Ultrasound in Cardiac Arrest

Reason Byte


Emergency Department Point-of-care Ultrasound in Out-of-Hospital and in-ED Cardiac Arrest

Resuscitation September 2016 - Pubmed Link

Additional Data from presentation at CAEP 2015



Take Home Points

1. Cardiac activity on initial ultrasound in PEA or Asystole arrests is associated with increased survival to hospital admission. OR 3.6 BUT it is not sufficiently specific or sensitive.

2. Patient with no cardiac activity and initial rhythm of asystole have low chance of survival to hospital discharge.

3. Ultrasound may identify causes of arrest leading to intervention. Higher survival in group who received pericardiocentesis.


Other Awesome FOAMed on this article

Ultrasound Podcast Part 1 and Part 2

EM Nerd

St. Emlyn's

EM in Focus

R.E.B.E.L. EM


Background

Bedside ultrasound can be useful during cardiac arrest to determine if there is cardiac activity. While this is intuitive, the exact manner in which is should be incorporated into cardiac arrest algorithms is unclear. If there is no cardiac activity seen, does this mean that further resuscitation is futile? If there is cardiac activity, is it worth while to continue resuscitation until there is not? Although many people use cardiac ultrasound for decision making during arrests, there was not too much evidence for it prior to this study.


Question

Is sonographic cardiac activity associated with survival (or nonsurvival) in cardiac arrest?


Population

20 centers, US and Canada

Inclusion:

  • Out-of-hospital cardiac arrest or In-ED cardiac arrest

  • Asystole or Pulseless Electrical Activity (PEA)

  • Had ultrasound performed during resuscitation

Exclusion:

  • Traumatic arrest

  • Resuscitation <5 minutes

  • If resuscitation was stopped after initial ultrasound

  • If resuscitation was stopped due to Do Not Resuscitate order


Who did the scans?

Emergency Physicians credentialed in bedside US


Design

Prospective, observational

Multivariate regression models


Primary Outcome - survival to hospital admission

Secondary Outcomes - survival to hospital discharge and ROSC (return of spontaneous circulation)


Intervention

Advanced Cardiac Life Support (ACLS) was employed

US performed at beginning and end of cardiac arrest, during pauses in CPR

Treating physician were not blinded to results

Resuscitation was continued for at least one round of medications after first ultrasound was performed


Scan

Probe unspecified - likely phased array or curvilinear array probe

Subxiphoid or parasternal long axis views of heart

“Cardiac Motion” defined as “any visible movement of the myocardium excluding movement of blood within the cardiac chambers or isolated valve movement.”


Results

953 enrolled, exclusions:

  • Resus ended after US (106)

  • DNR (8)

  • US not interpretable (1)

  • Incomplete timing data (3)

  • Resus ended before any ACLS medications given (42)

N = 793 analyzed over 3.5 years.


The Population

Overall

208 patients (26.2%) survived the initial resuscitation (return of spontaneous circulation, ROSC)

114 (14.4%) survived to hospital admission

13 (1.6%) surviving to hospital discharge

263 (33%) had cardiac activity on initial US (54% of PEA patients and 10% of Asystole patients)

Of those with cardiac activity:

ROSC 134 (51.0%)

Survival to admission 76 (28.9%)

Survival to discharge 10 (3.8%) - this is still very low ( but better compared to 0.6 without US cardiac activity)


Primary Outcome

Cardiac activity on ultrasound associated with survival to hospital admission - OR 3.6 (2.2 - 5.9). AUC = 0.762

This was the variable with the strongest relationship to survival at all end points.

BUT, another way to look at this is the test characteristics of ultrasound for detecting those who would not survive

-In asystole, US was 90% sensitive, 17% specific for survival to admission

-In PEA, US was 53% sensitive, 80% specific for survival to admission


Secondary Outcomes

Asystole + no cardiac activity, characteristics for nonsurvival to ROSC - Sens 91%, Spec 19%.

Asystole + no cardiac activity, characteristics for nonsurvival to hospital discharge - Sens 90%, Spec 0%

PEA + no cardiac activity, characteristics for nonsurvival to ROSC - Sens 60%, Spec 76 %

PEA + no cardiac activity, characteristics for nonsurvival to discharge - Sens 47%, Spec 91%

Cardiac activity on ultrasound associated with ROSC - 2.8 (1.9 - 4.2) AUC = 0.803

Cardiac activity on ultrasound associated with survival to hospital discharge - OR 5.7 (1.5 - 21.9) AUC = 0.825


Other Findings

Other variables associated with increased survival are various end points = PEA, Bystander witnessed, Shockable rhythm.

Patients who arrested in the emergency department had more survival to admission but not to hospital discharge.


The lack of cardiac activity on initial US was strongly associated with non-survival, BUT 3 patients (0.6%) of patients with no cardiac activity survived to discharge

  • 2 were initially Vfib

  • 1 was Asystole


Patients with pericardial effusion (n = 13) who underwent pericardiocentesis demonstrated higher survival rates (15.4%) compared to all others (1.3%)

Others have similarly advocated changing management based on echo during PEA

ERCAST - PEA made simple

ED ECMO - Pseudo PEA with Littmann article and Prosen article


Kappa 0.63 after reviewed by second emergency physician


Some patients lost or gained cardiac activity

-No activity to positive activity → 11.1%

-Positive activity to no activity → 11.7%


Average time recording ultrasound images was 4-5 seconds.


Limitations

Not blinded - ultrasound results could influence management.

Patient may have been excluded based on having no activity on initial ultrasound which would exclude patient who may have survived based on the ultimate data. However, as noted by authors, if these were included it would have biased the results as well because they would likely have still stopped resuscitation based on ultrasound.

Ultrasound could have affected rhythm interpretation (did not give data on how often ultrasound findings changed what was thought to be the rhythm based on monitor)

No assessment of cognitive function of survivors.

Survival to discharge is arguably more important end-point, but that was a secondary outcome.

Confounding differences between groups. More bystander witnessed arrests and bystander CPR in group with ultrasound activity. Shorter downtime and longer resuscitations for patients with cardiac activity.

Selection bias of sites. Since the resuscitations were not protocolized, there could have been differences.

Unclear how much of a role the longer resuscitation could have played (although no association found on multivariate regression)

We don’t know how useful ultrasound would be in Vfib/Vtach arrests based on this data.

Will cardiac activity on US lead to longer resuscitations or admission than necessary for no change in neurologically intact outcomes?


Take Home Points

1. Cardiac activity on initial ultrasound in PEA or Asystole arrests is associated with increased survival to hospital admission. OR 3.6 BUT it is not sufficiently specific or sensitive.

2. Patient with no cardiac activity and initial rhythm of asystole have low chance of survival to hospital discharge.

3. Ultrasound may identify causes of arrest leading to intervention. Higher survival in group who received pericardiocentesis.


Our score

5 Probes


Published on 12/19/16 01:00 AM
comments (0)