Multiple biomarkers for rapid rule-out of myocardial infarction: worth the added stress?

This editorial refers to ‘Adding stress biomarkers to high-sensitivity cardiac troponin for rapid non-ST-elevation myocardial infarction rule-out protocols’ by I.Z. Restan et al., pp. 201–212. Chest discomfort is one of the most common presenting symptoms among patients attending Emergency Departments (ED), representing ∼5% of all visits.1 Accurate assessment of chest pain is critical as it may be a manifestation of a life-threatening condition such as acute myocardial infarction (MI).1 However, a majority of patients with chest discomfort will be diagnosed with something other than MI, with an accompanying low risk; thus, these cases represent a substantial contributor to ED overcrowding.1 Improving the efficiency of the chest pain evaluation while maintaining safety has become a priority for clinicians and hospital systems. Over the past decade, significant advancements have been made in the evaluation of chest pain owing to greater refinement in biomarker testing: high-sensitivity cardiac troponin (hs-cTn) assays facilitate precise measurement of troponin concentrations at or below the 99th percentile upper reference limit with resultant higher accuracy for MI evaluation.2 To allow for convenient and reproducible application, pathways utilizing serial hs-cTn concentrations were developed to rapidly and safely diagnose exclude MI.3 This paradigm progressed further when data emerged that a single hs-cTn concentration at the time of ED presentation may be sufficient to safely discharge a reasonable proportion of patients presenting a few hours after chest pain onset.4 Combining these aforementioned strategies into accelerated diagnostic protocols has resulted in shorter ED stay times and reduced non-invasive testing, without compromising safety.5 As with all diagnostic means, however, improvement in diagnostic accuracy beyond hs-cTn remains something continuously studied. Why? To begin with, patients with unstable angina pectoris have—by definition—a troponin below the 99th percentile upper reference limit; such patients may remain high-risk and go under-recognized; availability of a biomarker strategy focused either on the presence of anatomic coronary artery disease rather than ischaemic injury would be expected to deliver superior performance for such patients.6 Further, while a low hs-cTn remains able to exclude risk in a large percentage, the vexing challenge of abnormal hs-cTn in those without MI remains an issue: the European Society of Cardiology (ESC) 0/1 h or 0/2 h algorithms triage ∼25% of patients as ‘observe’ with abnormal values for hs-cTn and ambiguous clinical picture.7 The optimal management of this specific cohort remains unknown. Non-invasive anatomical or ischaemic testing is often utilized for additional risk stratification in this group, but again, a combination of biomarkers could identify some patients at sufficiently low risk to defer testing.8 An additional challenging cohort to evaluate for MI are individuals with renal disease, with many patients meeting the observe or rule-in criteria due to chronic myocardial injury9 making additional biomarkers a plausible tool to improve the diagnostic accuracy of hs-cTn in affected individuals. Lastly, using hs-cTn alone, differentiation of the most common subtypes of MI, specifically, Type 1 MI from Type 2 MI, remains impossible; combining biomarkers to differentiate these entities derived from angiographically based adjudicated datasets may be fruitful.10 Because of the challenges created by these limitations, investigators continue to explore the value of alternative biomarkers or the possibility of combining hs-cTn with additional biomarkers to further improve the efficiency of chest pain evaluation. For instance, cardiac myosin-binding protein C is released very early after cardiac myocyte injury stimulating investigation into the value of this biomarker particularly among early chest pain presenters, however, the incremental benefit over hs-cTn assays appears to be limited.11 Additionally, a repeated exercise has been to merge troponin results with those from testing for copeptin, the C-terminal fragment of pro-vasopressin; copeptin has been marketed as a marker of acute physiologic stress with the concept that coronary ischaemia may lead to its early elevation. Despite some results suggesting copeptin does not add to hs-cTn,12 a recent randomized clinical trial suggested that combining pre-hospital copeptin with a single hs-cTn measurement may reduce ED stay time when compared to the ESC 0/3 h pathway without compromising safety.13 Thus, the exploration of copeptin in chest discomfort continues. Restan et al.14 explored the utility of combining hs-cTn with stress biomarkers, copeptin, or glucose (another marker of physiological stress), for diagnosing or excluding MI in a derivation cohort of 959 patients from the WESTCOR study and subsequently in an external validation cohort of 1272 patients from the multicentre, international, APACE (Advantageous Predictors of Acute Coronary Syndrome Evaluation) study. As with most prior studies examining this question, copeptin did not improve the accuracy of MI diagnosis when added to hs-cTnT [area under the curve (AUC) of 0.91 vs. 0.91 for hs-cTnT alone, P = 0.83] and decreased accuracy when added to hs-cTnI (AUC of 0.85 vs. 0.93 for hs-cTnI alone, P < 0.001).14 Similar results were seen when glucose was added to hs-cTnT (AUC of 0.89 vs. 0.91 for hs-cTnT alone, P = 0.05) and to hs-cTnI (AUC of 0.79 vs. 0.93 for hs-cTnI alone, P < 0.001). These findings were similarly replicated in the external validation cohort. Furthermore, adding copeptin or glucose to the ESC recommended single rule-out cut-offs for the Roche hs-cTnT (<5 ng/L) and the Abbott hs-cTnI (<4 ng/L) assays did not increase sensitivity but rather decreased efficacy. Collectively, these results suggest that taking on the added stress of factoring copeptin or glucose to hs-cTnI or hs-cTnT is unlikely to further improve the diagnosis of MI or the safety of rule-out when compared to current single rule-out hs-cTn values. Are additional biomarkers likely to further improve the sensitivity of current hs-cTn based rule-out pathways for MI? The data are nearly unanimous that the answer is no: present-day hs-cTn pathways such as the ESC 0/1 h pathway and High-STEACs pathways have a very high rule-out negative predictive value and extraordinarily low 30-day major adverse cardiovascular events associated with their use.5,15 Where added biomarkers might be expected to improve hs-cTn performance relates to the various circumstances where hs-cTn testing creates stress for clinicians: for evaluation of those with unstable angina, individuals with observation zone hs-cTn results, those with renal disease, and differentiation of Type 1 vs. Type 2 MI, among others. These are the more problematic issues with hs-cTn and where efforts should be focused on clinically meaningful improvements in how hs-cTn assays are deployed. Through innovative biomarker research and development, it is an inevitability we will utilize other markers with hs-cTn, but more likely than not, the reasons for doing so would be to sharpen the specificity of troponin (not its sensitivity) for Type 1 MI and to correctly reclassify individuals whose hs-cTn results are abnormal but ambiguous. Conflict of interest: J.L.J. is a Trustee of the American College of Cardiology, has received grant support from Novartis Pharmaceuticals, HeartFlow Inc., Innolife, Applied Therapeutics and Abbott Diagnostics, consulting income from Abbott, Janssen, Novartis, Pfizer, Merck, and Roche Diagnostics, and participates in clinical endpoint committees/data safety monitoring boards for Abbott, AbbVie, Amgen, Bayer, CVRx, and Takeda. C.P.M. is supported by the NHLBI T32 postdoctoral training grant (5T32HL094301-12). The views and opinions expressed in this article are those of the authors; they do not necessarily reflect the views of the Editors.