adults with cardiac arrest. Since the publication, the REASON trial (Gaspari 2016) has been published in 2016 as well as another systematic review. (Blyth 2012)

E. The Cochrane Library was searched for reviews on the subject. No reviews were found.

F. Google Scholar & Grey literature
No additional articles were found while looking at cross-references.

Search Outcome

Out of the 123 articles cited using the PubMed interface, only one was found to measure the interpretation of POCUS amongst physicians as a primary outcome. Of note, many of the studies evaluating prognostic accuracy of a bedside POCUS in cardiac arrest discussed about variability of interpretation but these results were mostly shown as part of the discussion, rather than a secondary outcome. We have included one trial (Gaspari 2016) since it is the only large multi-center study and the methods are clearly described for agreement between interpretations of US results.

The references of relevant papers were also scanned for any missed papers.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Hu et al 2017 USA	Convenience sample of 127 total participants, including faculty, fellows, and resident physicians specializing in emergency medicine, critical care, and cardiology. Survey respondents were allotted 20 seconds per slide to determine whether each of 15 video clips of patients in cardiac arrest were standstill or not.	Cross-Sectional convenience sample survey.	Primary: Variability in interpretation of cardiac standstill among respondents	Overall: moderate agreement (alpha = 0.47) Variable agreement across US clips, especially poor agreement with valve flutter, ventilation, weak contractions and profound bradycardia	Non-random, convenience sample of participants. Response bias - the people that participated in this study may have had special interest in ultrasound. Majority of respondents self-reported “basic skill level” which may skew analysis towards disagreement. However, no difference was noted when testing the higher level users. Most participants were residents (63%). Did not assess intra-rater reliability. Participants were not isolated from one another, so discussion among them may have occurred. It is also possible that in clips may have projected differently in the conference rooms than they would have appeared to point-of-care users when they were obtained.
			Secondary: Variability among subgroups based on specialty, training level and self-described POCUS experience	Moderate agreement seen across specialty, training level and US experience Fair agreement among critical care physicians
Gaspari et al. 2016 USA & Canada	Patients with non-traumatic, OHCA or in-ED arrest with PEA or asystole. N = 953 Exclusion: DNR, Resuscitation lasted < 5 minutes or Resuscitation not continued after initial ultrasound	Multicenter, Non-randomized, prospective, observational study at 20 hospitals in the USA or Canada	Primary : Survival to Hospital Admission	Cardiac activity on US: Increased Survival to Hospital Admission (OR 3.6; 2.2 – 5.9)	The primary endpoint was not patient centered. The measured outcomes were not linked to our three-part questions.
			Secondary: Survival to Hospital Discharge, ROSC	Increased Survival to Hospital Discharge (OR 5.7; 1.5 – 21.9) No Cardiac Activity on US: 0.6% still survived to discharge Agreement between initial US interpretation and review was substantial (k = 0.63).

Comment(s)

Point of Care Ultrasound (POCUS) is now widely used in the ED for cardiac arrest resuscitation purposes. A recent meta-analysis (Blyth 2012) concluded that cardiac standstill, on bedside echocardiography performed during cardiac arrest, is an extremely poor prognostic indicator. More recently, the REASON trial demonstrated that patients with PEA or asystole without cardiac activity on POCUS are extremely unlikely to survive to hospital discharge. (Gaspari 2016). However, even though cardiac standstill is part of basic learning skills in POCUS courses, the definition of cardiac activity varies in the literature, ranging from any organized contractile activity to any detected motion within the heart (atrial, valvular or ventricular) (Aichinger 2012). Therefore, the level of variability in interpretation of cardiac standstill between physicians is unknown. Only one study describing this issue was found and concluded there is only moderate agreement in providers determining cardiac standstill and especially poor agreement with valve flutter, ventilation, weak contractions and profound bradycardia. Of note, this study showed 6 seconds-clips in a classroom, so variability of interpretation while getting the images in an on-going CPR situation is unknown and one could think it may be higher with surrounding disturbing factors and the challenge of getting a clear image.

Clinical Bottom Line

Many studies showed significant decrease of survival when cardiac standstill is present in cardiac arrest. However, there appears to be considerable variability in interpretation of cardiac standstill among physician sonographers. The moderate level of agreement might reflect the absence of a consensus definition of cardiac activity. Therefore, we need more research to achieve a consensus definition so we can use cardiac standstill to make clinical decisions.

References

1. Hu K et al Variability in Interpretation of Cardiac Standstill Among Physician Sonographers Ann Emerg Med 2017 PMID: 28870394
2. Gaspari R et al. Emergency Department Point-Of-Care Ultrasound in Out-Of-Hospital and in-ED Cardiac Arrest. Resuscitation 2016 109: 33 – 39. PMID: 27693280