High-Sensitivity Cardiac Troponin in the Evaluation of Possible AMI

Introduction

High-sensitivity (hs) cardiac troponin (cTn) assays expedite the evaluation of patients with possible acute coronary syndromes (ACS) in the emergency department. Rapid screening protocols with hscTn have been proposed for patients for whom ruling-in or ruling-out acute myocardial infarction (AMI) is the primary issue. These protocols have not included the entire range of patients with possible AMI, such as those with end-stage renal disease, the critically ill, or those who present atypically. Protocols have early rule-out and early rule-in components for acute myocardial injury. The absence of acute myocardial injury does rule out AMI. However, having elevations simply means myocardial injury is present, with a specificity for AMI in published studies ranging from 60-75%. The diagnosis of AMI requires a rising and/or falling pattern of cTn with at least one value exceeding the 99th percentile limit of a normal reference population, i.e., the presence of acute cardiac injury but also a clinical situation where symptoms, electrocardiogram (ECG), or imaging findings of ischemia are present.¹ An isolated increase in cTn is not sufficient.¹ This brief review reflects on the benefits and gaps in the proposed approaches presently available.

cTn Assays

Guidelines recommend preferentially usable hscTn assays.² Assays should have imprecision measured as the coefficient of variation <10% at the 99th percentile value.³ HscTn assays measure cTn concentrations fivefold to 100-fold lower than conventional assays and should be able to detect cTn concentrations below the 99th percentile in >50% of normal individuals.² Currently, there are five hscTn assays that may meet these criteria.⁴ There is some controversy whether the Roche fifth-generation cTnT meets this criterion.⁵

HscTn values should be reported in whole numbers to avoid the confusion of multiple decimal places. For the hscTnT assay approved in the United States, partial harmonization between the fourth- and fifth-generation assays is present. Values >0.1 ng/ml with the fourth-generation assay are multiplied by 1,000 to obtain a hscTnT value. However, this is not the case with values below 0.1 ng/ml. A fourth-generation cTnT value of 0.01 ng/ml comports to a hscTnT value of 30 ng/L; a value of 0.03 ng/ml comports to a hscTnT value of 53 ng/L. Undetectable values (i.e., below the limit of detection of the assay, which is between 3 and 5 ng/L) can safely rule out MI using a single determination in low-risk patients.⁶Unfortunately, the US Food and Drug Administration will not allow the reporting of values below 6 ng/L, making this strategy problematic in the United States. Studies attempting to advocate for values <6 ng/L as cut-off values are not convincing.

False positive elevations in hscTn can be due to non-repeatable increases (so called "flyers") and/or secondary to heterophile antibodies and/or macrokinases. With hscTnT, chronic skeletal muscle disease also can cause elevations.⁷ False negatives are rare and usually due to anti-cTn antibodies. HscTnT assay values are lowered by hemolysis.⁵ Supplemental biotin ma y also interfere with the assay and lower values.

Not all HscTn assays are biologically equivalent. We endorse the International Federation of Clinical Chemistry recommendations for determination of the 99th percentile upper reference limit (URL), which include biomarkers but not imaging.⁸

Triage of Patients With Possible ACS

The majority of patients with unstable plaque events (type 1 AMI) should have a rising or falling pattern of hscTn values. However, clinicians must be aware that the rise is more rapid than the fall and take that into account when evaluating those who present more than 12 hours after the possible onset of symptoms. Because cTn release depends on the flow in the culprit vessel, open vessels will provide an earlier and more robust signal than closed ones. Overall, the frequency of elevations due to ACS should not increase significantly with hscTn assays.⁹ There will be an increase in the number of non-ST-segment elevation myocardial infarction diagnoses instead of unstable angina diagnoses. Many of these new elevations will represent type 2 AMIs due to supply-demand imbalance, and many will be in women. Women have lower cTn values at baseline and less extensive fixed coronary artery disease (CAD) when they present with possible AMI.¹⁰ If elevations of cTn occur with or without a changing pattern of values in the absence of clinical, ECG, and/or imaging evidence of ischemia, structural heart disease and other non-ACS conditions⁴ should be considered (Table 1).

Table 1: Noncoronary Elevation of cTn

Acute Conditions	Chronic Conditions
Imbalance of Demand/Supply
Tachy- or bradyarrhythmias	Tachy- or bradyarrhythmias
Aortic dissection
	Severe aortic valve disease
Cardiogenic, hypovolemic and septic shocks
	Hypertrophic cardiomyopathy
Respiratory failure
	Anemia
	Hypertension
	Left ventricular hypertrophy
Coronary embolism or vasculitis
Coronary spasm	Coronary spasm
Endothelial dysfunction	Endothelial dysfunction
Cocaine use
Non-ischemic Myocardial Damage
Cardiac contusion
Cardiac surgery
Radiofrequency or cryoablation therapy
Pacing or defibrillation shocks	Pacing or defibrillation shocks
Rhabdomyolysis with cardiac involvement
Myopericarditis	Myopericarditis
Cardiotoxic agents	Cardiotoxic agents
Some chemotherapeutics	Some chemotherapeutics
Carbon monoxide poisoning
Multifactorial Causes of Myocardial Damage
Heart failure	Heart failure
Takotsubo cardiomyopathy
Severe pulmonary embolism
	Pulmonary hypertension
Extreme exertion
Sepsis
Gastrointestinal bleeding
Rhabdomyolysis without cardiac involvement
Renal failure
	Infiltrative diseases such as sarcoidosis or amyloidosis
Severe acute neurological diseases, such as stroke or trauma
Skeletal myopathies

Ruling Out AMI Based on a Single hscTn Determination

Investigations show that when hscTn is undetectable, AMI is unlikely.¹¹ Validation studies for the single-hscTn approach have included only small numbers of patients with AMI who have had blood samples obtained early after symptom onset.¹² In the TRAPID-AMI (The High Sensitivity Cardiac Troponin T Assay for Rapid Rule-out of Acute Myocardial Infarction) study, the mean time from symptom onset to presentation was 1.9 hours, but it took 1.5 hours to obtain the first sample.¹³ Thus, there is some concern regarding this approach applied to early presenters sampled expeditiously. This approach is problematic in the United States because the lowest reportable value with hscTnT is above the limit of detection. In one validation study using a value of 6 ng/l or below, 8 AMIs were missed.¹⁴ In another study, the approach worked well only when using the relatively higher values for the 99th percentile URL.¹⁵ We have concerns about this approach because the confidence intervals around the values given the imprecision of the assay at these levels are larger than ideal. With time, the analyzers will increase precision at these low values, making this approach more attractive. A single-sample approach using a value <99th percentile URL can be used in patients with symptom onset >6 hours from presentation.

Rapid Rule-Out of AMI

Protocols trying to distinguish minor changes in hscTn (between <3 ng/L and >5 ng/L) have been suggested for use by some algorithms at 1 hour. In our view, the precision of the assays is not adequate to make these small distinctions accurately.⁴ We advocate the 2-hour approach, which allows for larger differences. The majority of patients can be triaged within 2 hours. Patients who have normal but increasing values that do not meet criteria for AMI at 2 hours need to be monitored longer. At our institution, we advocate serial hscTnT measurements at 0 and 2 hours to evaluate those patients with possible AMI. With hscTnT, the following most often can rule out AMI:

• Values less than or equal to the sex-specific 99th percentile that we advocate (10 ng/L for women and 15 ng/L for men)
• The absence of a delta of 4 ng/L or greater when integrated with the ECG, patient's history, and validated risk scores.

Roughly 20% of patients may have a change in values >3 ng/L but not >10 ng/L and will need a third sample. For high-risk patients and late presenters, clinical judgment should prevail. A flow chart of proposed use of hscTnT is depicted in Figure 1. The safety of this approach is currently unclear. Importantly, lower hscTn values have less likelihood of CAD after stress testing¹⁶ and better long-term prognosis. Conversely, the higher the hscTn value, the more adverse the prognosis even with values that are still within the normal range.¹⁷

Figure 1

Flow chart of hscTnT use in patients presenting with chest pain. Reproduced with permission from Sandoval et al.¹⁸

Ruling In AMI

A rising pattern with at least one value above the 99th percentile URL in patients with evidence of acute ischemia is necessary for the diagnosis of AMI.¹ Individualization is necessary in the elderly, women, and diabetic and postoperative patients who may be pauci-symptomatic.⁴

An absolute baseline cut-off value of hscTnT of 52 ng/L has been advocated in the rapid algorithms for the diagnosis of acute myocardial injury. This criterion is reasonable for patients with chest pain, but it will be too low for patients with chronic elevations, the elderly, and those who are critically ill. We have proposed a cut-off value of 100 ng/L as diagnostic for acute myocardial injury in the absence of end-stage renal disease and AMI in ischemic patients. If one extrapolates this cut-off to chest-pain patients, it accurately predicts AMI with a positive predictive value of 90% and the specificity >99%.¹⁹

Change criteria have also been advocated to diagnose acute myocardial injury. A change of 50% or more predicated on the reference change interval (the change in values that guarantees the change is not due to variation alone) has been suggested when baseline values are near the 99th percentile URL. However, when the baseline value is elevated, changes of only 20% are recommended to maintain sensitivity. The optimal delta for the diagnosis of AMI depends on the intended rule-in or to rule-out approach. Sensitivity and specificity data are recapitulated elsewhere.⁴ The use of an absolute delta change is superior to the percentage change because it provides a changing set of criteria depending on the baseline value, thus preserving sensitivity. When the absolute or relative delta is less than conjoint biological and analytical variation, some patients will have increased values due to biologic and analytical variation alone. A 10 ng/L delta for hscTnT over a period of 2 hours improves diagnostic specificity.

Patients who present late after the onset of symptoms (>12 hours) can have hscTn elevations without an easily discernible changing pattern because they are on the downslope of the time-concentration curve, which is slower than the upstroke, and thus need more observation. These patients could have structural heart disease alone or structural heart disease and unstable angina, or they could have AMI and be presenting late.

Sex is Important

Women have a higher burden of non-obstructive CAD and plaque erosion.²⁰ Their normal cTn values are lower for every assay studied. When they present with AMI, hscTn values are often lower. We endorse the use of sex-specific cut-off values with hscTn assays despite some controversy with the hscTnT assay. We advocate cut-offs of 10 ng/L for women and 15 ng/L for men for the hscTnT assay based on numerous normal-range studies.

Age

For hscTn assays, 99th percentile URL values are established from a reference population of subjects without heart disease, with ages 20-70 years.¹⁰ However, the majority of patients presenting to the emergency department have multiple comorbidities.

When one corrects for comorbidities, including with imaging, the differences between age groups are markedly mitigated.²¹ Thus, we do not recommend the use of different cut-off values for older patients to take into account their comorbidities. Healthy older patients may also be disadvantaged.

Patients Who do not Rule Out or Rule In

Such patients will require a third hscTnT sample. A delta of 12 ng/L at 6 hours identifies those with acute myocardial injury.

Symptomatic patients who do not exceed the 99th percentile URL but do exhibit a significant delta and symptomatic patients who exceed the 99th percentile URL but fail to have a significant delta change value are high risk and warrant additional testing.

Therapeutic Implications of hscTn for Treatment of AMI

Distinctions between type 1 and type 2 MI can be challenging. With use of hscTn assays, the frequency of type 2 MIs will increase. These patients often have critical illness, causing supply-demand imbalance. Some may also have CAD. Many of them do not require an invasive approach to their myocardial injury or even their type 2 AMI but instead require only control of the underlying problem leading to the event. This contrasts with those with type 1 AMI who benefit from an invasive strategy. That is why the ability to extrapolate from earlier assays to the present hscTn one is important. When the values are above 0.01 ng/ml or 30 ng/l with hscTnT, the data suggesting that an invasive strategy is superior holds. That has not been proven in randomized clinical trials with hscTn values below that level.

Conclusions

HscTn assays will help triage patients with ACS more accurately and rapidly than prior assays. They will also improve risk stratifications of patients presenting with chest pain, but the rapid algorithms have some gaps for clinicians to be aware of. For example, clinicians should not over-respond to small changes in values. In addition, alternative explanations for cardiac injury should be considered when hscTn is elevated but ischemia is not clearly present.

References: American College of Cardiology

1. Giannitsis E, Katus HA. Cardiac troponin level elevations not related to acute coronary syndromes. Nat Rev Cardiol 2013;10:623-34.
2. Apple FS. A new season for cardiac troponin assays: it's time to keep a scorecard. Clin Chem 2009;55:1303-6.
3. Jaffe AS, Ravkilde J, Roberts R, et al. It's time for a change to a troponin standard. Circulation 2000;102:1216-20.
4. Vasile VC, Jaffe AS. High-Sensitivity Cardiac Troponin for the Diagnosis of Patients with Acute Coronary Syndromes. Curr Cardiol Rep 2017;19:92.
5. Gunsolus IL, Jaffe AS, Sexter A, et al. Sex-specific 99th percentiles derived from the AACC Universal Sample Bank for the Roche Gen 5 cTnT assay: Comorbidities and statistical methods influence derivation of reference limits. Clin Biochem 2017;50:1073-7.
6. Pickering JW, Than MP, Cullen L, et al. Rapid Rule-out of Acute Myocardial Infarction With a Single High-Sensitivity Cardiac Troponin T Measurement Below the Limit of Detection: A Collaborative Meta-analysis. Ann Intern Med 2017;166:715-24.
7. Jaffe AS, Vasile VC, Milone M, Saenger AK, Olson KN, Apple FS. Diseased skeletal muscle: a noncardiac source of increased circulating concentrations of cardiac troponin T. J Am Coll Cardiol 2011;58:1819-24.
8. Apple FS, Sandoval Y, Jaffe AS. In Reply. Clin Chem 2017;63:1167-70.
9. Korley FK, Jaffe AS. Preparing the United States for high-sensitivity cardiac troponin assays. J Am Coll Cardiol 2013;61:1753-8.
10. Giannitsis E, Kurz K, Hallermayer K, Jarausch J, Jaffe AS, Katus HA. Analytical validation of a high-sensitivity cardiac troponin T assay. Clin Chem 2010;56:254-61.
11. Agewall S, Giannitsis E, Jernberg T, Katus H. Troponin elevation in coronary vs. non-coronary disease. Eur Heart J 2011;32:404-11.
12. Boeddinghaus J, Nestelberger T, Twerenbold R, et al. Direct Comparison of 4 Very Early Rule-Out Strategies for Acute Myocardial Infarction Using High-Sensitivity Cardiac Troponin I. Circulation 2017;135:1597-611.
13. Mueller C, Giannitsis E, Christ M, et al. Multicenter Evaluation of a 0-Hour/1-Hour Algorithm in the Diagnosis of Myocardial Infarction With High-Sensitivity Cardiac Troponin T. Ann Emerg Med 2016;68:76-87.e4.
14. Peacock WF, Baumann BM, Bruton D, et al. Efficacy of High-Sensitivity Troponin T in Identifying Very-Low-Risk Patients With Possible Acute Coronary Syndrome. JAMA Cardiol 2018;3:104-11.
15. Cullen L, Than M, Peacock WF. Undetectable hs-cTnT in the emergency department and risk of myocardial infarction. J Am Coll Cardiol 2014;64:632-3.
16. Lee G, Twerenbold R, Tanglay Y, et al. Clinical benefit of high-sensitivity cardiac troponin I in the detection of exercise-induced myocardial ischemia. Am Heart J2016;173:8-17.
17. Than MP, Aldous SJ, Troughton RW, et al. Detectable High-Sensitivity Cardiac Troponin within the Population Reference Interval Conveys High 5-Year Cardiovascular Risk: An Observational Study. Clin Chem 2018;64:1044-53.
18. Sandoval Y, Jaffe AS. Using High-Sensitivity Cardiac Troponin T for Acute Cardiac Care. Am J Med 2017;130:1358-65.e1.
19. Reichlin T, Schindler C, Drexler B, et al. One-hour rule-out and rule-in of acute myocardial infarction using high-sensitivity cardiac troponin T. Arch Intern Med2012;172:1211-8.
20. Vasile VC, Jaffe AS. High-Sensitivity Cardiac Troponin for the Diagnosis of Patients with Acute Coronary Syndromes. Curr Cardiol Rep 2017;19:92.
21. Della Rocca DG, Pepine CJ. What causes myocardial infarction in women without obstructive coronary artery disease? Circulation 2011;124:1404-6.
22. Reiter M, Twerenbold R, Reichlin T, et al. Early diagnosis of acute myocardial infarction in the elderly using more sensitive cardiac troponin assays. Eur Heart J 2011;32:1379-89.