Review
How acute changes in cardiac troponin concentrations help to handle the challenges posed by troponin elevations in non-ACS-patients

https://doi.org/10.1016/j.clinbiochem.2014.09.003Get rights and content

Highlights

  • Changes in cardiac troponin provide additional information in diagnosis of AMI.

  • Changes in troponin levels help to handle the challenges posed by troponin elevations in non-AMI.

  • Absolute changes provide higher diagnostic accuracy for AMI as compared to relative changes.

Abstract

Cardiac troponins (cTn) I and T are organ-specific, but not disease specific biomarkers. Although acute myocardial infarction (AMI) is the most important cause of cTn elevation, other cardiac disorders as well as primarily noncardiac disorders with cardiac involvement often are also associated with cardiomyocyte injury. Levels of cTn should be interpreted as quantitative markers of cardiomyocyte injury with the likelihood of AMI increasing with the level of cTn.

Similar to the level of cTn at presentation, acute changes in cTn help to differentiate chronic disorders, which show no change, from acute conditions, which usually show a rise from presentation to the second measurement at 1–3 h in the emergency department. Thereby, changes in cTn help to overcome some of the challenges posed by cTn elevations in non-AMI patients. Absolute changes in cTn provide a higher diagnostic accuracy for AMI as compared to relative changes. Again, the higher the absolute change, the higher the likelihood for AMI. Two caveats apply to the diagnostic use of cTn changes. First, patients with AMI may show no or only a minimal change when assessed around the peak of cTn release. Second, in addition to AMI, several other acute cardiac conditions including tachyarrhythmias, myocarditis, hypertensive crisis, and Takotsubo cardiomyopathy also may present with substantial cTn changes.

Introduction

Coronary artery disease (CAD) and acute myocardial infarction (AMI) are a major cause of death and morbidity worldwide [1]. In Europe and the United States, almost 15 million patients per year present to the emergency department (ED) with symptoms suggestive of AMI. Rapid identification and risk stratification of patients with AMI are of paramount clinical importance. Delayed rule-in increases morbidity and mortality; delayed rule-out prolongs time spent in the ED causing enormous costs for the health care system [2].

According to current guidelines [3], AMI can be diagnosed when there is evidence of myocardial necrosis in a clinical setting consistent with myocardial ischemia. Necrosis is defined by a significant rise and/or fall of cardiac troponin (cTn) with at least one value above the 99th percentile. The required magnitude or the preferred metrics for the rise and/or fall are not specified in the universal definition of AMI which are as follows: should clinicians use absolute or relative changes to best differentiate AMI from other conditions associated with elevations in cTn (Table 1) and what magnitude of change in cTn level is required in which time interval?

Determination of cTn is the method of choice to detect myocardial injury in clinical practice owing to their high analytical and clinical specificity. Analytical lower detection limit and precision of measurements cTn were substantially improved during the last years. Introduction of high-sensitivity cTn (hs-cTn) with high analytical sensitivity and precision of measurements (coefficient of variation < 10% at and below the 99th percentile) with consistently high analytical specificity allows detection of cTn in more than 50% of a healthy population [4].

The introduction of hs-cTn assays has improved diagnostic accuracy in the early diagnosis of AMI [5], [6], [7]. These improvements in assay sensitivity have increased the number of positive hs-sTn results (cTn values > 99th percentile) in conditions other than AMI, particularly in structural heart disease with chronic myocardial damage (e.g. chronic heart failure, stable CAD, left ventricular hypertrophy) [8], [9], [10], [11], [12], [13], [14], [15], [16] (Fig. 1). As these conditions sometimes may cause symptoms and ECG changes similar to that seen in AMI, they often are difficult to distinguish from AMI. The use of serial measurements to determine cTn changes over 1, 2, 3, or 6 h in the ED (delta cTn) is therefore helpful to differentiate between acute and stable cTn elevations [17] since the majority of patients with chronic myocardial damage do not display relevant cTn changes.

Fortunately, in 2014 the debate of whether absolute or relative changes in cTn are the preferred metrics in the early diagnosis of AMI can be based now on rather solid evidence. Two different approaches are currently applied when assessing the change criteria. The first is based on theoretical considerations regarding analytical and biological variability obtained from serial sampling in healthy volunteers [10], [18], [19]. This approach was e.g. followed in the 2007 NACB (National Academy of Clinical Biochemistry) practice guidelines [20]. The second approach is data-driven and was recently assessed in large prospective cohorts of patients with acute chest pain [21], [22], [23].

The NACB guidelines recommend a relative change of > 20% (> 3 SD of the coefficient of variation) [20], they did not consider biological variation. When we assume that the difference in troponin values at 2 time points is determined by an underlying acute cardiac event as well as analytical and biological variability (e.g. circadian changes), then a change in cTn has to exceed the variability possibly related to analytical and biological factors in order to be more likely related to the acute cardiac event and therefore clinically significant. Some groups [10], [18], [19], [24], [25] measured biological short- and long-term variations of the analyte for the determination of reference change values (RCVs) by intra-individual variability and inter-assay variability. Short-term biological variability was 14–58% in low ranges of hs-cTn as assessed in a healthy population, which led to a RCV of 38–86%. It is questionable whether these results can be extrapolated to a cohort with suspected AMI, because these comorbid patients do have much higher baseline cTn values. In addition, as we lack an alternative method applicable in clinical practice with similar sensitivity to hs-cTn to detect myocardial injury (imaging seems to have less than one-fifth the sensitivity or even less if the disease process is diffuse) [3], we do not know whether the observed change is due to analytical factors or in fact due to changes in biology. E.g. in a healthy person left ventricular enddiastolic pressure may be 2 mm Hg in the morning and perhaps 8 mm Hg after liberal fluid intake during breakfast. Of course, he/she is completely asymptomatic at both time points, and one incorrectly might postulate that biology is identical. Extrapolating the method described above to this scenario would require a change in left ventricular enddiastolic pressure of 400% to be considered clinically relevant, which of course is incorrect.

In the data-driven approach based on several large prospective diagnostic studies, absolute changes consistently provided significantly higher diagnostic accuracy as compared to relative changes to distinguish AMI from other causes of acute chest pain [22], [23], [26], [27]. The diagnostic superiority of absolute changes seemed to be at least partly explained by the following considerations: First, with the new hs-cTn assays it is possible to measure very low cTn concentrations. Even a small absolute increase or decrease in cTn in the normal range can lead to large relative changes, also in patients without AMI (e.g. a rise from 5 ng/L to 7 ng/L). Second, the recommended relatively high relative changes (113 to 266%) [28], [29], [30], [31] might often not be reached by patients with AMI who are presenting several hours after chest pain onset as their cTn levels may already be close to the plateau [32]. It is reassuring that until now, all published studies, differing in small details regarding the adjudication process from each other, have consistently found absolute changes to be superior to relative changes in the early diagnosis of AMI.

Some of these observations have already been incorporated in current recommendations of the biomarker study group of [33] the European Society of Cardiology. Clinicians should use an absolute change of ≥ 50% the 99th percentile at 3 h in patients with a baseline hs-cTn value  99th percentile and a relative change of ≥ 20% at 3 h in patients with a baseline hs-cTn > 99th percentile (Fig. 2). An alternative algorithm combining the absolute level of hs-cTnT at presentation with absolute changes within the first hour has recently been shown to provide a very high negative predictive value for the “rule-out” of AMI and a high positive predictive value for the “rule-in” of AMI (Fig. 3) [34]. A similar algorithm has been optimized for the differentiation of AMI from other cardiac conditions such as heart failure or arrhythmias (Fig. 4) [35].

The higher the baseline cTn value, the higher the positive predictive value for an AMI. In patients with very high levels of cTn (e.g. 50-times the 99th percentile), in such cases changes in cTn add little if anything to the presentation level. In contrast, changes in cTn add substantial information if the initial cTn level is in the normal range or only slightly above the 99th percentile.

One huge merit of hs-cTn delta values is the possibility for fast rule-out of AMI (using the baseline and the 1 h value) according to their high sensitivity and negative predictive values [34], [35]. Compared with the 3 to 6 hour time period for a follow-up cTn test sample as still recommended in the current guidelines, shortening to a 1-hour follow-up period would improve substantially patient management by reducing unnecessary hospitalizations and shortening observation time, therefore leading to more efficient use of ED resources. The safety for the hs-cTnT algorithm [34] seems to be excellent. Similar algorithms are currently derived and validated for hs-cTnI assays. Although the concept is the same for all assays, individual cut-off levels will be assay-specific to optimize diagnostic accuracy.

Two caveats apply to the diagnostic use of cTn changes. First, patients with AMI may show no or only a minimal change when assessed around the peak of cTn release. Second, in addition to AMI, several other acute cardiac conditions including tachyarrhythmias, myocarditis, hypertensive crisis, and Takotsubo cardiomyopathy also may present with substantial cTn changes. These caveats highlight that there is not a magic single change value that works in all setting. Rather, changes need to be considered in conjunction with all other clinical and laboratory information.

The universal definition of AMI is strongly based on the concept of using an absolute cTn value that has to be reached in order for the event to qualify for AMI: the 99th percentile [3]. This concept is universally accepted. However, it also has one key limitation, which clinical implication may be more profound than previously thought: the determination of the 99th percentile.

The methodology for the determination of 99th percentile cutoffs has come under scrutiny, e.g. selection of the reference population. The handling is very different according to the respective manufacturer and several different rules have been used for the definition of “healthy“ [36]. Depending on what one might designate the reference population, the results differ substantially as shown in the study by Collinson et al [37]. The 99th percentile cutoff will also not be the same for different subpopulations like elderly patients [38], patients with preexisting coronary artery disease (CAD) [39] and patients with chronic renal disease. In order to overcome the problem of the population-dependent underlying baseline value, one might consider as an alternative approach for the definition of AMI to use a definite delta change independently of the baseline hs-cTn value for diagnosis of AMI.

There is evidence for a better risk prediction with hs-cTn delta values: absolute changes outperformed relative changes in their accuracy to predict mortality in patients with suspected AMI [40]. The beauty of this analysis is that it is completely independent of the method used for the adjudication of AMI. However, it is important to highlight that as tools for risk prediction, changes were still inferior to hs-cTn presentation values. Apple et al [41] demonstrated an increased risk for cardiac event or death with deltas when hs-cTn at baseline was normal compared to the elevated baseline values. And Kavsak et al [27] found a significantly higher risk for death or AMI in patients in the highest tertile of absolute or relative changes, whereas in the Cox regression model the absolute changes were an even stronger predictor than relative changes. However, baseline values were not evaluated in that analysis.

Section snippets

Conclusion

Acute changes in cTn complement the quantitative information provided by the cTn level at presentation in the early diagnosis of AMI. Changes in cTn help to overcome some of the challenges posed by cTn elevations in non-AMI-patients. Absolute changes in cTn provide a higher diagnostic accuracy for AMI as compared to relative changes. The higher the absolute change, the higher the likelihood for AMI.

Declaration of interests

Professor Mueller has received research grants from the Swiss National Science Foundation (32003B_135434), the Swiss Heart Foundation, the European Union (Eurostars No. 5495), the Cardiovascular Research Foundation Basel, 8sense, Abbott, ALERE, Brahms, Critical Diagnostics, Nanosphere, Roche, Siemens, and the University Hospital Basel, as well as speaker or consulting honoraria from Abbott, ALERE, Brahms, Cardiorentis, Novartis, Roche, and Siemens. Karin Wildi and Raphael Twerenbold declare

References (41)

  • F.S. Apple et al.

    Biomarkers ITF on CA of C. Analytical characteristics of high-sensitivity cardiac troponin assays

    Clin Chem

    (2012)
  • T. Reichlin et al.

    Early diagnosis of myocardial infarction with sensitive cardiac troponin assays

    N Engl J Med

    (2009)
  • T. Keller et al.

    Sensitive troponin I assay in early diagnosis of acute myocardial infarction

    N Engl J Med

    (2009)
  • S. Agewall et al.

    Troponin elevation in coronary vs. non-coronary disease

    Eur Heart J

    (2011)
  • A. Wu et al.

    Short- and long-term biological variation in cardiac troponin I measured with a high-sensitivity assay: implications for clinical practice

    Clin Chem

    (2009)
  • J.T. Saunders et al.

    Cardiac troponin T measured by a highly sensitive assay predicts coronary heart disease, heart failure, and mortality in the Atherosclerosis Risk in Communities Study

    Circulation

    (2011)
  • J. a De Lemos et al.

    Association of troponin T detected with a highly sensitive assay and cardiac structure and mortality risk in the general population

    JAMA

    (2010)
  • C. Kawahara et al.

    Prognostic role of high-sensitivity cardiac troponin T in patients with nonischemic dilated cardiomyopathy

    Circ J

    (2011)
  • T. Omland et al.

    A sensitive cardiac troponin T assay in stable coronary artery disease

    N Engl J Med

    (2009)
  • P.A. Kavsak et al.

    High-sensitivity cardiac troponin I measurement for risk stratification in a stable high-risk population

    Clin Chem

    (2011)
  • Cited by (19)

    • Cardiac troponin T concentrations and patient-specific risk of myocardial infarction using the novel PALfx parameter

      2019, Clinical Biochemistry
      Citation Excerpt :

      Lastly, cTnT concentrations are often evaluated at several time points during the ED visit to evaluate cTnT change. In this report we only use the baseline cTnT level to calculate the risk of MI, as it is still being debated which change in the cTnT level signifies MI [3,42,43]. Indeed, over 25% of patients with non-ST elevation MI (NSTEMI) have less than a 20% change in the cTnT concentration over six hours, likely because they present late in the infarction process [18].

    • Recommendations for Institutions Transitioning to High-Sensitivity Troponin Testing: JACC Scientific Expert Panel

      2019, Journal of the American College of Cardiology
      Citation Excerpt :

      Rising or falling hs-cTn patterns result from a variety of underlying conditions, including pulmonary embolus, myocarditis, or sepsis, as well as AMI. The absolute baseline concentration, as well as the change in hs-cTn, is often one indication of whether an AMI has occurred versus another disease state (17,18). For instance, when type 1 MI has occurred, it is common to see rapid and substantial increases in hs-cTn over a few hours (19).

    • Resource utilization and outcome among patients with selective versus nonselective troponin testing

      2018, American Heart Journal
      Citation Excerpt :

      The timing and frequency of troponin testing was not standardized, consequently a significant number of patients had a single measurement and dynamic troponin patterns could not be assessed. Measurement of troponin change is potentially useful in non-ACS patients.35,36 The emergency department physician was likely aware of the initial troponin testing result which may have influenced the admission diagnosis coding.

    • Clinical Features and Treatment of Scleroderma

      2016, Kelley and Firestein's Textbook of Rheumatology: Volumes 1-2, Tenth Edition
    View all citing articles on Scopus
    View full text