[display_podcast]

Date: June 12th, 2018

Reference: DeMeester S et al. Implementation of a Novel Algorithm to Decrease Unnecessary Hospitalizations in Patients Presenting to a Community Emergency Department With Atrial Fibrillation. AEM June 2018

Guest Skeptic: Dr. Morgenstern is an emergency physician and the Director of Simulation Education at Markham Stouffville Hospital in Ontario. He is the creator of the excellent #FOAMed project called First10EM.com and an amazing photographer.  

Case: A 62-year-old Canadian is on vacation in up-state Michigan, and after a celebratory evening, presents to your emergency department with palpitations. “I’ve had atrial fibrillation a number of time before. Usually they just shock me and send me home.” Local practice is usually to treat rapid atrial fibrillation with a calcium channel blocker infusion and admit to hospital. As the conversation progresses, you wonder whether it might be safe to discharge some atrial fibrillation patients home for outpatient follow-up.

Background: Atrial fibrillation, rate control vs. rhythm control. This is a debate that has gone on for many years. It is like normal saline vs. Ringer’s lactate for fluid resuscitation, steroids vs. no steroids for sepsis, or Coke vs. Pepsi.

Atrial fibrillation is one of the most common dysrhythmias and patients often present to the emergency department with increased heart rates, chest pain and weakness among other presentations. The debate has been going on for years as to which is the best strategy to address these patients, rate or rhythm control.

In patients with chronic atrial fibrillation or unknown time of onset and a rapid ventricular response, rate control and consideration and initiation of anticoagulation therapy are the standard emergency department approach.

Both beta-blockers and calcium channel blockers are commonly used for rate control in the emergency department. SGEM#133 reviewed a study by Fromm C et al. comparing diltiazem vs. metoprolol in the management of atrial fibrillation or flutter with rapid ventricular rate in the emergency department (J Emerg Med 2015).

The SGEM bottom line was that the best available evidence shows that diltiazem will achieve more rapid rate control in patients with atrial fibrillation than metoprolol (NNT 2).

Dr. Ian Stiell and colleagues published an article in 2011 in Annals of EM looking at variation in recent-onset atrial fibrillation management in Canada and found a lot of variability. Rhythm control was selected in 42-85% of patients across hospitals and electricity was chosen as the primary strategy for rhythm control in 7-69%.

In the USA there is a fear of cardioverting someone in atrial fibrillation because it could cause them to throw a clot. What often happens is that most patients are rate controlled, admitted and cardiology is left to sort it out.

In Canada we do tend to cardiovert patients with recent onset atrial fibrillation. SGEM#88 looked at the effectiveness and safety of the Ottawa Aggressive Protocol to perform rapid cardioversion and discharge of patients with these arrhythmias.

The SGEM bottom line from that episode was that The Ottawa Aggressive Protocol appears to be highly effective in converting patients with recent onset atrial fibrillation or flutter back to sinus rhythm. The vast majority of patients (97%) were discharged home from the emergency department with 93% in normal sinus rhythm.

We are not going to solve the rate vs. rhythm debate on this show but we are going to address the admission rate observed in the USA.


Clinical Question: Can an emergency department algorithm for atrial fibrillation management decrease the number of patients admitted to hospital?


Reference: DeMeester S et al. Implementation of a Novel Algorithm to Decrease Unnecessary Hospitalizations in Patients Presenting to a Community Emergency Department With Atrial Fibrillation. AEM June 2018

  • PopulationAdult emergency department patients with a primary diagnosis of atrial fibrillation or atrial flutter.
    • Excluded: Individuals with an alternate primary diagnosis (ie sepsis), pregnant, or incarcerated patients.
  • Intervention: An algorithm for the management of atrial fibrillation developed as a collaboration between the emergency and cardiology departments.

  • Comparison: This is a before and after study. Pre-intervention data was collected over a 1-year period before the algorithm was developed. Post-intervention data was collected over a 1-year period after the algorithm was implemented. (They excluded the year around implementation, as uptake was gradual).
  • Outcome:
    • Primary Outcome: Rate of hospital admission.
    • Secondary Outcomes: Return visits to the emergency department at 3 and 30 days.

Dr. Susanne DeMeester

Dr. Susanne DeMeester completed her residency at the University of Maryland in 2006, She has practiced at St Joseph Mercy Hospital, a mixed community and academic ED, in Ann Arbor, Michigan.  She is the director of the emergency observation center and serves as the ED cardiology liaison.

Authors’ Conclusions: Implementation of a novel algorithm to identify and treat low-risk patients with AFib can significantly decrease the rate of hospital admissions without increased emergency department returns. This simple algorithm could be adopted by other community hospitals and help lower costs.” 

Quality Checklist for Observational Study:

  1. Did the study address a clearly focused issue? Yes
  2. Did the authors use an appropriate method to answer their question? Unsure
  3. Was the cohort recruited in an acceptable way? Yes
  4. Was the exposure accurately measured to minimize bias? Unsure
  5. Was the outcome accurately measured to minimize bias? Yes
  6. Have the authors identified all-important confounding factors? Unsure
  7. Was the follow up of subjects complete enough? Yes
  8. How precise are the results? Unsure
  9. Do you believe the results? Yes
  10. Can the results be applied to the local population? Unsure
  11. Do the results of this study fit with other available evidence? Yes

Key Results: There were 586 patients with atrial fibrillation in the pre-intervention year, and 522 during the post-intervention year. The mean age was around 70 and there was almost a 50/50 male female split. Overall, they appear to be well watched at baseline.


The algorithm decreased the number of patients admitted to hospital.


  • Primary Outcome: Admissions
    • 80.4% before vs. 67.4% after (difference 13%; p<001)
  • Secondary Outcomes:
    • There was no difference is 3 or 30-day ED return visits. The overall bounce back rate was very low – just less than 4% is both groups.
    • There were no deaths noted by database search
    • Length of emergency department stay was no different.

Listen to the podcast on iTunes to hear Susanne’s responses to our five nerdy questions.

1) Confounders: One of the limitations of a before and after study design is the possibility of the confounders. There has been a general trend toward more outpatient management in medicine. Is it possible that admissions might have decreased even without this algorithm, or that there were other factors, such as the Hawthorne effect, that contributed to the observed decline rather than the specific algorithm you developed?

2) External Validity: As you mention in the manuscript, atrial fibrillation management varies significantly around the world. Where I work, admission rates are much lower than described here, and a higher proportion of patients are managed with a rhythm control strategy. How might that impact the generalizability of your results?

3) Follow-Up: We noticed that, despite what sounds like incredible outpatient follow-up, more than 10% of patients did not show up for their scheduled outpatient follow-up visits. You designed your algorithm to be simple, delaying some testing and decisions about anticoagulation until the outpatient visit. Do you think that the difficulty with outpatient compliance could affect implementation of protocols like this elsewhere?

4) Clinical Judgment: You developed your algorithm in conjunction with the cardiology department. You note that there are no clear consensus guidelines to identify patients who can safely be managed as outpatients and those who need admission. So the criteria that you use were based on the expertise of your cardiologists and have not been validated. Do you think that these criteria add anything to simple, unstructured clinical judgement?

5) Adverse Events: Adverse events were inferred from return visits to the emergency department. If patients had adverse events, they may have been unhappy with their care, and decided to present to different hospitals. Is it possible that you missed some adverse events among discharged patients?

Comment on Authors’ Conclusion Compared to SGEM Conclusion: We agree that an algorithm might be used to decrease atrial fibrillation admissions, but its value will depend heavily on your current practice. This algorithm has not been validated in other populations.


SGEM Bottom Line: There are clearly patients with primary atrial fibrillation who can be managed safely as outpatients. There are no evidence-based criteria for identifying high-risk patients who require admission, so for now we will have to rely on clinical judgement.


Case Resolution: You are able to achieve rate control with an oral calcium channel blocker. The patient is asymptomatic, and your workup has not identified any high-risk features. After a shared decision-making conversation, you discharge the patient home with close follow-up with the cardiology department.

Clinical Application: Hospital admissions are very expensive, and inconvenient for patients. For high-risk patients, the hospital environment may help prevent bad outcomes, but for low-risk patients, we should consider outpatient management for atrial fibrillation.

Dr. Justin Morgenstern

What Do I Tell My Patient? You have a condition called atrial fibrillation, which basically means your heart is beating fast and irregularly. The management of this condition varies, depending on where you are in the world. Seeing as you are currently feeling well, and have no risk factors for stroke, we have a few options including trying to shock your heart back into its regular rhythm or trying to slow it down. Once we have the arrythmia controlled, we can discuss whether it would be a better option to admit you to hospital or to go home with close follow-up.

Keener Kontest: There was no winner last week. Listen to the SGEM podcast on iTunes to hear this weeks’ question. If you know the answer send an email to TheSGEM@gmail.com with “keener” in the subject line. The first correct answer will receive a cool skeptical prize.

SGEMHOP: Now it is your turn SGEMers. What do you think of this episode? Tweet your comments using #SGEMHOP. What questions do you have for Susanne and her team about atrial fibrillation? Ask them on the SGEM blog. The best social media feedback will be published in AEM.

Also, don’t forget those of you who are subscribers to Academic Emergency Medicine can head over to the AEM home page to get CME credit for this podcast and article. We will put the process on the SGEM blog:

  • Go to the Wiley Health Learning website
  • Register and create a log in
  • Search for Academic Emergency Medicine – “June”
  • Complete the five questions and submit your answers
  • Please email Corey (coreyheitzmd@gmail.com) with any questions or difficulties.

Remember to be skeptical of anything you learn, even if you heard it on the Skeptics’Guide to Emergency Medicine.