Does Observation for ACS Makes Sense?

Observation for ACS title image

I have dedicated a number of posts to the workup of low risk chest pain. There is the 5 part series demonstrating that our use of stress tests doesn’t make any sense. There is also a post demonstrating that CCTA has no role in these patients at this time. One topic that I have never covered is the role of observation units for low risk chest pain (mostly because we don’t use these in Canada; we just send patients home). Therefore, I am thrilled to host a guest post by Dr. Charles Murchison, modified from an original series of blogs on the County EM blog, addressing the question: Does Observation for ACS Make Sense?

This is a guest post by Dr. Charles Murchison. Charles is a PGY-4 EM resident at Kings County Hospital/SUNY Downstate Medical Center with an academic interest in evidence-based medicine. When not in the hospital he can be found surfing out on Rockaway Beach. 

Part 1: A History of Observation for Chest Pain

Accurate and expeditious diagnosis of acute myocardial infarction (MI) is one of the key charges of emergency departments and emergency providers. Chest pain is the second most common reason for emergency department visits in the United States, and coronary artery disease is the leading cause of death (1,2).

Up until the late 1980s, most patients presenting to the emergency department (ED) with acute chest pain were simply admitted to a coronary care unit (CCU) (3). Some of these patients had acute MI and were treated accordingly. However, the majority were low risk patients admitted simply to monitor for the development of acute MI with serial ECGs and cardiac enzyme measurement (CK-MB at that time). Over half the patients admitted to the CCU with acute chest pain were eventually “ruled out” for myocardial infarction (4). Admission of these “rule out MI” patients to an intensive care unit was costly, leading hospitals to seek ways of distinguishing the low risk patients earlier. Researchers developed decision aids to help identify low risk patients in the ED (5, 6), and ran studies to find the shortest amount of time in which MI could be ruled out. It was determined that 12-24 hours was an acceptable observation period for ruling out MI (7, 8). Yet still, these patients were being admitted to the CCU for this observation period.

In 1991, ED doctors at Brigham and Women’s Hospital published a study describing a “coronary observation unit” in which patients with low risk chest pain were monitored in an ED observation unit to rule out MI rather than being admitted to the hospital (9). They used their own clinical algorithm to determine low risk, which they defined as <10% risk of acute MI. Patients underwent ECG every 12 hours and CK-MB testing every 8 hours for 24 hours. Cardiology consultation was optional. If both ECG and cardiac enzymes remained negative during the observation period, the patient was discharged home. The investigators found no increase in cardiac events or death between the patients placed in the ED observation unit and the controls who received usual care. Three years later, they published a follow-up paper from the same observation unit with a larger patient cohort. The results again confirmed that ED observation was safe, and in addition, they found significant cost savings for the hospital (10).

An important side note from this study: Prior to discharge, the patients in the ED observation unit were supposed to undergo an exercise stress test. The authors admitted that stress testing has poor sensitivity and specificity, yet stated that it provided a “reasonable estimation of prognosis” for the patients in whom MI had already been ruled out. In this study, it is unclear what happened to the patients with abnormal stress test results.

Since 1991, observation units have been widely adopted, and it is likely most of them have a “rule out MI” pathway. The pathways at each hospital differ but typically include 12 to 24 hours of telemetry monitoring with serial ECGs and cardiac enzyme testing. Most will also include an inpatient stress test or one scheduled within 72 hours as per the 2014 AHA/ACC guideline recommendations (11). Optional additions to the pathway include cardiology consultation and a rest echocardiogram.

Now that we are 30 years from the first observation unit, we must re-evaluate the system. Technology and knowledge has changed since the advent of the observation unit. For one, we have an entirely new cardiac enzyme that is more sensitive and specific for myocardial necrosis than CK-MB. Troponin I and troponin T rise within 3-4 hours of myocardial injury, and can remain elevated for up to 14 days (12). Troponin measurement allows us to pick up more MIs than CK-MB could. Recent studies show that many patients who have negative CK-MB measurement will have positive troponin levels (13). A more sensitive test should decrease the rate of missed MI in the ED and therefore obviate the need for further monitoring.

In addition to the new cardiac enzymes, we have improved clinical prediction tools to identify low risk chest pain patients. The TIMI score, GRACE score, and more recently, the HEART score are all validated scoring systems. Emergency providers have warmly embraced the HEART score (and HEART pathway) given its simplicity, accuracy, and improvement in ED workflow. Through the use of these clinical prediction tools, we not only avoid CCU admission for patients with chest pain, we have identified a large subset who are so low risk that they can be safely discharged from the ED. (JM note: even without the HEART score, I think I can identify patients who are low enough risk to go home. It is essentially everyone with negative troponins and normal ECGs. We manage essentially all these patients as outpatients in Canada.)

So now we are identifying more MIs in the ED and safely discharging a large swath of low risk chest pain patients. Who is left for observation? Are there any patients who still benefit from this practice? 

Part 2: Identifying ACS

There are two of the main goals of placing a patient on observation for chest pain:

  • Identify patients having cardiac chest pain who may benefit from acute intervention – in other words, to identify patients with acute coronary syndrome (ACS).
  • Risk stratify patients for adverse cardiac events.

This section will discuss the process of identifying ACS in the ED and the observation unit. 

So what is ACS?

ACS is an acute thrombosis in a coronary vessel causing varying degrees of occlusion (14). Classically, ACS was divided into ST-segment elevation myocardial infarction (STEMI), non-ST-segment elevation myocardial infarction (NSTEMI), and unstable angina (UA). STEMI is thought to result from sudden complete occlusion of a coronary vessel with subsequent transmural infarction, whereas NSTEMI and UA involve thrombosis that either does not entirely occlude the vessel or occurs in a vessel with good collateral circulation. NSTEMI and UA are distinguished by the presence or absence of infarction, i.e. elevated cardiac biomarker level. (JM note: We use ST elevation as a surrogate, but the important distinction is between MI with occlusion of the coronary artery and MI without occlusion of the artery, because the ECG is an imperfect tool.)

Does the ED actually miss ACS that often?

Several studies since the late 1980s have shown that EDs are actually quite good at identifying MI (15, 16, 17). Each study found a 2% rate of missed MI in the ED. However, the conclusion from all three of these papers was that ED providers are not good at identifying MI. To support this argument the authors ignored data supporting good outcomes and embellished on findings suggesting bad outcomes for patients. Who wants to publish an article suggesting everything is just fine?

As an example, the NEJM study from 2000 (16) found that of the 19 patients with missed MI, 2 of them (11%) had ECGs that were later interpreted as ischemic by a blinded cardiologist. In their conclusion, they state that there was an “11% incidence of failure by the emergency department clinician to detect ST-segment elevations… in the electrocardiograms of patients with myocardial infarction”. That is a gross misrepresentation of the data. Out of over 10,000 chest pain patients and 950 patients diagnosed with MI, there were 2 who were missed despite having potentially ischemic ECGs, and 19 missed over all. They should have concluded: “Out of 10,000 chest pain patients, emergency physicians only missed 2 MIs (0.02%) whose ECG suggested ischemia”. 

In all three studies the mortality rate between missed MI and those admitted with identified MI was the same – about 10%. However, despite the reassuring numbers, the authors attribute this to insufficient data. They say the outcomes must be worse for patients with missed MI, and they just didn’t have large enough numbers to show it.

The scary, albeit misleading, conclusions of these studies led to a good amount of fear among emergency physicians and hospitals, shaping modern day practice. 

Related post: Stress tests part 1: How often do we miss MI?

Does repeating ECGs and cardiac enzymes in the observation unit help identify ACS?

This is what observation units were originally designed for, but is it still necessary? Troponins are more sensitive than CK-MB and become positive within 3-4 hours, with 6 hours being the maximum amount of time it would take for us to see a positive test result (18, 19). Given that we already did (hopefully) serial ECGs and “0 and 3-hour troponin” in the ED, is further testing going to add anything? (JM comment: the exact timing of the troponin will depend on what troponin assay you are using.) If we are really worried, can we not repeat the ECG and troponin in 6 hours in the ED? This isn’t a prohibitively long length of stay. “Observation unit crowding” and handoffs to other providers in the observation unit who may not know the patient are a common source of medical errors and patient harm (20), so this must factor into our thinking of the risks and benefits of observation for the patient.

Let’s think about it another way. We often use the HEART pathway to discharge chest pain patients after 3 hours, if they are deemed low-risk. What we are essentially saying is we have ruled out ACS. Why wouldn’t that timeline also apply to “high-risk” patients? We can either rule out ACS in 3 hours or we cannot – the level of risk should not change this fact. (JM comment: I think this is a very important point, and why I actually dislike the HEART score. I make the diagnosis of MI based on the ECG and troponins. A very high risk history is also important, but the other factors in the HEART score are sort of irrelevant for my immediate management. Sure, a patient with multiple risk factors is high risk, but if they aren’t having acute ACS today, they don’t need to stay in the hospital. They need risk factor modification, but that can happen as an outpatient.)

Can a rest echocardiogram in the observation unit help identify ACS?

Most observation units these days will include a resting echocardiogram as part of their observation pathway for patients with chest pain. This may help further risk stratify a patient for ischemic heart disease and congestive heart failure, but does it help with our fundamental question of identifying current ischemia or infarction? Looking for regional wall motion abnormalities would be the primary means for this. Studies have shown that rest echo to identify regional wall motion abnormality is on average only about 87% sensitive for ACS or MI, and not at all specific (21, 22, 23, 24). In the study that showed the highest sensitivity of echo (93%) for identifying ischemia, 2.3% of the patients who did not have regional wall motion abnormality were actually having MI (21).

This conclusion intuitively makes sense. Echo can be limited by patient’s body habitus, the ability of the cardiologist to interpret the images, variants in anatomy, etc. Also, if you see a wall motion abnormality, how do we know it is linked to what is happening now? Unless we have an echo from one week ago, identifying an abnormality as one data-point is unlikely to be helpful. Just as an ECG without comparison to a prior has limited utility. (JM comment: I have seen regional wall motion abnormalities used to differentiate between occlusional and non-occlusional MIs, but I can’t imagine this picking up MIs that are ECG and troponin negative, especially seeing as these misses only seem to happen in 1-3 in 1,000 chest pain patients we see.)

What about that older patient with a scary story and a bunch of risk factors but a non-diagnostic ECG and negative serial troponin? Isn’t that the ideal observation patient?

Let’s do this thought experiment. You work in a perfect ED where the providers accurately interpret every ECG and all causes of non-ACS chest pain have been ruled out. In this ED, STEMIs will be identified by ECG and NSTEMIs will be identified by a positive troponin. STEMI and NSTEMI are not candidates for most observations units. Therefore, the only patients going to the observation unit are those with unstable angina (and a lot of patients without ACS at all). So let’s take a deep dive into the murky world of unstable angina.

What is unstable angina? Is it even real?

Probably not. Look at it this way: the person who created the first guidelines for the diagnosis and treatment of unstable angina (UA) no longer believes it to be a real entity (25). The definition was created at a time when we were still using CK-MB as evidence of myocardial necrosis. As mentioned earlier, troponin is far more sensitive and specific than CK-MB. By 2000, AHA guidelines estimated that more than one-third of patients previously thought to have UA would be reclassified as NSTEMI, given they would have had a positive troponin (26).

Since then, troponins have become even more sensitive with a lower cut point for an abnormal result, further classifying more patients as NSTEMI rather than UA. And now, we have “high-sensitivity” troponins. Almost all patients with chest pain due to ischemia will have detectable high sensitivity troponin (27), and even slight elevations in high-sensitivity troponins have been associated with adverse outcomes (28). It is likely that the upper limit of normal will become progressively lower with time. While this will result in lower specificity for diagnosing ACS, it will lead to higher sensitivity of identifying NSTEMI, and possibly result in the end of UA as a diagnosis. Patients will either be classified as having stable angina of varying severity or MI. 

(JM comment: I tend to agree. I very rarely make the diagnosis of UA. However, I think this distinction requires us to be VERY good at reading ECGs (a lot of the cases of missed MI that people point to as unstable angina actually have ischemic ECGs). There are very smart doctors who disagree. Dr. Smith (of the ECG blog) has many posts about cases of unstable angina (although often with ECG learning points). I think there will be occasional patients with ischemia that for whatever reason results in the first 2 troponins coming back negative. However, these patients will be rare enough that they shouldn’t have an entire system of care built around them, and can probably be addressed by either admitting patients with very high risk histories, or occasionally keeping very high risk patients for a third troponin. This may happen to me once or twice a year, at most.)

For now though, many of us do not have high-sensitivity troponin assays available, so we may still see some people with “unstable angina”.

How do you diagnose unstable angina?

Unstable angina is typically defined as angina that is either worsening, occurs at rest or is new, coupled with either transient ST elevation (>1 mm and <20 minutes), ST depression (>0.5 mm) or TWI (>3 mm) in at least 2 contiguous leads OR in a patient with known CAD (as diagnosed by angiography or prior MI) (29, 30).

As you can see from this definition, a patient with chest pain, known CAD, and no ECG changes could technically have UA. Add a negative troponin and that seems like the patients who we would typically see in an observation unit.

Should we be sending unstable angina to obs? What is the appropriate treatment of unstable angina?

By definition, patients with unstable angina are patients with ACS, and they should presumably be treated aggressively for ACS. The following is a brief, up to date review of the literature on treating unstable angina. Of note, unstable angina is often grouped with NSTEMI and called NSTE-ACS.


A 2009 Cochrane Review concluded that nitrates reduce short-term mortality in patients with ACS (31). The studies in this review used IV nitrates, primarily IV nitroglycerin.

2014 AHA guidelines recommend sublingual nitroglycerin to be administered to all patients with chest pain concerning for ACS, followed by IV nitroglycerin if the pain is refractory to sublingual medication (11).

(JM note: This topic might need its own deep dive at some point.)


In 2002, the Antithrombotic Trialists Group performed a meta-analysis of aspirin use in various cardiovascular disorders including acute stroke and acute MI (32). They found that aspirin at initial presentation, and then low dose daily, reduced serious vascular events in every subset of patients. One of these subsets was patients with unstable angina.

2014 AHA Guidelines recommend giving 162 to 325 mg non-enteric coated aspirin, chewed, to all patients with suspected ACS (11).

Dual Antiplatelet Therapy

The CURE trial established the benefit of clopidogrel, with aspirin, given immediately on patient presentation with ACS (33). Patients presenting within 24 hours of suspected NSTE-ACS – defined by ECG changes or elevated cardiac biomarkers – were randomized to receive aspirin plus a loading dose of 300 mg clopidogrel, and then 75 mg daily thereafter, compared to control receiving just aspirin. They found reduced recurrent ischemia, MI, heart failure, and death in the group receiving aspirin + clopidogrel. These benefits were seen early (in the first 30 days) and long-term (over the next 9 months). 

2014 AHA guidelines recommend dual antiplatelet therapy for treatment of NSTE-ACS (11).


A 2008 Cochrane Review concluded that heparins do not improve mortality, but do reduce the risk of MI in the first week – from 5% to 2% – in patients presenting with ACS (34). The studies included both low molecular-weight heparin (LMWH) and unfractionated heparin (UFH). It should be noted that this risk reduction in MI disappeared at 30 days and at 3-6 month follow up.

There is considerable controversy as to whether heparin benefits patients with NSTE-ACS, given its known, significant bleeding risk. For more reading see this great post by the website NNT: NNT – Heparin for ACS.

Regardless of the data, 2014 AHA guidelines recommend the use of heparin in all NSTE-ACS (11).

(JM comment: As I outline in this post, I think anticoagulants in NSTEMI are clearly harmful based on the evidence we have and absolutely shouldn’t be used.)

Fibrinolytic Therapy

The TIMI IIIB trial found no benefit and increased risk of severe hemorrhage for fibrinolytic therapy in patients with NSTE-ACS, (35).

2014 AHA Guidelines do not recommend fibrinolytics as a treatment for NSTE-ACS (11).

Invasive strategy

A 2016 Cochrane review concluded that an early invasive strategy (within 2 hours) in patients with unstable angina or NSTEMI does not improve mortality when compared to a selective invasive strategy (36). A selective invasive strategy involves performing catheterization only on patients with ongoing ischemia – recurrent infarction, angina at rest, dynamic ST changes on ECG, and definitive inducible ischemia on provocative testing. They found that patients who underwent early invasive therapy had a 2% decreased risk of MI at one year, but a 3% increased risk of MI during the actual procedure. It should also be noted that most patients in this meta-analysis had positive biomarkers, i.e. had NSTEMI. When the authors analyzed only those patients with negative biomarkers, i.e. unstable angina, they found increased mortality with an early invasive strategy.

2014 AHA guidelines recommend risk stratifying patients to see who would benefit from an early invasive versus delayed invasive strategy (11). This may change as the Cochrane Review came out after these recommendations.

Related post: Stress Tests Part 4: Revascularization and the Value of Stenting

Should unstable angina be admitted to an observation unit?

After evaluating the evidence and recommendations for treatment of unstable angina, it seems clear that patients whom we suspect have unstable angina, i.e. ACS, may not be suitable for the observation unit. That may include the older patient with a concerning history and known CAD, but a non-specific ECG and negative initial troponin. (A patient with “clinically important” unstable angina or ACS – one in whom there may be a significant risk of short-term complications and aggressive treatment may benefit – will likely have ongoing/recurrent symptoms and demonstrate dynamic ECG changes and/or detectable troponin on serial testing). Regardless of current evidence to the contrary, AHA guidelines still recommend aggressive medical therapy, and possibly even an early invasive strategy for patients with unstable angina.


In conclusion, observation and the interventions and tests that come with it are unlikely to help us identify ACS any better than our initial ED evaluation. Troponin will almost always become positive within 6 hours in patients with NSTEMI, so this can reasonably be identified within an ED visit. For those patients who are high-risk – concerning stories or ECGs or ongoing chest pain – consider the possibility of unstable angina and remember that the treatment for this should be timely and relatively aggressive. I would argue that placing a concerning patient on observation is not appropriate. Cardiology should be consulted early in such patients. For patients with non-concerning stories and ECGs, ACS can be ruled out in an ED visit, regardless of the patient’s age and risk factors.

Part 3: Risk Stratifying for Adverse Cardiac Events

Observing patients for chest pain to “rule out ACS” may no longer be relevant. The tests we do in the observation unit are unlikely to help us more accurately determine which patients are having ACS – STEMIs and NSTEMIs should be identified in the ED, and unstable angina may no longer exist. Even if unstable angina is real, it is ACS and should be managed accordingly. So, for that high-risk patient with a concerning story but a non-diagnostic ECG and negative troponin, you should probably decide in the ED whether or not this is ACS. If it is ACS – treat accordingly. If not – they do not need observation to “rule out ACS”.

Which brings us to our final topic. The second goal of observation is to categorize a patient’s short-term risk of adverse cardiac events, which if determined to be high, would ideally lead to an intervention that reduces this risk. This assessment of risk is supposed to happen after ACS has been ruled out.

How do we risk stratify a patient’s risk of major adverse cardiac events (MACE)?

There are a few scoring systems for risk stratification of chest pain patients, but the one that is most valid in the ED today is the HEART score. The evidence suggests that if chest pain patients have a low enough HEART score (0-3), they can be discharged home safely. However, if a patient’s HEART score indicates an unacceptably high risk for major adverse cardiac events (MACE) over the next 6 weeks, they should be either placed in an observation unit or admitted. The question is whether there are any tests we can do that further risk-stratify a patient’s risk for MACE, or even better, are there procedures to reduce this risk.

Will stress testing help better risk stratify a patient’s 6-week MACE?

Stress testing can be done to see if people have “inducible ischemia”, which can be a sign of coronary artery disease (CAD). However, stress testing is not accurate at diagnosing CAD, defined as >50% stenosis in a coronary artery on catheterization. At best, it is 80% sensitive using stress echo, and at worst, it is 70% sensitive using exercise ECG. Stress testing and stress echo are also nonspecific (80% specificity) for CAD (37). (JM comment: explored further in this post.)

As far as helping better stratify a patient’s risk for 6-week MACE, no studies show that stress testing helps. A positive stress test may identify patients at risk for long-term adverse outcomes – 6 months to 5 years – but not at 6 weeks (38, 39, 40, 41, 42). Furthermore, recent studies show that the short-term risk of adverse events after ruling out MI is extremely low, and stress testing in this situation confers no added benefit (43, 44, 45).

Yet, AHA guidelines still recommend stress testing these patients and it is unclear why. One cited study that supports stress testing showed that inpatient stress tests result in better outcomes in patients who have been ruled out for MI (46). Taking a closer look at this study, we see that it is deeply flawed.

First, the investigators identified adverse outcomes as death, MI or unstable angina. They defined unstable angina as chest pain at rest or a positive stress test. This involved incorporation bias – the reference standard uses the test that is being studied. (JM comment: this flawed circular logic appears throughout the stress testing literature. We think stress testing is good because it identified these patients with unstable angina. We think these patients have unstable angina because of a positive stress test. And round and round we go.) Also, dying and having a positive stress test are very different outcomes, yet they are lumped together as equal.

Second, their recommendation for inpatient stress testing came from this data: 23% of patients scheduled for outpatient stress test had an adverse outcome compared to only 10% who received an inpatient stress test. But what about the people who never got a stress test at all? Only 9% had an adverse outcome. They had the best outcomes. Yet the authors don’t mention this anywhere in the paper. The title should have been “Inpatient stress testing is associated with better outcomes than outpatient stress testing, but no stress testing results in the best outcomes”.

The evidence is fairly conclusive that stress testing offers no benefit to patients in whom ACS has been ruled out. But let’s say we do one anyway, and it’s positive. A patient with a positive stress test will likely go for coronary angiography. As we mentioned, stress testing is (at best) 80% specific for CAD, so 20% of these patients will have no evidence of CAD and will undergo an invasive procedure with associated procedural risk and no benefit. (JM note: and we have already said that patients with negative troponins don’t benefit from invasive management.)

Will medical therapy reduce a patient’s 6-week MACE?

Aspirin, anti-hypertensive medication, statins, beta-blockers, and lifestyle modifications have all been shown to improve mortality in patients with CAD, but again this is all based on long-term mortality benefit (32, 48, 49, 50, 51). A reduction in long-term mortality may also mean a reduction in short-term adverse events, but this has not been specifically studied. More importantly, we can ensure that the patient is on an appropriate medical regimen in the ED without placing them in observation. (JM comment: aside from aspirin, these interventions are just part of good primary care. Everyone should exercise, have a good diet, and ensure their blood pressure etc are appropriately managed. The only decision we have to make is whether to start aspirin, and I think we can manage that choice without invasive testing.)

Will PCI reduce a patient’s 6-week MACE?

The 2007 COURAGE trial evaluated patients with stable CAD and found no mortality benefit or decreased rate of MI for PCI versus medical therapy. They did find decreased rates of angina, but this is not the outcome we are worried about (52). (JM comment: And COURAGE was an unblinded trial, so the subjective outcome of angina is not trustworthy. In the only blinded trial – ORBITA – there was no benefit in terms of patient symptoms. This has further been confirmed by the ISCHEMIA trial, which was published after this post was written.) In 2014, a meta-analysis was published showing placing stents in patients with CAD does not reduce death, MI, or angina (53).

Will CABG reduce a patient’s 6-week MACE?

The current indications for CABG include left main stenosis, triple-vessel disease with an LVEF <40%, and two-vessel disease with proximal LAD stenosis. However, as with medical therapy and PCI, studies on CABG measure long-term mortality. They show 4% reduced mortality (over 5-8 years) with CABG versus medical therapy in selected patients with these indications (54). However, intraoperative mortality during CABG is 1%, so you can imagine short-term survival is likely to be lower.


Stress testing and coronary angiography cannot help us further risk stratify for short-term adverse outcomes. Medical therapy, PCI, and CABG have only been studied with respect to their long-term benefit, so it is unclear if they offer any reduction in short term adverse events. Keeping a patient in the hospital for further testing or initiation of treatment is not necessary and such testing can be done as an outpatient.


Observation units were developed in the 1990s to help hospitals save money. We did not have the sophisticated scoring systems or lab tests that we have today, therefore we needed 12 to 24 hours to rule out MI in patients presenting with chest pain, and we needed somewhere to do this outside of a CCU. This is no longer the case. We can safely rule out MI in 3-4 hours in the ED with contemporary troponin testing.

Most observation patients today fall into two categories: those in whom you have ruled out MI but are still concerned that they may be having ACS (the elusive unstable angina patient) and those in whom you have ruled out MI but they are “too high-risk for adverse events” to send home (the patient whose discharge “looks bad on paper”). I would argue that neither of these patients benefit from observation.

If unstable angina is real and we are concerned a patient is having unstable angina, they have ACS and should be treated accordingly. It is not an easy diagnosis to make by any means, and the decision to pull the trigger and involve cardiology will not be straightforward. Yet I would argue that the decision should be made in the ED. Using history, exam, and (perhaps serial) ECGs, emergency providers should determine ACS or not. If they are concerned, get cardiology involved early and start treatment.

If nothing in the history, exam, and ECG sounds like ACS, the patient can be safely discharged from the ED. Patients who are not having ACS, but are “high-risk for short term adverse events” are unlikely to benefit from observation or admission. Stress testing, angiography and all the downstream treatments have never been shown to reduce a patient’s risk for short-term adverse events.

Finally, a word on medical research. Through the course of reading for this topic, I’ve realized that so many studies are deeply flawed, yet somehow make it into official recommendations and therefore affect our practice. It’s impossible to read all primary literature and analyze all the data on your own, but I am now going to try to do this more. Maybe choose a few topics you are interested in and dig deep into the literature. (JM comment: Absolutely. Evidence based medicine is easy. That is why I can run a blog.) The results have really surprised me. It seems strange, but a blog post might be more reliable than a New England Journal of Medicine article. 


  1. Emergency Department Visits. CDC National Center for Health Statistics. Accessed June 1, 2019:
  2. Heart Disease Facts and Statistics. CDC. Accessed Jun 1, 2019:
  3. Lee TH, et al. The coronary care unit turns 25: historical trends and future directions. Ann Intern Med. 1988 Jun;108(6):887-94.
  4. Pozen MW, et al. A predictive instrument to improve coronary-care-unit admission practices in acute ischemic heart disease. A prospective multicenter clinical trial. N Engl J Med. 1984 May 17;310(20):1273-8.
  5. Pozen MW, et al. The usefulness of a predictive instrument to reduce inappropriate admissions to the coronary care unit. Ann Intern Med. 1980 Feb;92(2 Pt 1):238-42.
  6. Tierney WM, et al. Predictors of myocardial infarction in emergency room patients. Crit Care Med. 1985 Jul;13(7):526-31.
  7. Lee TH, et al. Sensitivity of routine clinical criteria for diagnosing myocardial infarction within 24 hours of hospitalization. Ann Intern Med. 1987 Feb;106(2):181-6.
  8. Lee TH, et al. Ruling out acute myocardial infarction. A prospective multicenter validation of a 12-hour strategy for patients at low risk. N Engl J Med. 1991 May 2;324(18):1239-46.
  9. Gaspoz JM, et al. Outcome of patients who were admitted to a new short-stay unit to “rule-out” myocardial infarction. Am J Cardiol. 1991 Jul 15;68(2):145-9.
  10. Gaspoz JM, et al. Cost-effectiveness of a new short-stay unit to “rule out” acute myocardial infarction in low risk patients. J Am Coll Cardiol. 1994 Nov 1;24(5):1249-59.
  11. Amsterdam EA, et al. 2014 AHA/ACC guideline for the management of patients with non-ST-elevation acute coronary syndromes: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation. 2014 Dec 23;130(25):2354-94.
  12. Morrow D, et al. National Academy of Clinical Biochemistry Laboratory Medicine Practice Guidelines: Clinical Characteristics and Utilization of Biochemical Markers in Acute Coronary Syndromes. Circulation. 2007;115:e356–e375
  13. Hamm CW, et al.The prognostic value of serum troponin T in unstable angina. N Engl J Med. 1992 Jul 16;327(3):146-50.
  14. 14. Kumar, Amit et al. Acute Coronary Syndromes: Diagnosis and Management, Part I. Mayo Clin Proc. 2009 Oct; 84(10): 917–938.
  15. McCarthy BD et al. Missed diagnoses of acute myocardial infarction in the emergency department: results from a multicenter study. Ann Emerg Med. 1993 Mar;22(3):579-82.
  16. Pope JH et al. Missed diagnoses of acute cardiac ischemia in the emergency department. N Engl J Med. 2000 Apr 20;342(16):1163-70.
  17. Lee TH et al. Clinical characteristics and natural history of patients with acute myocardial infarction sent home from the emergency room. Am J Cardiol. 1987 Aug 1;60(4):219-24.
  18. Morrow, D et al. National Academy of Clinical Biochemistry Laboratory Medicine Practice Guidelines: Clinical Characteristics and Utilization of Biochemical Markers in Acute Coronary Syndromes. Circulation. 2007;115:e356–e375
  19. Apple FS et al. Role of monitoring changes in sensitive cardiac troponin I assay results for early diagnosis of myocardial infarction and prediction of risk of adverse events. Clin Chem. 2009 May;55(5):930-7. doi: 10.1373/clinchem.2008.114728. Epub 2009 Mar 19.
  20. Cheung, D et al. Improving Handoffs in the Emergency Department. Annals of Emergency Medicine 2010, Vol 55, Issue 2.
  21. Sabia P et al. Value of regional wall motion abnormality in the emergency room diagnosis of acute myocardial infarction. A prospective study using two-dimensional echocardiography. Circulation. 1991 Sep;84(3 Suppl):I85-92.
  22. Peels CH et al. Usefulness of two-dimensional echocardiography for immediate detection of myocardial ischemia in the emergency room. Am J Cardiol. 1990 Mar 15;65(11):687-91.
  23. Sasaki H et al. Utility of echocardiography for the early assessment of patients with nondiagnostic chest pain. American Heart Journal 1986 112(3): 494-497.
  24. Kontos MC et al. Early echocardiography can predict cardiac events in emergency department patients with chest pain. Ann Emerg Med. 1998 May;31(5):550-7.
  25. Braunwald E et al. Unstable angina: is it time for a requiem? Circulation. 2013 Jun 18;127(24):2452-7. doi: 10.1161/CIRCULATIONAHA.113.001258.
  26. Braunwald E et al. ACC/AHA guidelines for the management of patients with unstable angina and non-ST-segment elevation myocardial infarction: executive summary and recommendations. A report of the American College of Cardiology/American Heart Association task force on practice guidelines (committee on the management of patients with unstable angina). Circulation. 2000 Sep 5;102(10):1193-209.
  27. Bonaca MP et al. Prognostic performance of a high-sensitivity assay for cardiac troponin I after non-ST elevation acute coronary syndrome: Analysis from MERLIN-TIMI 36. Eur Heart J Acute Cardiovasc Care. 2015 Oct;4(5):431-40. doi: 10.1177/2048872614564081. Epub 2014 Dec 23.
  28. Eggers KM et al. Value of cardiac troponin I cutoff concentrations below the 99th percentile for clinical decision-making. Clin Chem. 2009 Jan;55(1):85-92. doi: 10.1373/clinchem.2007.101683. Epub 2008 Nov 6.
  29. Cannon CP et al. Comparison of early invasive and conservative strategies in patients with unstable coronary syndromes treated with the glycoprotein IIb/IIIa inhibitor tirofiban. N Engl J Med. 2001 Jun 21;344(25):1879-87.
  30. Platelet Receptor Inhibition in Ischemic Syndrome Management (PRISM) Study Investigators. A comparison of aspirin plus tirofiban with aspirin plus heparin for unstable angina. N Engl J Med. 1998 May 21;338(21):1498-505.
  31. Perez MI et al. Effect of early treatment with anti-hypertensive drugs on short and long-term mortality in patients with an acute cardiovascular event. Cochrane Database Syst Rev. 2009 Oct 7;(4):CD006743. doi: 10.1002/14651858.CD006743.pub2.
  32. Antithrombotic Trialists’ Collaboration. Collaborative meta-analysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk patients. BMJ. 2002 Jan 12;324(7329):71-86.
  33. Yusuf S et al. Effects of clopidogrel in addition to aspirin in patients with acute coronary syndromes without ST-segment elevation. N Engl J Med. 2001 Aug 16;345(7):494-502.
  34. Magee KD et al. Heparin versus placebo for acute coronary syndromes. Cochrane Database Syst Rev. 2008 Apr 16;(2):CD003462. 
  35. Anderson HV et al. One-year results of the Thrombolysis in Myocardial Infarction (TIMI) IIIB clinical trial. A randomized comparison of tissue-type plasminogen activator versus placebo and early invasive versus early conservative strategies in unstable angina and non-Q wave myocardial infarction. J Am Coll Cardiol. 1995 Dec;26(7):1643-50.
  36. Fanning JP et al. Routine invasive strategies versus selective invasive strategies for unstable angina and non-ST elevation myocardial infarction in the stent era. Cochrane Database Syst Rev. 2016 May 26;(5):CD004815.  
  37. Arbab-Zadeh, Armin. Stress testing and non-invasive coronary angiography in patients with suspected coronary artery disease: time for a new paradigm. Heart Int. 2012 Feb 3; 7(1): e2.
  38. Shaw LJ et al. Use of a prognostic treadmill score in identifying diagnostic coronary disease subgroups. Circulation. 1998 Oct 20;98(16):1622-30.
  39. Bholasingh R et al. Prognostic value of predischarge dobutamine stress echocardiography in chest pain patients with a negative cardiac troponin T. J Am Coll Cardiol. 2003 Feb 19;41(4):596-602.
  40. Colon PJ  et al. Long-Term Value of Stress Echocardiography in the Triage of Patients with Atypical Chest Pain Presenting to the Emergency Department. Echocardiography. 1999 Feb;16(2):171-177.
  41. Sicari R et al. Stress echo results predict mortality: a large-scale multicenter prospective international study. J Am Coll Cardiol. 2003 Feb 19;41(4):589-95.
  42. Lindahl B et al. Risk stratification in unstable coronary artery disease. Additive value of troponin T determinations and pre-discharge exercise tests. FRISK Study Group. Eur Heart J. 1997 May;18(5):762-70.
  43. Natsui S et al. Evaluation of Outpatient Cardiac Stress Testing After Emergency Department Encounters for Suspected Acute Coronary Syndrome. Ann Emerg Med. 2019 Apr 5. pii: S0196-0644(19)30054-X.
  44. Foy AJ et al. Noninvasive Testing in Emergency Department Patients with Low-Risk Chest Pain: Does the Evidence Support Current Guidelines? Cardiol Rev. 2016 Nov/Dec;24(6):268-272.
  45. Sandhu AT et al. Cardiovascular Testing and Clinical Outcomes in Emergency Department Patients With Chest Pain. JAMA Intern Med. 2017 Aug 1;177(8):1175-1182.
  46. Manini AF et al. Adverse cardiac events in emergency department patients with chest pain six months after a negative inpatient evaluation for acute coronary syndrome. Acad Emerg Med. 2002 Sep;9(9):896-902.
  47. Ringqvist, I et al. Prognostic value of angiographic indices of coronary artery disease from the Coronary Artery Surgery Study (CASS). J Clin Invest. 1983 Jun; 71(6): 1854–1866.
  48. Thompson, A et al. Antihypertensive Treatment and Secondary Prevention of Cardiovascular Disease Events Among Persons Without Hypertension. A Meta-analysis. JAMA. 2011 Mar 2; 305(9): 913–922.
  49. Pedersen TR et al. High-dose atorvastatin vs usual-dose simvastatin for secondary prevention after myocardial infarction: the IDEAL study: a randomized controlled trial. JAMA. 2005 Nov 16;294(19):2437-45.
  50. Bangalore S et al. β-blockers and cardiovascular events in patients with and without myocardial infarction: post hoc analysis from the CHARISMA trial. Circ Cardiovasc Qual Outcomes. 2014 Nov;7(6):872-81.
  51. Suzuki T et al. Frequency and impact of lifestyle modification in patients with coronary artery disease: the Japanese Coronary Artery Disease (JCAD) study. Am Heart J. 2012 Feb;163(2):268-73.
  52. Boden WE et al. Optimal medical therapy with or without PCI for stable coronary disease. N Engl J Med. 2007 Apr 12;356(15):1503-16.
  53. Stergiopoulos K et al. Percutaneous coronary intervention outcomes in patients with stable obstructive coronary artery disease and myocardial ischemia: a collaborative meta-analysis of contemporary randomized clinical trials. JAMA Intern Med. 2014 Feb 1;174(2):232-40.
  54. Long-term results of prospective randomised study of coronary artery bypass surgery in stable angina pectoris. European Coronary Surgery Study Group. Lancet. 1982 Nov 27;2(8309):1173-80.

Title image by Sandro Kradolfer on Unsplash

Cite this article as:
Murchison, D. Does Observation for ACS Makes Sense?, First10EM, June 15, 2020. Available at:

Leave a Reply

10 thoughts on “Does Observation for ACS Makes Sense?”

%d bloggers like this: