Biochemical markers and the era of troponin
Robert H. Christenson, Ph.D., DABCC, FACB

Dr. Christenson is professor of pathology and professor of medical and research technology, University of Maryland School of Medicine, and director, rapid response and clinical chemistry, University of Maryland Medical Center, Baltimore, Maryland.

The acute coronary syndromes (ACSs) represent a disease continuum ranging from unstable angina, associated with reversible myocardial cell injury, to frank ST segment elevation myocardial infarction (MI) with large areas of necrosis. The common pathophysiological feature of the ACSs is instability and disruption of atherosclerotic plaques in coronary artery(s). Slide 1 shows the ACSs spectrum divided, based on the electrocardiogram (ECG), into no ST elevation and ST elevation MI, and the diagnoses of unstable angina, no ST elevation MI (NSTEMI), and Q wave and non–Q wave MI. Although clinical presentation and the ECG are critically important tools for diagnosis and management of the ACS patient, the biochemical markers cardiac troponin I (cTnI) and cardiac troponin T (cTnT) are considered fundamental for the diagnosis, monitoring, risk stratification, and therapeutic management of the ACSs patient.

Troponin T and I assays became available in the early to mid-1990s. The basis for these assays are antibodies against the proteins troponin (Tn) I and T, which represent two of the three-member [troponin C-troponin T-troponin I] protein complex (Slide 2) that is essential for the contraction of striated muscle. Although troponin T and I are found in all striated muscle, these proteins have cardiac-specific isoforms that differ in amino acid sequence from those in skeletal muscle, allowing development of immunoassays that specifically measure only the cardiac isoforms. A distinct benefit of troponin over traditional biochemical markers, including creatine kinase (CK)-total, CK-MB and myoglobin measurements, is that use of this marker allows risk stratification of acute coronary syndromes patients.1,2 Slides 4 and 5 represent survival curves from large studies showing that patients who are troponin positive are at significantly greater risk of adverse outcome including MI and death after the index event. Systematic reviews have confirmed these results, and indicate that there is no significant difference between the performance of cTnT and cTnI provided that serial sampling is performed out to 12 to 16 hours after presentation.3,4 The "troponin era" has been based on evidence that cTnI and cTnT are tissue specific and extremely sensitive biochemical markers for indicating myocardial necrosis, and therefore identification of high-risk patients.

Data indicating that troponin-positive patients are at increased risk have been so consistent and significant that the use of troponin as a potential aid for guiding therapeutic intervention was formally proposed.5 Since this proposal, a number of reports have examined the potential role of troponin for managing patients to reduce or "abort" mechanisms causing this increased risk. Fortuitously, troponin assay development has been largely coincident with advances in the knowledge of the thrombotic process and platelet function. It is now widely recognized that the composition of the thrombus is different for patients with unstable angina/no ST elevation, being predominately platelet rich, versus those with ST elevation MI, whose thrombus is mainly fibrin rich (Slide 6). This difference in composition has led, in part, to the development of novel treatment strategies for unstable angina/no ST elevation MI patients targeted at mechanisms of platelet function shown in Slide 7. Several large randomized trials have been conducted to test the efficacy of these treatment strategies.

In the FRISC study, a randomized study of low molecular-weight heparin (LMWH), peak cTnT measurements were examined, finding that patients who were in placebo arm were at significantly greater risk for adverse events including death and MI.6 Thus troponin appeared to identify patients who benefited from LMWH. In a separate study,7 the CAPTURE investigators examined troponin values in patients randomized to receive placebo or the c7E3 Fab inhibitor of GP IIb/IIIa function (abciximab). Importantly, this study showed that the risk of MI or death by 6 months was confined to patients with elevated cTnT; in these cTnT positive patients, events were significantly reduced in patients receiving abciximab versus placebo (Slide 8). This issue was studied further in PRISM,8 which examined a different GP IIb/IIIa inhibitor termed tirofiban. Both cTnT and cTnI measurements were performed in PRISM, which also showed significantly better outcomes in the treatment arm (Slide 9). A third GP IIb/IIIa inhibitor, lamifiban, was examined in PARAGON B. Although the benefit demonstrated for lamifiban versus placebo did not achieve statistical significance in PARAGON B, cTnT positive patients showed significant benefit consistent with other studies (Slide 10). Overall, the troponins identified patients who are most likely to benefit from GP IIb/IIIa therapy and appear to be a more appropriate risk stratification tool for this purpose than angiography.9

Note that the results of GUSTO IV must be considered in the context of these other trials relating troponin, administration of GP IIb/IIIa inhibitors, and outcome. GUSTO IV randomized abciximab therapy in comparison with placebo in relatively low-risk patients with chest pain >5 minutes, having either positive troponin or ECG changes (Slide 11). It was a surprise to many that this study showed no benefit of abciximab versus placebo in this patient cohort. A contrast between GUSTO IV and previous GP IIb/IIIa studies was that the earlier trials included higher risk patients having both ECG deviation and troponin positive results. This difference offers the possible explanation that the benefit of GP IIb/IIIa inhibitors in cTnT-positive patients in the absence of ECG changes may not be as substantial as in patients with both risk indicators. Thus the GUSTO IV findings may relate to the relatively low-risk patient cohort enrolled, or an inherent issue with abciximab administration.9 Further analysis of GUSTO IV data is necessary; however, the substantial evidence supporting the role of troponin positively in the medical management of high-risk patients remains valid.9

Troponin measurements in acute coronary syndrome patients are also important for the diagnosis of MI. Over the past 20 years the diagnosis of MI has been based on WHO criteria, which included at least two of the following three features: 1) symptoms of ischemia, 2) characteristic ECG changes, and 3) a typical rise and fall in cardiac markers. There have been issues with these diagnostic criteria because a large proportion, perhaps one third, of patients who are diagnosed as having MI do not present with the typical symptoms of chest pain. Also, diagnostic ECG findings are relatively insensitive, occurring in only about 45% of patients who are eventually diagnosed with MI. This uncertainty was largely responsible for the designation of biochemical markers, traditionally CK-MB, as the "gold standard" for the diagnosis of MI.10 Over the past decade, cTnT and cTnI have also demonstrated excellent performance characteristics for diagnosis of MI by WHO criteria.11 However, as is frequently the case when a new diagnostic tool is more sensitive than the benchmark, with initial troponin implementation the marker was believed by many to yield many "false positive" results, that is, troponin results were positive when the patient ruled out for MI using traditional markers as the "gold standard." Because the troponins are a more sensitive indicator of myocardial necrosis than CK-MB, a substantial proportion of patients who ruled out for MI by WHO criteria had elevated troponin levels. Later evidence revealed that patients with positive troponin had indeed suffered a cardiac event.12

Troponin measurement in the setting of myocardial ischemia has become the preferred biomarker for indicating myocardial damage because it has nearly absolute tissue specificity and high sensitivity, thereby reflecting even microscopic myocardial necrosis.12 The data are so compelling that the recent consensus redefinition of MI by the joint European Society of Cardiology/American College of Cardiology (ESC/ACC) committee, Slide 12, has designated troponin measurements as the central indicator of myocardial necrosis, that is, MI (Slide 13). It is the view of the ESC/ACC committee that evidence of any amount of myocardial necrosis caused by ischemia should be labeled as MI.12 As part of this definition, the ESC/ACC committee has recommended this 99th percentile of a reference control population, that is, normal healthy individuals, as the cutoff for indicating necrosis (Slide 14). Traditionally, MI cutoffs have been derived from receiver operator characteristic (ROC) curve studies of WHO-defined MI versus non-MI cases; ROC derived cutoffs are much higher than will be indicated by this 99th percentile. Redefinition will present many challenges for laboratory methods, clinical work-ups, and research in the area of acute coronary syndromes.

From a laboratory perspective, an important technical impact of the ESC/ACC's redefinition of MI will be a 4-fold to 30-fold decrease in the troponin cutoff indicating necrosis, depending on the method (examples shown in Slide 15). Further, the technical goal of troponin assays at the 99th percentile cutoff has been set at a CV of 10% for total imprecision. Of the numerous commercial troponin assays available, there are perhaps 2 of about 15 that are close to meeting this imprecision goal.

An important clinical issue: Will multiple cutoffs be necessitated by the new ESC/ACC redefinition? Clearly the 99th percentile cutoff will mandate a low troponin value for MI diagnosis; however, a second higher cutoff may be needed to guide administration of GP IIb/IIIa inhibitors. As illustrated in Slide 16, if the 99th percentile indicating MI diagnosis is 0.02 ng/mL, and the evidence is that there is no benefit in treating patients with GP IIb/IIIa inhibitors unless the troponin level is 0.1 ng/mL, then two troponin cutoffs appear to be appropriate.

From epidemiological and research perspectives, the ESC/ACC committee recognized that the MI redefinition will result in a substantial step function increase in MI prevalence. This is because approximately one third of unstable angina patients are troponin positive and will be classified as MI (Slide 17). The new MI definition recognizes that the release pattern of troponin is frequently delayed after myocardial necrosis, and that high diagnostic sensitivity is not achieved until relatively late after the acute event.12 This delay emphasizes that serial sampling out to 16 hours after the acute event is necessary for diagnosis of myocardial necrosis with high sensitivity and specificity (Slide 18).

Currently troponin assays are not the perfect clinical tool. The problem, in part, has been the lack of assay standardization. For cTnT measurement, there have been three generations of quantitative assays, and two generations of whole-blood qualitative tests. Assay harmonization has been achieved between these several cTnT assay-formats, because all cTnT assays have been produced by a single manufacturer (Roche Diagnostics). As a result, standardization of cTnT measurements is not currently an issue within the laboratory community. Standardization for cTnI assays, on the other hand, presents a far more complicated situation. There are numerous cTnI assays that have been developed and marketed by a variety of manufacturers. This has led to a situation where cTnI measurements using different methods on identical specimens may differ by 100-fold,13 creating a substantive problem for the clinical and laboratory community, particularly as the utilization of cTnI measurements increases. There is currently an effort spearheaded by the American Association for Clinical Chemistry to standardize cTnI assays.14

The clinical importance of troponin measurements has changed the testing requirements for biochemical marker testing. Along with increased sensitivity at low concentrations, there has been a movement toward rapid bedside testing, also termed point-of-care (POC). Many of these POC assays are qualitative (positive/negative), rather than quantitative (numerical) in result. The accuracy of these tests, compared with their laboratory counterparts, is critical. To assess the testing characteristics of a POC troponin assay, a pilot study was performed which intended to 1. assess the analytical equivalence of bedside versus laboratory measurements; and 2. compare the "vein to brain" time for bedside versus lab testing.

Slide 19 summarizes the design of the study. Briefly, patients' specimens were simultaneously sent to the central laboratory for troponin measurement and performed immediately on the bedside troponin device from Spectral Diagnostics (Toronto, Ontario, Canada). The Spectral troponin test yields either a positive or negative result; lab results were categorized in a binary manner, as either above (positive) or below (negative) the institution's cutoff. Slide 20 shows an overview of the pilot study's results for specimens from 939 patients. Over 93% of the specimens provided the same results for the central lab and POC troponin testing (either both positive or both negative). Although about 4% of the lab tests were classified as falsely negative and about 3% were falsely positive compared with the central lab results, there was no significant difference between the results from central laboratory testing and those for the POC troponin assay.

Although the troponin tests yielded equivalent results analytically, the comparative vein-to-brain times for the POC and central lab assays were very different (Slide 21). As shown, the POC troponin test had a mean time of 15 minutes, whereas the central lab time from sample collection to clinician awareness was 128 minutes. Slide 22 shows the very wide distribution of reporting times from the central lab. In summary, this pilot study showed an 8-fold faster availability of bedside results versus laboratory testing. There was no significant difference between the troponin results for laboratory and bedside testing. From this study, we believe that clinicians prefer rapid turnaround of results and that bedside troponin testing is a potentially valuable complement to ED triage because it could expedite decision-making in patients with chest pain.

The National Heart Attack Alert Program has previously defined the 4Ds (door, data, decision, drug), and determined that the goal for providing the 4Ds should be 30 minutes (Slide 23). Troponin testing has now been established as an essential element of the data, and therefore must also be provided to clinicians in an accurate and timely fashion to optimize patient care.

References

1. Ohman EM, Armstrong PW, Christenson RH, et al. Cardiac troponin T levels for risk stratification in acute myocardial ischemia. GUSTO IIA Investigators. N Engl J Med 1996;335:1333-1341.
2. Antman EM, Tanasijevic MJ, Thompson B, et al. Cardiac-specific troponin I levels to predict the risk of mortality in patients with acute coronary syndromes. N Engl J Med 1996;335:1342-1349.
3. Olatidoye AG, Wu AH, Feng YJ, Waters D. Prognostic role of troponin T versus troponin I in unstable angina pectoris for cardiac events with meta-analysis comparing published studies. Am J Cardiol 1998;81:1405-1410.
4. Ottani F, Galvani M, Nicolini FA, et al. Elevated cardiac troponin levels predict the risk of adverse outcome in patients with acute coronary syndromes. Am Heart J 2000;140:917-927.
5. Roberts R, Fromm RE. Management of acute coronary syndromes based on risk stratification by biochemical markers: an idea whose time has come. Circulation 1998;98:1831-1833.
6. Lindahl B, Venge P, Wallentin L. Troponin T identifies patients with unstable coronary artery disease who benefit from long-term antithrombotic protection. Fragmin in Unstable Coronary Artery Disease (FRISC) Study Group. J Am Coll Cardiol 1997;29:43-48.
7. Hamm CW, Heeschen C, Goldmann B, et al. Benefit of abciximab in patients with refractory unstable angina in relation to serum troponin T levels. C7E3 Fab Antiplatelet Therapy in Unstable Angina (CAPTURE) Study Investigators. N Engl J Med 1999;340:1623-1629.
8. Heeschen C, Hamm CW, Goldmann B, et al. Troponin concentrations for stratification of patients with acute coronary syndromes in relation to therapeutic efficacy of tirofiban. PRISM Study Investigators. Platelet Receptor Inhibition in Ischemic Syndrome Management. Lancet 1999;354:1757-1762.
9. Bhatt DL, Topol EJ. Current role of platelet glycoprotein IIb/ IIIa inhibitors in acute coronary syndromes. JAMA 2000;284:1549-1558.
10. Christenson RH, Azzazy HM. Biochemical markers of the acute coronary syndromes. Clin Chem 1998;44:1855-1864.
11. Keffer JH. Myocardial markers of injury. Evolution and insights. Am J Clin Pathol 1996;105:305-320.
12. 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;36;959-969.
13. Apple FS, Maturen AJ, Mullins RE, et al. Multicenter clinical and analytical evaluation of the AxSYM troponin-I immunoassay to assist in the diagnosis of myocardial infarction. Clin Chem 1999;45:206-212.
14. Christenson RH, Duh SH, Apple FS, et al. Standardization of cardiac troponin I assays: round robin of ten candidate reference materials. Clin Chem 2001;47:431-437.