Can We Agree on Cardiac Standstill?

Background: Point of Care Ultrasound (POCUS) has gained wider use in resuscitation of patients presenting with cardiac arrest. POCUS can play an important role in determining the etiology of arrest as well as being used to determine the presence or absence of mechanical activity. The REASON study demonstrated that patients with PEA or asystole without cardiac activity on POCUS are extremely unlikely to survive to hospital discharge though this study did not investigate the more important question of a survival with a good neurologic outcome. (Gaspari 2016). In light of this as well as other evidence, some have proposed that the absence of cardiac activity may be adequate to declare resuscitation futility and stop resuscitative efforts. However, there is a lack of agreement on what defines cardiac activity (see table 1). Due to this, and other factors, the level of agreement between physicians in interpretation of cardiac standstill is unknown.

Clinical Question: How much variability exists in the interpretation of cardiac standstill on POCUS amongst physicians?

Article: Hu K et al. Variability in Interpretation of Cardiac Standstill Among Physician Sonographers. Ann Emerg Med 2017. PMID: 28870394

Population: Survey respondents were residents, fellows and faculty attending ED-ICU combined ultrasound rounds at a combined conference (three academic institutions).

Outcomes:

  • Primary: Variability in interpretation of cardiac standstill among respondents
  • Secondary: Variability among subgroups based on specialty, training level and self-described POCUS experience

Design: Cross-Sectional convenience sample survey. Participants were shown 15, 6-second deidentified ultrasound clips of patients with pulseless arrest and given up to 20 seconds to answer whether the clip demonstrated cardiac standstill or not.

Excluded: Physicians that did not register demographic information (specialty, training level etc).

Definitions of Cardiac Activity:

Definitions:

Krippendorff’s alpha = assessment of interrater reliability among physicians, similar to Fleiss’s kappa but thought to be more reliable in situations with missing data points

  • Range of 0 to 1:
    • 0 = poor agreement
    • Customary to require alpha ≥0.80 for agreement, but tentative conclusions can be drawn at alpha ≥0.667
    • 1 = perfect agreement

Primary Results:

  • Physician sonographers included: 127
    • Emergency Medicine: 94 (74%)
    • Critical Care: 15 (12%)
    • Cardiology: 4 (3%)
  • Training Level
    • Resident: 80 (63%)
    • Fellow: 11 (9%)
    • Attending: 24 (19%)
  • POCUS Skill Level
    • Basic: 68 (54%)
    • Advanced: 37 (29%)
    • Expert: 7 (6%)

Critical Findings:

  • Interrater Agreement (Primary Outcome)
    • Overall: moderate agreement (alpha = 0.47)
    • Variable agreement across US clips
  • Interrater Agreement by Subgroup
    • Moderate agreement seen across specialty, training level and US experience
    • Fair agreement among critical care physicians

Strengths:

  • Study asks an important diagnostic question that has not been previously addressed
  • Majority of respondents self-reported “basic skill level” which may make study more generalizable (in comparison to just having expert level operators)

Limitations:

  • Non-random, convenience sample of participants
  • Majority of respondents self-reported “basic skill level” which may skew analysis towards disagreement

Authors Conclusions: “According to the results of our study, there appears to be considerable variability in interpretation of cardiac standstill among physician sonographers. Consensus definitions of cardiac activity and standstill would improve the quality of cardiac arrest ultrasonographic research and standardize the use of this technology at the bedside.”

Our Conclusions: We agree with the authors that this study, though limited by it’s methodology and convenience sample, demonstrates significant disagreement between providers as to the interpretation of cardiac standstill on POCUS.

Potential to Impact Current Practice: This study cannot be used to impact clinical practice but should cause some pause for thought for providers who are currently using POCUS in cardiac arrest as part of their protocol for termination based on cardiac standstill.

Bottom Line: The use of POCUS to declare death and stop resuscitation is predicated on a clear ability to establish the presence of cardiac standstill. The low level of agreement of providers likely reflects the absence of a single definition of standstill as well as calls for increased training and assessment. A unified definition of standstill would be extremely helpful not only in future research but in cardiac arrest care as well.

For More on This Topic Checkout:

References:

  1. 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

Post Peer Reviewed By: Salim R. Rezaie (Twitter: @srrezaie)

The post Can We Agree on Cardiac Standstill? appeared first on R.E.B.E.L. EM - Emergency Medicine Blog.

Turn it (All the Way) Up: Flush Rate O2 for Pre-Oxygenation

Background: There has been a lot of buzz recently about the importance of pre-oxygenation in emergency airway management.  The recent publication of the ENDAO trial [2], a randomized clinical trial of ApOx vs no ApOx also emphasized this point.  In the review article accompanying this trial by John Sackles [3] he brought up the point that most patients in this study were intubated in less than 1 – 2 minutes.  In this scenario, preoxygenation alone would likely provide an adequate oxygen reservoir to prevent hypoxemia and that apneic oxygenation would only be helpful in the patients who exhausted their oxygen reserves (i.e. prolonged intubations). Although, apneic oxygenation has recently come into favor in emergency intubation, the issue that should maybe warrant greater consideration is proper preoxygenation. The optimal method of pre-oxygenation however, is often debated: bag-valve mask (BVM), nonrebreather (NRB), or simple face mask. 

What They Did:

  • Crossover trial with healthy volunteers (each volunteer was pre-ox with all 4 experimental techniques)
  • Patients pre-oxygenated for 3 minutes with:
    • NRB at 15L/min (NRB-15)
    • NRB with flush rate (>40L/min) oxygen (NRB-Flush)
    • BVM device with oxygen at 15L/min (BVM-15)
    • Simple mask with flush rate (>40L/min) oxygen (SM-Flush)
  • Flush rate oxygen achieved by rotating the flowmeter dial counterclockwise until it could not be turned farther (Engineer not affiliated with the study evaluated flush rate flow of oxygen through flowmeters and confirmed flush rate in this study was most likely 50 – 54L/min)

Outcomes:

  • Primary Outcome: FeO2 in a single exhaled breath after 3 min pre-oxygenation (device accurate for measuring FeO2 within 1-3%)

Inclusion:

  • ED staff at Hennepin County (These patients could also have history of well-controlled chronic respiratory disease and/or facial hair)

Exclusion:

  • Symptomatic respiratory disease at participation
  • Smoking history greater than 5 pack-years
  • Pregnancy
  • Younger than 18 years

Results:

  • 26 subjects enrolled
    • Mean age: 31 years
    • Mean BMI: 24kg/m2
    • Mean baseline FeO2: 17.3% (95% CI 16.9 – 17.6%)
  • Critical Results

Strengths:

  • FeO2 measured with oxygen gas analyzer (Handi+ model R218P12) with manufacturer-reported accuracy within 1 – 3%
  • Oxygen gas analyzer calibrated with 100% oxygen before each preoxygenation trial (i.e. 4x per subject)
  • Each trial followed by 2 minutes of breathing room air and renitrogenation confirmed with FeO2 for each subject being measured between trials and returned to subject’s baseline value

Limitations:

  • Small, single center study without evaluation of patient oriented outcomes
  • Neither investigators nor subjects were blinded to preoxygenation device
  • Healthy volunteers used (i.e. results must be confirmed in a critically ill patient population)
  • Average BMI ~ 24 kg/m2 which may not generalize to the standard ED population depending on where you work
  • BVM may have had mask leaks which would underestimate the true efficacy of the BVM for pre-oxygenation
  • Study not designed to demonstrate superiority of NRB-Flush compared with NRB-15 or simple mask with flush rate oxygen
  • The value of flush rate oxygen may vary depending on the flowmeter model available at each institution

Discussion:

  • Important point made in this paper is that in a dyspneic, anxious, or agitated patient a tight mask seal may be difficult to achieve with a BVM.
  • BVM requires the use of a team member to hold the mask in place where a NRB does not (i.e. Use of NRB frees up a pair of hands)
  • Also many BVM devices lack a 1-way valve needed to achieve adequate oxygenation
  • How flush rate oxygen works in a non-sealed system: To provide near 100% inspired oxygen in a non-sealed system of oxygen delivery, the oxygen flow rate must exceed the inspiratory flow rate of the patient and the dead space nitrogen in the mask/upper airway must be flushed out between breaths.

Author Conclusion: “Preoxygenation with NRB-Flush was noninferior to BVM-15.  NRB with flush rate oxygen may be a reasonable preoxygenation method for spontaneously breathing patients undergoing emergency airway management.”

Clinical Take Home Point: Although the results of this study need to be confirmed in a critically ill patient population, it appears that flush rate oxygen via a NRB mask is non-inferior to BVM mask at 15L/min.

References:

  1. Driver BE et al. Flush Rate Oxygen for Emergency Airway Preoxygenation. Ann emerg Med 2017; 69: 1 – 6. PMID: 27522310
  2. Caputo N et al. EmergeNcy Department use of Apneic Oxygenation Versus Usual Care During Rapid Sequence Intubation: A Randomized Controlled Trial (the ENDAO Trial). Acad Emerg Med 2017. [epub ahead of print]. PMID: 28791755
  3. Sackles JC. Maintenance of Oxygenation During Raid Sequence Intubation in the Emergency Department. Acad Emerg Med 2017 [epub ahead of print]. PMID: 28791775

Post Peer Reviewed By: Anand Swaminathan (Twitter: @EMSwami)

The post Turn it (All the Way) Up: Flush Rate O2 for Pre-Oxygenation appeared first on R.E.B.E.L. EM - Emergency Medicine Blog.

Predicting Dysrhythmias After Syncope

Background: Syncope, the sudden, brief loss of consciousness followed by spontaneous, complete recovery is a common presentation to the Emergency Department (ED). It represents a significant portion of overall admissions to the hospital because while many cases are benign (i.e. vasovagal syncope), some may result from a life-threatening cause (ventricular tachydysrhythmia, myocardial ischemia, gastrointestinal bleeding etc). Admission or 24 hour observation is often sought for continuous dysrhythmia monitoring but they are infrequently found. A decision instrument directed at stratifying patients to high or low risk for a dysrhythmia causing syncope (and thus subsequent risk for a recurrent dysrhythmia) would be useful in managing patients with syncope.

Article: Thiruganasambandamoorthy V  et al. Predicting short-term risk of arrhythmia among patients with syncope: the Canadian syncope arrhythmia risk score. Acad Emerg Med 2017. PMID: 28791782

Clinical Question: Can a decision instrument be developed for predicting the 30-day risk of dysrhythmia or death after ED presentation for syncope?

Population: Patients > 16 years of age who presented with syncope within 24 hours of the event. Syncope defined as a brief loss of consciousness (< 5 minutes)

Outcomes:

Primary: Death or dysrhythmia (including procedural interventions for dysrhythmia) within 30 days.

Design: Multicenter prospective cohort study

Excluded: Prolonged loss of consciousness (> 5 minutes), change in mental status after syncope, witnessed seizure, head trauma causing loss of consciousness, patients unable to provide proper history due to alcohol or drug intoxication or patients with a language barrier.

Primary Results:

  • Enrolled patients: 5,358
  • Incomplete outcomes assessment: 348/5358 (6.5%)
  • Inter-rater reliability assessment
    • 207 patients (4.1%)
    • Kappa = 0.89 (95% CI: 0.79-0.98)
  • Admission rate: 9.5%
  • Comorbid Conditions
    • Hypertension (31.6%)
    • Diabetes (10.0%)
    • CAD (11.6%)
    • Congestive Heart Failure (3.6%)

Critical Findings:

  • Primary outcome at 30-days (death or dysrhythmia)
    • 2.1% (106/5010) 95% CI 1.7 – 2.5%)
    • 0.9% (45/5010) suffered primary outcome after discharge from the hospital
    • 0.57% (29/5010) had pacemaker placement performed
  • Decision tool derivation
    • Researchers identified 39 candidate predictors
    • Final model with eight predictors
      • Three clinical evaluation predictors, three investigation predictors and two final ED diagnoses
      • Vasovagal predisposition (warm-crowded place, prolonged standing, fear, emotion or pain)
  • Decision tool score
    • Ranges from -2 to +8 (lower score = lower risk)
    • Threshold Scores
      • Score > 0
        • Sensitivity: 97.1% (95% CI: 91.6 – 99.4%)
        • Specificity: 53.4% (95% CI: 52.0 – 54.9%)
        • (+) LR = 2.0
        • (-) LR = 0.05
      • Score > 6
        • Sensitivity: 9.8%
        • Specificity: 99.6%
        • (+) LR = 245
        • (-) LR = 0.91

Strengths:

  • Large, prospective study addressing an important clinical question that can potentially affect millions of patients
  • First study of it’s kind to attempt to identify risk factors for short term development of dysrhythmia or death
  • Primary outcome is patient centered
  • Used the validated, European Society of Cardiology (ESC) guidelines for arriving at the final ED diagnosis
  • Researchers worked extensively to track down patients to gauge outcomes
    • Structured review of all medical records related to index and subsequent ED visits, hospitalizations and hospital record deaths
    • Scripted telephone follow-up at 30-days
    • Reviewed health records at all local adult hospitals in Ontario and administrative health database (NetCare)
    • Ontario coroner’s office records reviewed

Limitations:

  • Although multi-center, was single system and single country
  • Primary outcome was a composite where the outcomes weren’t of equivalent impact on the patient (death and dysrhythmia)
  • Requires external validation that includes sites in various countries prior to implementation
  • 20% of eligible patients were not enrolled in the study
  • Patients were overall young (mean age 53.4 years) and relatively healthy limiting application of the decision tool
  • “Vasovagal Predisposition” is subjective and may be difficult to reliably gauge. This factor will require further study for inter-rater reliability
  • Final diagnosis of vasovagal and cardiac syncope subjective (though inter-observer agreement in this study was high)
  • Vasovagal predisposition and vasovagal syncope as the final diagnosis are likely interdependent factors and not independent items
  • Incomplete data on parts of decision tool (i.e. only 54.2% of patients had troponin performed)
  • Troponin assays were different at different institutions and difficult to compare directly

Authors Conclusions:

“The Canadian Syncope Arrhythmia Risk Score can improve patient safety by identification of those at-risk for arrhythmias and aid in acute management decisions. Once validated, the score can identify low-risk patients who will require no further investigations.”

Our Conclusions:

This prospectively derived decision instrument may be helpful in stratifying syncope patients to high or low risk for dysrhythmia but requires external validation prior to consideration for implementation.

Potential to Impact Current Practice:

Because dysrhythmia is a common reason for admission to hospital and further workup of patients presenting with syncope, this decision tool has the potential to be used to support a decision of discharge and follow up and avoid low yield inpatient evaluations if externally validated.

Clinical Bottom Line:

The prospective derivation of this decision instrument is a positive step towards risk stratification of syncope patients for subsequent dysrhythmias. It is critical to remember that there are numerous other causes of syncope (ectopic pregnancy, PE, ACS, aortic dissection, GI bleeding etc) that should be considered in patients as well. Additionally, because the population in this study was relatively young and healthy, evaluation of a higher risk cohort would be useful in future studies.

For More on This Topic Checkout:

Post Peer Reviewed By: Salim Rezaie (Twitter: @srrezaie)

The post Predicting Dysrhythmias After Syncope appeared first on R.E.B.E.L. EM - Emergency Medicine Blog.