Acute myocardial infarction (AMI) is a major cause of morbidity and mortality worldwide. Accordingly, prompt identification of patients with AMI is critical, as established therapies exist that are of proven benefit.1,2 More than 15 million patients per year present to emergency departments with symptoms compatible with acute myocardial ischaemia in the US and Europe; only 20–30%, however, are eventually proven to have acute ischaemic heart disease due to AMI in half of these patients.1 Crowding in emergency departments, in part due to delays in evaluating this large number of patients, is a major problem.
Electrocardiography and cardiac troponin (cTn) now form the diagnostic cornerstones of clinical assessment.1,2 Often, an electrocardiogram (ECG) alone is insufficient to diagnose AMI: first, because significant ECG changes are absent in many AMI patients and, second, because S-T segment deviation is observed in many other conditions.1–3 cTns, structural proteins unique to the heart, are sensitive and specific biochemical markers of myocardial damage.1,2,4 Fully automated, rapidly available contemporary assays for cTn are superior to other available biomarkers for the diagnosis of AMI.5 In addition, elevated cTns identify patients with acute coronary syndrome at high risk because of adverse anatomy and pro-coagulant activity.6 These patients benefit from aggressive anticoagulation, an early invasive strategy and glycoprotein IIb/IIIa blockade.1,2,4
One limitation of contemporary cTn assays is that circulating levels may not become detectable for three to four hours.1,2,7 At times, diagnosis requires serial sampling for six to 12 hours. Delays in diagnosing disease (‘ruling in’) hold back prompt treatment, while delays in excluding disease (‘ruling out’) interfere with evaluation of alternative diagnoses – both of which contribute to overcrowding in emergency departments and costs exceeding several billion US dollars per year.8
Most standard cTn assays are unable to measure cTn in healthy persons.9 Thus the true upper normal limit, the 99th percentile, is defined by the assay and not by biology. Recently, improvements in cTn assay technology have allowed manufacturers to provide automated assays that meet the quality specifications set out by the International Federation of Clinical Chemistry.2 These assays have a lower limit of detection below the 99th percentile of a normal reference population and a low level of imprecision (coefficient of variation <10%) at that value.10 This feature is critical, since cTn values above the 99th percentile of a normal reference population are, at present, a conditio sine qua non for the diagnosis of AMI.
Cardiac Troponin is an Integral Part of the Universal Definition of Acute Myocardial Infarction
cTnI and cTnT are the current, gold-standard biomarkers for the detection of myocardial necrosis, of which AMI is the most important cause. In 2000, experts from the European Society of Cardiology, the American Heart Association and the American College of Cardiology agreed on a universal definition of AMI based on cTn. These recommendations have changed slightly since their publication.2 The definition requires a rising and/or falling pattern of cTn values with at least one value above the 99th percentile of a normal reference population in patients with evidence of myocardial ischaemia.
Evidence of myocardial ischaemia includes:
- typical symptoms;
- typical ECG changes;
- new Q waves;
- new loss of viable myocardium; and
- new regional wall motion abnormalities.
In a patient presenting with chest pain, a rise (and/or fall) in cTn above the 99th percentile of a normal reference population is a prerequisite for the clinical diagnosis of AMI. Assuming appropriate timing of the evaluation (values near peak may not appear to rise and/or fall), the diagnosis of AMI should not be made if the patient does not show a rise (and/or fall) of cTn values above the 99th percentile of a normal reference population, irrespective of the clinical presentation, ECG findings, or the findings of coronary angiography.
Sensitive and Highly Sensitive Cardiac Troponin Assays
Newer cTn assays have limits of detection far below the 99th percentile of a normal reference population.10–15 However, the terms ‘sensitive’ and ‘highly sensitive’ have been used indiscriminately by their manufacturers. There is no consensus about when the terms ‘sensitive’, ‘highly sensitive’ or ‘ultra sensitive’ should be applied. Some assays allow the detection of cTn in about 50% of a normal reference population, others in up to 90% of a normal reference population.9 One reasonable option is to use ‘sensitive’ for the former and ‘highly sensitive’ for the latter.9 However, it is unclear whether these analytical differences improve their clinical performance. Evidence from one large, multicentre study suggests that this might not be the case for the diagnosis of AMI.16
More Sensitive, Contemporary Cardiac Troponin Assays Facilitate Early Diagnosis of Acute Myocardial Infarction
Two large, prospective, multicentre studies have unequivocally documented that the present generation of sensitive cTn assays has a higher diagnostic accuracy than conventionally used assays. These data build on pilot data from several previous studies.7 Our group examined the diagnostic accuracy of two contemporary sensitive cTn assays and two novel assays, one of which is said to be ‘highly sensitive’ on the basis of detectable cTn values in almost all normal subjects at the time of emergency department presentation.
cTn levels were determined in a blinded fashion using: the Architect® cTnI assay (Abbott Laboratories, Illinois, US); the highly sensitive cTnT and cTnI assays (Roche Diagnostics, Basel, Switzerland); the Centaur TnI Ultra (Siemens Healthcare Diagnostics, Erlangen, Germany); and a standard assay, the fourth-generation cTnT assay (Roche Diagnostics, Basel, Switzerland) in 718 consecutive patients presenting with symptoms suggestive of AMI.
The final diagnosis was adjudicated by two independent cardiologists based on all available medical records pertaining to the patient, including standard cTn assay values. In full accordance with current guidelines,2 the diagnosis of AMI required at least one value above the 99th percentile (with <10% coefficient of variation) as well as a rising and/or falling pattern with the standard cTn assays. AMI was the adjudicated diagnosis in 123 patients (17%). The diagnostic accuracy, as quantified by the area under curve (AUC) of the receiver operating characteristic (ROC) curve, were: 0.96 for the Abbott Architect® cTnI assay; 0.96 for the Roche highly sensitive cTnT assay; 0.95 for the Roche highly sensitive cTnI assay; and 0.96 for the Siemens Centaur TnI Ultra assay. All were significantly higher than the standard, fourth-generation Roche cTnT assay (0.90). The benefit of sensitive cTn assays was most pronounced in patients presenting early after the onset of chest pain.16 These findings were confirmed in a second multicentre study.17
The superiority of more sensitive assays, especially in patients with recent-onset chest pain, could facilitate the more rapid identification of patients with AMI, thereby improving the effectiveness of evidence-based AMI treatment.1,2 In addition, this approach could allow a more reliable rule out of AMI. Used in conjunction with clinical assessment and ECG, this approach should reduce the percentage of patients requiring prolonged evaluation in the emergency department, with attendant savings in cost and a reduction in emergency department crowding.
Cardiac Troponin in Chronic Cardiac Disease
Sensitive cTn assays detect cTn in many patients with stable coronary artery disease (CAD) and may help in their risk stratification. Contrastingly, in most patients with stable CAD, cTn levels were below the limit of detection for conventional assays; when detected, however, they did suggest an adverse prognosis.18 Detection of low levels of cTn using more sensitive assays provides further information.19 Cardiac troponin T levels, elucidated using a highly sensitive assay, were above the 99th percentile in about 10% of patients with stable CAD. After adjustment for other prognostic indicators, there was a strong and graded increase in the incidence of cardiovascular death and heart failure.
Increased risk was evident even below the limit of detection of conventional cTnT assays and below the 99th percentile of the novel assay. The observation of elevated cTn levels in patients with stable CAD, however, highlights the need to scrutinise the performance of sensitive cTn assays in the diagnosis of AMI, particularly in patients with CAD and other disorders potentially associated with elevated cTn levels – such as stable chronic kidney disease and heart failure – as well as in the elderly.
Even more sensitive cTn assays are being developed.20 They will allow for detection of cTn in most healthy individuals and will further increase the accuracy of cTn evaluations at emergency department presentation, although our data did not show clinically significant differences between presently available and ‘highly sensitive’ assays. Thus, the term ‘troponin positive’ should be avoided. ‘Detectable’ levels will become the standard and will have to be distinguished from ‘elevated’ levels. As AMI is not the only cause of myocyte necrosis, these assays may improve the detection of myocardial injury in patients with other acute cardiac but non-coronary disorders, such as acute heart failure, pulmonary embolism, arrhythmias and myocarditis, and also myocardial injury in patients with chronic cardiac disorders. Detailed clinical assessments will be critical to differentiate AMI from other causes of myocardial injury, which will become more pertinent as sensitivity improves. At present, this will be a clinical judgement, as most of the tools that we use in clinical practice to help with these distinctions, such as ECG, echocardiography, cardiac magnetic resonance imaging, computerised tomography angiography and coronary angiography, have not been validated for this indication. Also, the most appropriate management for these groups remains unknown.
The differential diagnosis of small amounts of myocardial injury – and, therefore, mild elevations of cTn – is broad. Serial changes, documented by a second measurement to differentiate acute cardiac disorders (showing a rise and/or fall) from chronic cardiac disease (which usually exhibits constant cTn levels) will likely be helpful.
However, at present, is it unclear whether absolute or relative changes would better separate acute from chronic cTn elevations, or what change should be considered significant. Preliminary data suggest that an absolute change of 30% of the cTn level at the 99th percentile within 6 hours might be reasonable criteria but these criteria will likely be assay dependent.
Ongoing studies need to define the best algorithms for applying data from both contemporary and novel sensitive cTn assays in clinical practice. Rule-out and rule-in algorithms, as well as the timing of the second measurement, must be fine-tuned for each assay. Until data from large studies have validated more rapid rule-out algorithms, we recommend that clinicians stick to the guidelines and repeat the measurements after six hours.1,2
We and others are currently attempting to develop ways of facilitating the triage of patients who present to the emergency department. For example, a conjoint strategy with copeptin, a quantitative marker of endogenous stress, could improve the diagnosis of AMI.21 This strategy is undergoing evaluation in two large, multicentre trials.
Sensitive, contemporary cTn assays already deliver an excellent diagnostic performance at emergency department presentation and could significantly improve the early diagnosis of AMI, particularly in patients with recent-onset chest pain. Ongoing studies must define the best algorithms to rapidly rule in and rule out AMI.