Ok here we are the main show, the reason for reading, the Trump of testing…well you get the idea. The whole goal of a protocol is to identify a low risk group of patients and then do something to make them even lower risk (<1%) to send them home without admitting them all. So first who is low risk. Well from all the above studies it really seems like all these patients tested are already pretty low risk. I mean most of the MACE was found on index visits and 30 day mortality is really low.
CLINICAL DECISION TOOLS
To identify patients at low risk for ACS we need some sort of clinical decision tool (CDT) or physician Gestalt (assuming doctors are able to make decisions without a lab test). Be warned because there are a million different chest pain CDT’s. Remember that we need a CDT because for a baysian type analysis we need to identify the pre-test probability (incidence) of disease then we can use a couple of troponin and ECGs to come up with a post test probability. So to that end what CDT’s are available to date. Well there is physician gestalt and a myriad CDT’s e.g (GRACE, TIMI, VANCOUVER, HEART, EDACS, etc.) which are usually an acronym for their components. As far as the literature goes probably TIMI and HEART are most popular, however in practice physicians seem to have a “whole lotta love” for the HEART score. I will say one important consideration is to always keep a checklist handy when using any CDT as it is easy to forget an element. In this review of clinicians ability to perform a HEART score, it was incorrectly scored in 15.2% of cases and most errors occurred in the Risk factors’ element (18). Another interesting factoid from the Dubin study was that clinicians with 10-15 years of experience had a significantly higher proportion of MACE in their admitted cohort (15% vs 6%) than clinicians with 0-5 years experience. Meaning experienced clinicians might be better at picking the sick patients(18).
If I had to guess, other than a nominal difference likely most CDT’s probably perform similarly if used without serial lab testing. The HEART (0-3) and TIMI (0-1) scores perform pretty similarly in this 2015 meta-analysis with -LR of 0.2 and 0.3 respectively (27). In another study of 255 patients comparing HEART score or Clinician Gestalt (using all data including troponin and ECG in the ED), the authors looked at acute MI or MACE 6 weeks after presentation. In this group 29% had MACE and 26% had acute MI. HEART score identified 33% of total patients to be low risk and Gestalt identified 31% of patients to be low risk. Of the cohort of Gestalt patients 57% of those were also low risk by HEART. For HEART 5/75 (6%) patients had ACS in the next 6 weeks vs the Gestalt which had 10/79 (13%) patients had ACS. This gave a -LR for HEART score of 0.15 vs a -LR for Gestalt of 0.35 (28). That’s not bad considering they used both attending and PGY1-4 residents from the Netherlands (isn’t Amsterdam there?). The authors also note that these differences were not statistically significant. However, there its not usual for clinicians to stop after one troponin in these types of patients, so lets look at these in combination with serial troponin.
ACCELERATED DIAGNOSTIC PROTOCOLS
These protocols are known as “rapid rule outs” or accelerated diagnostic protocols (ADP). All the ones I will discuss will utilize a “conventional” troponin assay and not the “high sensitivity assay” since those are newly approved and not universally available. The first large one was the ASPECT Trial in 2011 and looked at an observational cohort of 3582 patients followed for 30 days. The classified 352 of these patients as “low risk” by a TIMI score =0 then did two conventional troponins 2 hours apart. Now this was done in New Zealand and Southeast Asia so the group was a bit sicker with a total of 11.8% of all comers having MACE. Of the 352 patient in the protocol 0.9% had MACE. This ADP has a -LR of 0.1 but a specificity of only 14% so a lot of patients wont fall into the ADP defined by TIMI = 0 (22). Notice that this ADP only identified 10% of the cohort as LOW RISK because they used a TIMI of 0. Looks like a decent start since with have a <1% MACE but that was only 10% of the included population.
The next large ADP was done also done by Than from above and also recruited patients from 2 Urban hospitals, New Zealand and Australia. This group was also “sick” and had a MACE of 15.3%. They did 2 iterations one with TIMI =0 and a second with TIMI = 1. In the TIMI =0 ADP they found a -LR of 0.01 and a Sp of 23.4% with 1/392 (0.25%) having MACE. In the recalculation with TIMI = 1, they found a Sn of 97%, Sp of 44.8% and 38% of the population able to be categorized as low risk. This time they had 9/750 (1.2%) false negative results (23). So TIMI = 1 still seemed to perform well and got us to “around” 1%.
Than et al. performed another ADP however this one was an RCT which put patients into a 2 hour ADP or 12 hour observation period. In this trial they randomized 270 patients to each group. This 540 patient cohort also had a high incidence of MACE (14.9%) overall. In the ADP they identified 19% of patients discharged without MACE at 30 days vs observational group which discharged only 11%. Again a small proportion of the total chest pain patients that present but great outcomes of no MACE in this ADP (24).
Another trial of ADPs was called the IMPACT trial which was a NON-randomized trial that looked at low, intermediate, high risk patients (E) by classifying them based on the “IMPACT” protocol (D). In this schema 244/1366 (18%) of patients were classified as low risk and 789/366 (58%) were classified as intermediate risk and the rest as high risk. The low and intermediate risk patients then had normal serial ECG’s and 2 troponin separated by 2 hours. Interesting in the low risk group 180/244 (73%) had no further testing and no 30 day ACS events, in the intermediate risk group 106/789 had no further testing and of those only 1 had a 30 day ACS event. Also interesting is that in the intermediate group 683/789 had some provocative testing and 13/683 (1.9%) had 30 day ACS events (25).
The next study I’ll discuss on this topic is the comparison of TIMI, HEART and clinician gestalt by Singer et al. which was a prospective multi center observational study performed in the United States. Here, 434 patients were enrolled and scores were prospectively calculated with 2 troponins THREE hours apart with a primary outcome of acute MI. Of the total of 434 patients 80 (18%) had acute MI. That’s pretty high for a U.S. study. The group prospectively described by TIMI=0 and 2 troponins, identified only 30 of the 434 (7%) patients but had zero missed acute MI’s. The group prospectively described by HEART score of <=3 and 2 troponins, identified 138/374 (36%) however had 4% missed MI. The group prospectively described by Clinician Gestalt and 2 troponins as low risk, identified 118/394 (30%) of patients and had zero missed acute MI. Note of all the studies discussed (and on the inter-webs) a 4% miss rate for HEART score <=3 is really high and not the norm. This likely has to do with how high the incidence of acute MI (18%) was in this cohort. However another possibility and a REALLY important and interesting fact here is that “It is also worth noting that half or more of the false negative cases for … HEART score with 2 [conventional] Tn measurements, (3 cases) involved the use of point of care assays.” This is definitely something that should be studied and considered! Subsequently, the authors lowered the cut point of the HEART score to now re-classify low risk as <=2. Now they found ZERO missed acute MI’s but only classified 19% of patients as low risk, as opposed to 36% when a cutoff of 3 is used (26). So the winner here? Clincian Gestalt and 2 troponins with the highest ration of low risk categorized patients to misses. Again Yay us!
FURTURE DIRECTIONS (aka My Wish List)
So where do we stand with all this. I think the answer still hasn’t been found exactly. But I think some one take away point is that we over work up No- and Low- risk chest pain patients who could be sent home with no further testing. So the answer to this would be EVERY ED have a protocol that is agreed upon by all services, written out, and is easily available. This would promote conformance while developing a “standard of care” for that community. As a starting point, I like this one by Scheuermeyer with some minor changes (30). Based on my review of the literature, the second troponin is what is likely making the patient low risk and every study done should share that data point in the results. We should reserve the CCTA and observation units for patients with actual risk. Stress testing is useless. The AHA guidelines should be changed to remove the wording of any provocative testing time limit and replace it with “consider provocative testing at some point down the line if the clinician at presentation feels it is necessary”. I’d also like to see the AHA (since they are the “Standard of Care”) support adding somewhere to a recommendation that a protocol that finds a threshold for missed AMI or MACE be <2%. Yes I said <2%! Because isn’t that what we use for PE (29)?!
That being said, I think a great study or protocol would be:
1. Determine what your primary end-point would be (acute MI, MACE, etc)
2. Determine the incidence of that end-point be at YOUR locale.
3. Stratify patients to no risk, low risk, intermediate risk, and high risk.
4. Discharge No risk patients right away (age <40, no suspicious clinical findings, normal ECG, and no other diagnosis of “deadly chest pain) with no lab work.
5. Stratify Low risk patients by TIMI (0 or 1), HEART (2 or 3) or Clinician Gestalt (only for providers with >5 years experience).
6. Perform a 2 hour (or 3 hour) troponin and repeat ECG on those low risk patients with chest pain and discharge them home if both troponins are normal.
7. Intermediate Risk patients can get a CCTA, ED observation unit, cardiology consult or something; maybe even do a 6 hour trop (This area needs the most study).
8. I think we all know what to do with high risk patients.
ADP = accelerated diagnostic protocols aka rapid rule out
CAS = coronary calcium score
CCTA = Coronary CT Angiogram
CDT = clinical decision tool
FFR = Fractional Flow Reserve
LR = Likelihood Ratio (Positive or Negative)
MACE = major cardiac adverse events usually including: cardiac death, non-fatal MI, dysrhythmias, unstable angina requiring hospitalization, and late revascularization
MI = AMI = myocardial infarction
MPI = myocardial perfusion scan
PCI = Percutaneous Coronary Intervention
SE = Stress echo
Sn = Sensitivity
Sp = Specificity
ST = Stress Testing
A. In 2000, the definition of acute myocardial infarction changed to: ‘typical rise and gradual fall of cardiac troponin, or more rapid fall of CK-MB, with at least one of the following: ischemic symptoms; the development of pathological Q waves; ECG changes indicative of ischemia (ST-segment elevation or depression); coronary artery intervention (2)
B. If you are reading this note then either you like my sarcasm or haven’t gotten my feelings for ST. Obviously ST in age group less than 40 is about as useful as a rubber ladder on a hot day. So it kills me when I see this in the community. I think this article sums it up nicely: “Routine stress testing added little to the diagnostic evaluation of this patient group and was falsely positive in all patients who consented to diagnostic coronary angiography” (13).
C. For ease of reading (and writing) all numbers are expressed as integers rounded to the nears whole number unless written with a decimal point for shock and awe.
D. This risk stratification has some clinical face value but is clearly unique to Australasia and has unknown generalizability to the US.
E. Risk was defined in a very complicated manner as follows:
1. Intermediate Risk: No high risk features, and any of:
* Chest pain or discomfort during rest within past 48 h, or pain that was repetitive or prolonged, but currently resolved
* Age > 65 years
* Coronary heart disease: prior myocardial infarction with LVEF ! 40% or known coronary lesion > 50% stenosed
* Two or more of: known hypertension, family history, active smoking, hyperlipidaemia
* Diabetes (with atypical symptoms of ACS)
* Chronic kidney disease, eGFR < 60 mL/min/1.73 m2 (with atypical symptoms of ACS)
* Prior aspirin use
2. Low Risk:
1. * No intermediate or high risk features, and any of:
* Onset of angina symptoms during the past month
*Worsening in severity or frequency of angina
*Lowering of angina threshold
1. Pope JH et al. Missed Diagnoses of Acute Cardiac Ischemia in the Emergency Department. NEJM. 2000 April; 342(16):1163-70.
2. Alpert JS. Myocardial infarction redefined–a consensus document of The Joint European Society of Cardiology/American College of Cardiology Committee for the redefinition of myocardial infarction. J Am Coll Cardiol. 2000 Sep; 36(3):959-69.
3. Swap. Value and Limitations of Chest Pain History in the Evaluation of Patients With Suspected Acute Coronary Syndromes. JAMA. 2005;294:2623-29
4. Dezman ZD. Utility of the History and Physical Examination in the Detection of Acute Coronary Syndromes in Emergency Department Patients. West J Emerg Med. 2017;18(4)752-60.
5. Pattanshetty DJ, Bhat PK, Gandhi S, Pillai DP, Aneja A. Comparing stress testing and fractional flow reserve to evaluate presence, location and extent of ischemia in coronary artery disease. Indian Heart Journal. 2015;67(1):50-55.
6. Amsterdam. 2014 AHA/ACC Guideline for the Management of Patients With Non–ST-Elevation Acute Coronary Syndromes: A Report of the American College of Cardiology/American Heart Association Task Force on Practice GuidelinesCirculation. 2014;130:e344-e426.
7. Anderson. 2012 ACCF/AHA Focused Update Incorporated Into the ACCF/AHA 2007 Guidelines for the Management of Patients With Unstable Angina/Non–ST-Elevation Myocardial Infarction. Circulation. 2013;127:e000-e000.
8. Froelicher V. The Electrocardiographic Exercise Test in a Population with Reduced Workup Bias: Diagnostic Performance, Computerized Interpretation, and Multivariable Prediction. Ann Intern Med. 1998;128:965-9
9. Gianrossi R, Detrano R, Mulvihill D, Lehmann K, Dubach P, Colombo A, et al. Exercise-induced ST depression in the diagnosis of coronary artery disease. A meta-analysis. Circulation. 1989;80:87-98.
10. Ayerbe et al. Clinical assessment of patients with chest pain; a systematic review of predictive tools. BMC Cardiovascular Disorders (2016) 16:18
11. Pijls. Measurement Of Fractional Flow Reserve To Assess The Functional Severity Of Coronary-Artery Stenoses. N Engl J Med 1996;334:1703-8.
12. Young. Meaning of zero coronary calcium score in symptomatic patients referred for coronary computed tomographic angiography. European Heart Journal Cardiovascular Imaging (2012)
13, 776–785 13. Hermann. The Limited Utility of Routine Cardiac Stress Testing in Emergency Department Chest Pain Patients Younger Than 40 Years. Ann Emerg Med. 2009;54:12-16.
14. Chang. Does Coronary Artery Calcium Scoring Add to the Predictive Value of Coronary Computed Tomography Angiography for Adverse Cardiovascular Events in Low-risk Chest Pain Patients? Academic Emergency Medicine 2011; 18:1065–1071.
15. Litt. CT Angiography for Safe Discharge of Patients with Possible Acute Coronary Syndromes. N Engl J Med 2012;366:1393-403.
16. Hoffman. Coronary CT Angiography versus Standard Evaluation in Acute Chest Pain. N Engl J Med 2012;367:299-308.
17. Redberg. Coronary CT Angiography for Acute Chest Pain. N Engl J Med 2012;367:375-376.
18. Ras M, et al. Secondary analysis of frequency, circumstances and consequences of calculation errors of the HEART (history, ECG, age, risk factors and troponin) score at the emergency departments of nine hospitals in the Netherlands. BMJ Open 2017;7:e017259
19. Nørgaard. Fractional Flow Reserve Derived From Coronary Computed Tomography Angiography in Suspected Coronary Artery Disease The NXT TrialJ Am Coll Cardiol 2014;63:1145–55.
20. Koo. Diagnosis of Ischemia-Causing Coronary Stenoses by Noninvasive Fractional Flow Reserve Computed From Coronary Computed Tomographic Angiograms. J Am Coll Cardiol 2011;58:1989–97.
21. Douglas. Outcomes of Anatomical versus Functional Testing for Coronary Artery Disease [PROMISE Investigators]. N Engl J Med 2015;372:1291-300.
22. Than. A 2-h diagnostic protocol to assess patients with chest pain symptoms in the Asia-Pacific region (ASPECT): a prospective observational validation study. Lancet 2011; 377: 1077–84.
23. Than. 2-Hour Accelerated Diagnostic Protocol to Assess Patients With Chest Pain Symptoms Using Contemporary Troponins as the Only Biomarker. J Am Coll Cardiol 2012;59:2091–8.
24. Than et al. A 2-Hour Diagnostic Protocol for Possible Cardiac Chest Pain in the Emergency Department A Randomized Clinical Trial. JAMA Intern Med. 2014;174(1):51-58.
25. Cullen. Improved Assessment of Chest pain Trial (IMPACT): assessing patients with possible acute coronary syndromes. Med J Aust. 2017 Aug 4;207(5):195-200.
26. A.J. Singer et al. Missed myocardial infarctions in ED patients prospectively categorized as low risk by established risk scores. American Journal of Emergency Medicine 35 (2017) 704–709.
27. Fanaroff. Does This Patient With Chest Pain Have Acute Coronary Syndrome? The Rational Clinical Examination Systematic Review. JAMA. 2015;314(18):1955-1965.
28. Visser A, HEART score and clinical gestalt have similar diagnostic accuracy for diagnosing ACS in an unselected population of patients with chest pain presenting in the ED et al. Emerg Med J 2015;32:595–600
29. Kline. Clinical criteria to prevent unnecessary diagnostic testing in emergency department patients with suspected pulmonary embolism. J Thromb Haemost 2004; 2: 1247–55.
30. Scheuermeyer. Safety and Efficiency of a Chest Pain Diagnostic Algorithm With Selective Outpatient Stress Testing for Emergency Department Patients With Potential Ischemic Chest Pain. Ann Emerg Med. 2012;59:256-264.