Biomarker strategies: the diagnostic and management process of patients with suspected AMI

Introduction: the clinical challenge of acute myocardial infarction (AMI) diagnosis

Acute chest pain is known as the “cardinal symptom” of AMI but the majority of patients presenting to the emergency department (ED) with chest pain are discharged from the ED without this diagnosis [1]. Many patients with suspected acute coronary syndrome (ACS) present with atypical symptoms, like atypical chest pain, or arrive at the ED free of pain. Still, the term “chest pain” triggers a sophisticated and time consuming diagnostic work-up in most of these patients. The current (2015) European Society of Cardiology (ESC) Guidelines for the management of ACS in patients presenting without persistent ST-segment elevation recommend a full work-up including electrocardiogram (ECG) and serial troponin-testing also in patients with a clinically low likelihood for ACS in order to not miss patients with AMI [2]. The security standards of our current times largely exclude a discharge of patients with chest pain from the EDs based on a clinical diagnosis. Given the increasing use and consequently crowding of EDs worldwide, there is a strong clinical and public health need to achieve a faster but safe rule-in and rule-out of myocardial infarction to direct patients onto the correct management pathway. Copeptin, a new, non-cardio-specific marker, reflects the body’s vasopressin-level, and increases immediately after endogenous and hemodynamic stress. It has been shown to guide a safe early rule-out and discharge of AMI in troponin negative patients. On the other hand, high sensitivity assays for cardiac troponin may be sufficient to rule-out AMI without serial measurements or after shorter time-intervals. A number of approaches for a faster rule-in and rule-out of AMI are currently under research and evaluation and some have already been integrated into current guidelines and/or implemented into the clinical routine in selected centers. Figure 1 summarizes these different approaches, all of which use one or more biomarkers to guide clinical decision-making.

Figure 1:

Current approaches for an early rule-in and rule-out of myocardial infarction, which are being discussed, researched, tested and practiced.

It is important to note that the time from the onset of symptoms until the first blood draw varies widely and is extremely difficult to assess, as patients are influenced by stress, anxiety and possibly waiting times. hsTn, high sensitive troponin; ACS, acute coronary syndrome.

In this article, we will summarize and discuss the different diagnostic strategies for patients with suspected AMI including the use of copeptin.

(Very) low levels of troponin at admission

With the introduction of new highly sensitive troponin assays, it became possible to diagnose AMI at an earlier stage and with higher accuracy and at the same time to refine risk stratification to patients with very low detectable troponin levels. This was first shown by Keller et al. who tested over 1800 patients with suspected AMI at different time points after presentation to the ED [3]. A number of studies have since shown that patients with low levels of troponin measured with high sensitivity assays are at low-risk for adverse events in the short-, medium-, and long-term future [4, 5]. This applies to very low troponin-levels (below the level of detection, LoD), as well as to levels below the 99th percentile cut-off. Bandstein et al. tested this in a population of almost 15,000 patients above the age of 25 years who presented themselves to an ED with acute chest pain and had at least one high sensitive troponin (hsTn) tested during their stay in the ED. Patients were divided into three groups according to an hsTn level (a) above the 99th percentile, (b) below the 99th percentile but above the level of detection and (c) below the level of detection. In this population, hsTn below the 99th percentile had a negative predictive value (NPV) for AMI within 30 days of 96.9 (96.9–97.5) % and for death within 30 days of 99.6 (99.4–99.8) %. In patients with hsTn below the LoD, the NPV was 99.8 (99.7–99.9) % for AMI and 100 (99.9–100) % for death within 30 days, respectively [6]. Vafaie et al. were able to confirm these findings in a sub-study of the BIC-8-trial [7]. In their population of 882 patients with low- to intermediate risk for ACS, no death or AMI occurred in patients within 30 days after admission, if they had hsTn values below the LoD at admission [8].

Shah et al. evaluated the negative predictive value (NPV) of hsTnI values in a prospective cohort of over 6000 ED patients with suspected ACS, consisting of three different cohorts (derivation cohort n=4870, internal validation cohort n=1126, external validation cohort n=308). In patients with troponin values below the 99th percentile of derivation cohort, a troponin concentration <5 ng/L provided a negative predictive value of 99.6% [95% confidence interval (CI) 99.3–99.8]. Notably, the NPV was lower in patients who were tested for troponin within 2 h after the onset of symptoms (97.6%, 95% CI 95.8–99.2) than in patients who had had symptoms for more than 2 h (99.8%, 95% CI 99.6–100.0) [9].

Despite these encouraging results it has to be noted that there are no interventional randomized controlled trials to confirm these findings – all studies which have been published so far were diagnostic studies where the biomarker results did not impact patient management. These studies are suited to evaluate the diagnostic performance of novel diagnostic approaches, as they use central adjudication of the final diagnosis as a reference standard, but lack the ability to anticipate the logistics and effects of their application in the clinical setting. The patient populations of these trials differ widely. In the BIC-8 cohort [8], a low-risk cohort with a mean age of 54.1 (±15.6) years, a mean GRACE score of 80.3 (±27.6) and a median TIMI risk score of 1 (0–3), 34.6% of the patients had hsTnT values below the LoD. Bandstein et al. [6] enrolled a very-low-risk population attending with chest pain and at least 1 hs-cTnT measurement and 62.8% of their patients had hsTnT below the LoD. In the higher-risk population of Shah et al., 88% of patients in the derivation cohort had values above the LoD, and 26% above the 99th percentile. In this cohort, 16% and 4% had an index AMI type I and II, respectively, 6% were classified as having myocardial injury; the validity of this paper is hampered by conflicting numbers given in the tables and figures [9].

In summary, there is good evidence that hsTn levels below the LoD might help to identify healthy people, but test populations of current populations are inconclusive and prospective, controlled, randomized studies are lacking. Still, this concept is a potential future alternative in a subgroup of low-risk patients to the validated and established copeptin/troponin protocol [7].

The troponin blind and gray zones

In an ideal world, biomarker results give the treating physician a clear guidance on the presence or absence of the disease in question. Even though the minority of biomarkers provide this ideal, a basic requirement is the easy and intuitive handling of biomarker results in clinical practice, especially for inexperienced doctors.

With troponin, we are currently moving towards a complex concept of different cut-off values and strategies. The idea of introducing separate cut-off values for rule-in and rule-out of myocardial infarction may be sensible from an academic point of view, but is difficult to implement in a clinical setting. Patient management is further complicated if there is a range of biomarker values which do not allow for a clear interpretation, the so-called “gray zones” or “observe zones”. Algorithms as proposed in current guidelines (2), where rule-in and rule-out of patients is based on a cut-off and/or a delta value between test results at 0 h and 1 h, whilst a large and variable proportion of patients is left in an observational gray zone may be difficult to implement in an actual clinical setting. Numbers of patients assigned to this observational zone are relatively high. Nestelberger et al. were able to show that around 25% of patients in their observational cohort of patients with suspected AMI (n=2213) were assigned to be observed by the ESC 2015 hsTn 0 h/1 h algorithm [10]. In a similar study by Mueller et al. (n=1282) 22% of patients were assigned to the observation zone [11]. A study by Neumann et al. reported a 8.2% mortality in patients in the observation zone, as compared to a 1.0% mortality in the rule-out and 6.7% in the rule-in group [12].

High levels of troponin at admission

There is plenty of evidence that troponin levels are proportionally related to risk. It has recently been suggested to use this relation and make use of the quantitative information of absolute troponin levels. Following this concept, in the same way that very low levels of troponin have a high NPV, very high levels may have a high positive predictive value (PPV) and may be able to guide an early rule-in (diagnosis) of AMI [13]. It has to be noted, that these high levels do not require hs troponin assays, as they can reliably be measured with conventional assays.

A rule-in strategy using one troponin value at admission as well as a strategy using different troponin levels as a continuum of risk assessment needs to be tested in an interventional design before it can be introduced into clinical practice and, again, it needs to be evaluated whether a strategy with different cut-off values is practicable for doctors treating emergency patients. A recent expert paper summarizes the current evidence [14].

Copeptin

An alternative for different cut-off values of one marker might be the combination of different tests in a dual-marker approach. The most promising strategy in this respect is combined testing of troponin and copeptin at admission to the ED. Copeptin is an unspecific marker of hemodynamic stress, which is elevated in a number of conditions, including sepsis, shock, acute exacerbation of COPD and ischemic stroke. As it is not specific to myocardial ischemia and subsequent necrosis, it is not elevated in patients with true unstable angina pectoris. The BIC-8 study was able to show a comparable safety of an early rule-out process with combined troponin/copeptin testing at admission as compared to serial troponin measurements [7]. Slagman et al. were recently able to confirm results from a number of observational trials [15, 16] in an analysis of serial samples taken from patients with AMI, illustrating the complementary rise and fall of the two biomarkers with a very early rise of copeptin after the onset of symptoms, thus compensating the slow increase of troponin [17].

Copeptin has been included into the 2015 ESC Guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation, with a I-A recommendation for the combination with troponin and a IIb-B recommendation for the combination with hsTn [2]. These differing recommendations are based on reports that copeptin may not add diagnostic information in combination with a high sensitivity troponin assay. BIC-8, as the only interventional study, let each study site use their in-house troponin assay and did not differentiate between high sensitive and conventional assays. In a BIC-8 substudy, copeptin did not significantly improve the diagnostic accuracy of myocardial infarction as compared to hsTn alone but significantly improved the negative predictive value and sensitivity [8].

Figure 2 shows how the guidelines valued copeptin in a modification of the standard diagnostic picture.

Figure 2:

Integration of the recommendations of the ESC [2] with respect to copeptin into the standard diagnostic algorithm of patients with chest pain and no ST-segment elevations in the ECG.

The guidelines an early rule-out of AMI as shown in this algorithm. For further details on copeptin in the management of patients with suspected ACS please refer to Figure 3. *Copeptin can be used with contemporary sensitive troponin tests (recommendation I-A) or hs Tn assays (recommendation Iib-B) **very high Tn values have been consented resecently as >5xURL [14] ***the delta change depend on the used hsTn assay (ESC guidelines) [2]. hsTn, high sensitive troponin; ULN, upper limit of normal.

The clinical case box below shows a typical patient in the ED, with a low to intermediate risk of ACS, where AMI could be excluded fast without putting the patient at risk.

Perspectives

The specialist discipline of Cardiology has the great luxury of having a marker available which provides 100% organ-specificity. On top of its proportional relation to risk and the possibilities of early rule-in and rule-out of AMI in patients with suspected ACS, it also offers the possibilities of individualized patient management. It has recently been shown, that troponin might be useful in the decision of invasive vs. non-invasive therapy as well as medication strategies in ACS patients [18]. With copeptin, we are provided with a marker which increases almost instantly after the onset of AMI and which might therefore balance troponin gray zone levels [19]. Figure 3 summarizes the current strategies for the initial diagnostic approach in patients with suspected myocardial infarction. Any strategies which require a second blood sample lead to decision times beyond 90 min.

Figure 3:

Integrated management algorithm for the early disposition of suspected ACS in the ED.

ED, emergency department; ACS, acute coronary syndrome; STEMI, ST-elevation myocardial infarction; ECG, electrocardiogram; NST-ACS, non-ST-elevation acute coronary syndrome; PCI, percutaneous coronary intervention; CPU, chest pain unit; URL, upper reference limit; hsTn, high sensitive troponin.

Conclusions

The clinical challenge of early AMI diagnosis is mainly related to the uncertainties of the initial presentation. Very low troponin levels may help in the future but important data are lacking. With the currently clinically available strategies using troponin, a patient with suspected ACS requires a monitor place and further chest pain unit (CPU) work-up as soon as a second troponin is scheduled, causing high costs for the health care system and putting many patients at an unnecessary risk. Up to 50% of patients will have troponin test results in the gray zone requiring ongoing observation.

Copeptin in combination with hsTn allows immediate disposition of the patient prior to any accelerated troponin based protocol, which follows if clinical suspicion is high or if copeptin is positive.