From the last two posts (Part 1 and Part 2) we know that observing patients for chest pain to “rule out ACS” may no longer be relevant now that we have new diagnostic modalities and updated evidence. 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. We then reasoned that even if unstable angina is real, it is technically 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 discussion. 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). 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. 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? They had a 9% 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 really 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 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.
Will coronary angiography help better risk stratify a patient’s 6-week MACE? The CASS trial in 1983 showed that results of cardiac catheterization help predict 6-year mortality risk, but not short-term risk (47). Let’s say that a patient undergoes stress testing and it shows reversible ischemia. He then goes for catheterization which shows severe CAD. The cardiologist decides to either start him on medical therapy, perform percutaneous coronary intervention (PCI), or refer him for coronary artery bypass surgery (CABG).
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.
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). 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.
Conclusion SUMMARY 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 technically 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] ECG, 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. The results have really surprised me. It seems strange that a blog post might be more reliable than a New England Journal of Medicine article. Publication bias is real. REFERENCES (continued from OBS for ACS Part 2) 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.
Charles Murchison
Latest posts by Charles Murchison (see all)
0 Comments