Improvements in the Early Diagnosis of Acute Myocardial Infarction using a Cardiac Troponin T-high Sensitive Assay
In the first talk, Christian Mueller, a Professor at the University Hospital in Basel, discussed the unmet needs of current troponin assays and the improvements made in the early diagnosis of acute myocardial infarction (AMI) using the latest cardiac troponin T-high sensitive (cTnT-hs) assay.
There is a Clinical Need for High Sensitive Cardiac Troponin Tests
To put the new assay into context, Professor Mueller first explained the clinical problem in using conventional cardiac troponin (cTn) as a marker indicative of AMI. When patients report chest pain, three important variables are used to make the diagnosis: detailed patient history, 12-lead electrocardiogram (ECG) and cTn as a preferred biomarker to document myocardial cell necrosis. If the ECG shows significant ST-segment elevation, the management of the patient is clearly defined. However, for the vast majority of patients with AMI who do not have significant ST-segment elevation or depression, the cTn assay is the most important diagnostic test.
The limitation of conventional cTn assays is that it takes approximately three to four hours before a detectable increase in peripheral blood cTn levels can be detected (see Figure 1). In current clinical practice, a second cTnT test is therefore required six to eight hours after the onset of a suspected AMI and patients require constant ECG monitoring during those hours. Furthermore, the inability to make a confirmed diagnosis of AMI for up to six to eight hours after the onset of chest pain means that there is a delay in adequate therapy for those patients who are later confirmed or ‘ruled-in’ to have AMI, allowing more cardiac damage to occur and subsequently increasing the risk of morbidity. The need for monitoring and further tests also contributes to the overcrowding of emergency departments and increases overall costs. Moreover, patients with benign disease who are later excluded or ‘ruled-out’ from having AMI, suffer increased anxiety while awaiting results.
The Cardiac Troponin T-high Sensitive Assay is Superior to Conventional Tests
The conventional cTnT assay is not able to detect low concentrations of cTnT <0.01μg/l and lacks of precision until cTnT levels reach 0.035μg/l (see Figure 2). Furthermore, conventional cTn assays do not detect cTn in healthy volunteers. Thus, it has been difficult to establish the normal range of cTn.
However, by using the cTnT-hs assay the normal range of cTnT has been shown to be below 0.014μg/l and any higher result is likely to be pathological (as defined by the 99th percentile recommended cut-off value derived from a healthy population). Furthermore, when detecting cTnT concentrations above 0.2μg/l, the cTnT-hs assay provides values identical to those attained using conventional cTnT assays.
The Cardiac Troponin T-high Sensitive Assay
The cTnT-hs assay can detect cTnT levels equivalent to that of ~100 dead cardiac cells, whereas conventional tests require ~1,000 cardiac cells to have died before cTnT levels are high enough to detect. Furthemore, the cTnT-hs assay can detect cTnT in 80% of healthy subjects.
The Clinical Effectiveness of the Cardiac Troponin T-high Sensitive Assay
When considering the effectiveness of the novel cTnT-hs assay, it was hypothesised that the troponin-blind interval would be shortened and that the diagnostic accuracy at the onset of chest pain would increase, allowing for a more rapid diagnosis to either rule-in or rule-out AMI. These hypotheses have proven to be correct. In the ongoing prospective, international and observational Advantageous Predictors of Acute Coronary Syndromes Evaluation (APACE) trial, unselected consecutive patients with symptoms suggestive of AMI who presented to the emergency department were recruited (currently n=1,600). The trial did not interfere with management decisions. After all the tests were completed, two independent experts applied a gold-standard diagnostic method using all clinical information available within 60 days of follow-up (history, physical, ECG, cardiac troponin assay, chest X-ray, coronary angiography, exercise testing, computed tomography [CT] angiography and endoscopy). The experts were blinded with respect to the Tn assay value being assessed as part of the study. The preliminary results showed that early diagnosis of AMI was improved using new generation troponin assays, particularly for patients who presented within three hours of the onset of symptoms, when used in conjunction with the 12-lead ECG and a detailed patient history (see Figure 3).1
Professor Mueller concluded that the clinical application of the cTnT-hs assay will allow for a diagnosis to be made more quickly than with conventional Tn assays (two versus six hours). Future work includes defining a rule-in and rule-out algorithm for patients with or without AMI, respectively.
A New Approach Using a Cardiac Troponin T-high Sensitive Assay for the Diagnosis of Non-ST-segment Elevation Myocardial Infarction
In the second presentation, Evangelos Giannitsis, Professor at the Angiology and Pneumology Medical Clinic in Heidelberg, Germany, reviewed the need for serial sampling to determine the kinetics of cTnT-hs. He began by explaining that with the increased sensitivity of the hs-Tn assays, more patients may be detected with elevated troponin levels, but not all these patients would have AMI, some would have non-acute coronary syndrome (non-ACS), chronic or acute myocardial damage. This has been demonstrated by a study that investigated the clinical sensitivity and diagnostic accuracy of a hs-TnI assay.2 By decreasing the troponin cut-off point from >0.1μg/l to >0.006μg/l, the sensitivity was increased from 55 to 96% in the baseline sample; however, the specificity decreased from 93 to 33%.
Professor Giannitsis further explained how these results were relevant in clinical practice when patients present with acute symptoms, such as shortness of breath and chest pain and patients with suspected pulmonary embolism (PE). He presented his data (yet to be published) from the chest pain unit (CPU) registry in Heidelberg (n=3,500 between 1 June and 30 November 2009) showing that 24.4% of patients admitted to the CPU were tested positive for cTnT using a high-sensitivity assay. Importantly, 69% of these troponin-positive patients did not have acute coronary syndrome (ACS) but had other disorders. For example, patients who were later diagnosed with acute heart failure (AHF) or PE had a median circulatory troponin concentration of 44.4ng/l and 68ng/l, respectively.
One problem in the diagnosis of ACS is that the magnitude (delta) of the rise and fall of cTnT levels required by the universal definition of myocardial infaraction (MI) has not yet been clearly defined.3 To address this, Professor Giannitsis conducted a small pilot study that investigated the cTnT-hs assay in patients who were retrospectively confirmed to have unstable angina (UA) or evolving non-ST-segmen-televation myocardial infarction (NSTEMI) and initially tested negative for troponin and later tested positive for troponin.4 The patients with UA had low concentrations of cTnT at admission and relatively little change over the course of three hours (median cTnT-hs concentration at admission and at three hours were 4.25ng/l and 2.89ng/l, respectively; n=29).
In comparison, patients with NSTEMI had slightly higher concentrations of cTnT-hs at admission, which either increased or decreased at three hours (median 21.45ng/l and 55.3ng/l, respectively; n=10). Using these data, Professor Giannitsis and his team determined that a doubling of the cTnT-hs concentration within three hours of admission was the optimal delta change for predicting evolving NSTEMI.4 Diagnostic performance of serial blood sampling using the 99th percentile cut-off in evolving MI showed that at three hours, the sensitivity was 100% and the specificity was 77%. For 20% and 117% deltas, sensitivities were 85% and 69%, respectively and specificities were 57% and 100%, respectively (see Figure 4 and Table 1).4 These preliminary results were generated based on a preselected group of patients with ACS. Therefore, the delta change analysis was extended to a larger unselected population cohort of CPU patients (n=775) including ACS and non-ACS patients. The results are expected to be published soon.
Professor Giannitsis also emphasised the importance of considering the biological variation of cardiac biomarkers by presenting results from a recent study on the biological variability of cTn measured with a hsTn assay.5 The results of this study showed that serial changes in values for individual patients are particularly important for interpreting results:
- from patients with increases in peripheral troponin at low concentrations measured with very high-sensitivity assays;
- from patients presenting with chest pain (short term); and
- for evaluating drugs for cardiotoxicity (long-term).
Professor Giannitsis concluded that a combination of kinetic changes and peak cTn concentration is required to make an accurate diagnosis.
Diagnosis and Mechanistic Significance of Cardiac Troponin T-high Sensitive Assay in Non-ST-elevation Acute Coronary Syndrome – Results from the Rule Out Myocardial Infarction Using Computer Assisted Tomography Study
James L Januzzi, Associate Professor of Medicine at Harvard Medical School and Director of the Cardiac Intensive Care Unit at Massachusetts General Hospital, US, started the final talk by stating that clinical data are often available long before clinicians understand the mechanism behind the results, as was the case with the cardiac biomarker, troponin.
We know that in the absence of risk factors for heart disease or ACS, very low cTnT levels are expected and that even small concentrations of cTnT in the context of ACS predict a higher risk for complications. Despite this and even though consensus statements recommend the clinical use of the 99th percentile of a normal population as the upper reference limit for cTn assays, most clinicians still use conventional assays that are not sensitive enough to detect the 99th percentile value (13–14pg/ml with cTnT-hs). However, the newly developed cTnT-hs assay delivers such performance.
The arrival of the cTnT-hs assay has led to the following two questions. Firstly, what does increased sensitivity do to the specificity of the assay and secondly, what does an elevated cTnT-hs mean in a patient without an obvious ACS? To address these questions, Professor Januzzi presented data from the recently published Rule Out Myocardial Infarction Using Computer Assisted Tomography (ROMICAT-1) hsTnT cross-sectional study, which included 377 randomised patients (mean age 53.7 years, 64.2% male) with chest pain and low-to-intermediate likelihood for ACS.6 Patients underwent a 64-slice cardiac CT on average four hours from initial presentation and a sample of blood was taken at the time of cardiac CT for the measurement of cTnT using a high-sensitivity assay. A total of 37 patients (9.8%) with chest pain appeared to have ACS and 340 patients (90.2%) showed no signs of ACS. Furthermore, coronary CT angiography showed significant differences in structure and function between the ACS and the non-ACS group, including more prevalent cardiac structure and function abnormalities such as coronary disease. The cTnT-hs assay was effective when determining whether or not patients had ACS (see Figure 5). Patients with UA also had increased levels of peripheral cTnT even though not every patient with UA tested positive for troponin using the cTnT-hs assay. The cTnT-hs assay was superior to the less sensitive assays for the diagnostic evaluation of chest pain (see Table 2), with 30% of ACS patients reclassified from UA pectoris to NSTEMI by the cTnT-hs assay.
Mechanistic Associations Between Cardiac Structure and Function Using Cardiac Computer Tomography and the Cardiac Troponin T-high Sensitive Assay
In multivariable analyses of the ROMICAT-1 hsTnT study, independent predictors of log-transformed cTnT-hs included age, presence/extent of coronary artery disease (CAD), cardiac structure, cardiac function and amino-terminal pro-B-type natriuretic peptide (NT-proBNP) values.6 When clinical variables were considered in patients without ACS as a function of cTnT-hs above or below the 99th percentile, patients without ACS who had cTnT-hs above 0.13pg/ml tended to be older, were more likely to have more complex medical histories (including diabetes, hypertension, increased blood serum lipids, prior CAD), as well as being on medications for coronary disease. Multivariate linear regression analyses showed that in patients without ACS, significant continuous variables predictors were age, left ventricular mass and increased NT-proBNP values.
Professor Januzzi concluded his talk by emphasising that the data show that cTnT-hs assays strongly predict ACS in patients presenting with chest pain and are superior to standard assays for this indication, as it is more sensitive than the conventional cTn tests. It also allows many patients previously diagnosed with UA pectoris to be correctly reclassified to have NSTEMI. Furthermore, an increase in cTnT-hs value reflects a true signal for heart disease, even in the absence of ACS and as such, the cTnT-hs assays allow prediction of adverse outcomes across a wide spectrum of patients. Among a cohort of low-to-intermediate risk chest pain patients in the speaker’s emergency department, cTnT-hs >13pg/ml were associated with ACS, a more complex medical history, increased NT-proBNP and cystatin C, coronary heart disease and cardiac size and function. Moreover, in patients without ACS an increase in cTnT-hs was associated with increased age, increased NT-proBNP levels, prevalent coronary artery disease, increased cardiac chamber size and cardiac function.
Overall, the symposium presented compelling evidence that the cTnT-hs assay is superior to conventional assays. Professor Mueller concluded the symposium by saying: “We face a problem in the early diagnosis of AMI that is related to the lack of sensitivity in early presenters with conventional tests. This limitation seems to be overcome with the cTnT-hs assay because it is more sensitive, precise and opens a window in the low pathological range and in the normal range. These two features combined allow this test to improve the early diagnosis of AMI in patients presenting with acute chest pain to the emergency department.”