What Would You Do?: What is Your Admission Threshold?

Home / Journal Club / What Would You Do?: What is Your Admission Threshold?

chest pains

A 42 y male with a history of Diabetes and Hypertension walks into your Emergency Department with a complaint of chest tightness. He was just seen at a previous well-respected hospital where he had the same complaint. The gentleman had been arranged for a follow up cardiology appointment. The pain has been pretty constant, giving him some shortness of breath. He also has noticed that when he climbs the stairs from the train (a climb he does daily) he noticed some worsened dyspnea about 2 days ago. Since then he has been able to do it just fine. The pain does not radiate to his arm or jaw and is not positional. He denies any pulmonary embolism type symptoms (for all you avid PE-ologists). His pain had gotten slightly better since the visit to the last hospital but this morning around 4 AM the pain came back. He was sweating a little bit with the pain. He had taken Motrin 800mg before coming and now the pain has totally resolved. His EKG looks like the one below.


What tests do you run on this gentleman? Is the EKG concerning for ACS? If he has negative cardiac enzymes do you let him go to the Cardiologist? What is your threshold for admission? What would elements of this or any patient’s history when it comes to chest pain make you push the admit button? In sum…what is your management of this situation?

I look forward to your answers and sharing your knowledge and experience with me.



  • alvarezzy

    just finished essentials of EM in SF, and here’s one of the things I learned… in 1959, they would have done sham thoracotomy to ligate their internal mammary gland. Cobb LA. Evaluation of internal mammary ligation by double-blind technic. N Engl J Med 1959; 260: 115–118. It showed better exercise tolerance post ligation of said gland in pt’s with acute coronary syndrome.

    will have my full answer after i finish my 7 in a row overnight stint at the county starting tonight..


  • andrewchertoff

    Here’s what I heard:

    A 42 y male with a history of Diabetes and Hypertension walks into your Emergency Department with a complaint of chest tightness, giving him some shortness of breath, worsened exertional dyspnea about 2 days ago, this morning around 4 AM the pain came back, he was sweating, and his EKG abnormal.

    What tests do you run on this gentleman? Admission. (For functional testing, no further ED testing except another set.)

    Is the EKG concerning for ACS? Yes.

    If he has negative cardiac enzymes do you let him go to the Cardiologist? NO!

    What is your threshold for admission? See above short paragraph – why would I discharge that?

    What would elements of this or any patient’s history when it comes to chest pain make you push the admit button? Acute definable exertional change is always an admission for me. Every day i wake up and have the very same level of function – if it suddenly changed, I would assume there is a cause. Diaphoresis – sympathetic stimulation (innervated by sympathetic “cholinergic” neurons) just isn’t right. Risk factors – this man has CAD (HTN + DM at the sweet young age of 42), we just don’t know how significant. Sx onset and presentation at 4am – chiding of Bronx behavior aside, you know you would not get out of bed at 4am unless you had to.

  • sannman

    Diffuse ST elevations with no reciprocal changes. Are we supposed to think pericarditis, given that an NSAID helped resolve his pain? Still, I don’t think this is a question of admitting so much as it is a question of whether or not to start anticoagulation. The story is worrisome and this guy needs frequent serial EKGs early on.

  • pik

    OK: the story is concerning, we can all agree. No one wants to do nothing, just to have his doc send him back to the ED in the a.m.

    1. His pain’s gone. That’s great! Now I can pontificate on what to do without that whole time is heart thing.
    2. His EKG is abnormal. That sucks. Now I have to decide whether to do some kind of aggressive or interventional thing.

    SO: Do I need to call Cards? Well, he has diffuse STE’s. Which isn’t the end of the world, some 2/3 of STE aren’t occlusive CAD. Does it look ugly? Not really, looks like the happy upslopy kind. Do we ignore those types of STE in mult. risk factor guy? No, we ignore them in young black males who are in the ED for toe pain. (Referencing JAMA article on STE here).

    1st thing we do is look for reciprocal changes. ANY reciprocal change cannot be pericarditis. Next thing we do is pay attention to funny late T changes: biphasic things or inverted u waves. V1 looks off to me, makes me worry. So if I’m worried, off I go to Steve Smith’s website to see how bad those STE’s with high voltage R’s really are. (http://download.journals.elsevierhealth.com/pdfs/journals/0196-0644/PIIS0196064412001606.pdf) Since you guys cut off the computer calc. of QTc, I have to guess at it but since the rate ain’t too weird, I’m hoping it’s close to 370-380. Pop in the values, you get 21 or so, which is less than 23 so no heparin or Plavix or nothin for this guy, just ASA and time.

    BUT- that V1 would prompt me to do a repeat EKG especially for recurrent chest pain. And even though you said nothing PE-ish, I’d consider a neb for exertional dyspnea and chest tightness. Delta trop negative and no obs unit in sight? Yeah, I might then send him off to his doctor.


  • To summarize, we have a diabetic patient presenting with non-active chest pain with ST elevations in the anterior leads.

    Firstly, if you listen to a lot of Amal Mattu, he has stated several times that the new (well maybe not so new) is that “diabetes = CAD.” He uses the example from the cardiology literature in the Heart Journal from 2005 that found, essentially that DM + ACS + negative cardiac enzymes = same risk of adverse events as non-DM + ACS + positive cardiac enzymes. The sample size of this study is good enough with around 1600 patients, but their prevalence of cardiac biomarker elevation is higher than that of our population (at least in my experience so far); so if you feel the prevalence of NSTEMI in this study is higher than that of our patients in the Bronx, then maybe you will think these findings don’t apply to us.

    So…I have used this patient to admit diabetic patients regardless of the presenting ECG.

    But this guy’s ECG isn’t normal. He’s got ST elevations with the “classic” J-point notching as talked about in some of our Wednesday conferences and Friday follow-up rounds. However, the only literature we ave on J-point notching (again, previously thought to be “classic” of early repolarization) is from Smith, when he found that 14% of subtle anterior STEMIs have J-point notching. In this same study, the only good study on this topic of STEMI versus early repolarization, he found only 3 ECG findings indicative of Early Repol versus STEMI: R wave in V4 > 17mm, QTC < 384, and the degree of ST elevation in V3. He derived a formula, not yet externally validated, with a positive LR of 9.2.

    Take a look at my post I made for Michelle Lin of Academic Life in EM:


    It's the most comprehensive document on the subject on the internet or in literature right now.

    See if you think this information will be useful in this patient.

    Then ask yourself if the ECG even matters.

    Will you two set this patient (keep in mind that some institutions 3 set ROMI from the ED and possibly perform a stress ECG if there is a chest pain unit attached) … or will you admit? Most NYC providers or conservative chest pain-equipped providers would admit this patient.

    I think there is reason to do either, provided either you or the patient can arrange appropriate follow-up AND this patient remains chest pain-free in the ED.

  • alvarezzy

    i would admit this patient. NOT because he’s diabetic but because it is his second visit and progressive sx. 2nd visits ALWAYS warrant a more careful look, perhaps the next level of diagnostics, etc unless it’s for wound check =)

    i think admitting all diabetics with chest pain from an operations point of view is impractical. I remember how easy this was during residency–Monte west side. chest pain. no prior stress test. admit. no questions asked. Now, working in a busy county hospital, this does not fly, and for good reason. It supersaturates the ED/inpatient floors.

    For the risk averse, I can understand why they would admit. Missed MI is the #1 reason why ED docs are getting sued.
    Academic Emergency Medicine. Volume 17, Issue 5, pages 553–560, May 2010

    But we simply can’t practice for fear of being sued.

    Here’s a summary from EMRAP November 2012:
    On sending patient’s home with unstable angina:
    Pope et al. Missed diagnoses of acute cardiac ischemia in the Emergency department. N Engl J Med. 2000 Apr 20;342(16):113-170. Not a great study, but demonstrates people with UA sent home do not die at a higher rate than if they were admitted.

    On low risk chest pain and stress testing. And low risk for ACS is attributed to “Patients who present to the ED with chest pain (or its equivalent) but have no electrocardiographic changes or elevation in cardiac biomarkers after an appropriate interval”. Kosowsky JM. Approach to the ED patient with “low-ˇrisk” chest pain. Emerg Med Clin North Am 2011Nov; 29(4):721‐7. http://pdvc.hng.co.in/Articles_PDF/CCE_8Low%20risk%20chestpain.pdf Here they found no added benefit for doing stress testing.

    On early PCI for UA with nl troponin. Cochrane Database Syst Rev. 2010 Mar 17;(3):CD004815
    “Compared to a conservative strategy for UA/NSTEMI, an invasive strategy is associated with reduced rates of refractory angina and rehospitalization in the shorter term and myocardial infarction in the longer term. However, the invasive strategy is associated with a doubled risk of procedure-related heart attack and increased risk of bleeding and procedural biomarker leaks” Also, “Subgroup analysis found that patients who did not have a positive troponin had a slight increase in mortality when stents were placed.”

    At essentials this month, Mike Weinstock (author of Boucebacks) mentioned his study with Weingart, Newman, etc al. “What is the risk of life threatening arrhythmia, inpatient STEMI, arrest, and death in admitted ED patients with chest pain, non-ischemic ECG, and 2 negative initial troponins?” Not yet published. Retrospective cohort study. 5 years. Inner city. 45K patients screened. Subject: 18+ with 1. Chest pain 2. Chest tightness 3. Chest burning 4. Chest pressure as chief complaints. Half got admitted. 5.5K had neg trop x 2. Icu admissions excluded. Total study population n=3.3K. result: Only 4/3,376 had adverse outcomes. 2 had cardiac arrest, 2 had inpatient stemi. 2 of the 4 had event during stress test. Interesting data. Stay tuned!

    The EMRAP group, however, cautioned that the current AHA guideline still states that low risk CP may be discharged but need stress testing within 72 hours.

    So we have confusing recommendations. What I find helpful is discussing this with the patient. This “shared decision making” discussion is important and key to document. “I think from our workup so far, you are considered to be low risk for ACS… Here are some options… What would you like to do?”

    Now, back to our patient… I would admit this patient because of the bounceback, HIGH risk features of his chest pain, and progressive sx. As an alternative, I would call his cardiologist and discuss the plan (and recent sx). I would argue that the cardiologist at this point would not risk sending him home. If the cardiologist still recommends dc, I would order an echo, and risk stratify that way. New onset CHF = admission. There’s also the Obs unit. Ultimately, I’d discuss these with the patient and get his input. If he wants to stay, done. If he wants to go home, I would call back his cardiologist to see if they can push the visit sooner and what other things we can do to expedite his care. All of this would be found in the chart.

Leave a Reply