Troponin Embedded in Clinical Decision Protocols: The Devil is in the Details Free

Measurement of cardiac troponin (cTn) is a key component defining myocardial infarction (MI) and myocardial injury in the Universal Definition of Myocardial Infarction guidelines (1). Improvement in the analytical sensitivity of cTn assays led to high-sensitivity cTn (hs-cTn) I and T assays, which allow detection of small cTn changes early in the release curve. Use of hs-cTn assays also led to subsequent development of early and rapid rule-out and rule-in protocols for the emergency department (ED), important for cost containment and managing overcrowding. These diagnostic and predictive clinical decision protocols (CDPs) have evaluated the ability to measure hs-cTn on presentation and again after 1, 2, or 3 hours. Early rule-out testing is used to predict that if additional tests were to be performed a diagnosis of MI would not be confirmed, allowing expeditious discharge at low risk of adverse events. A positive result (high positive predictive value group for MI) suggests that if further testing is performed the patient will have a diagnosis of MI or non-MI myocardial injury. Many of these CDPs have demonstrated acceptable safety and efficacy (2). One such CDP is recommended by the European Society of Cardiology (ESC), with a preference for a 0/1 h serial sampling protocol to rule-in or rule-out MI in approximately 75% of patients (3). The ESC document suggests a 0/1 h protocol is better than a 0/2 h protocol, which they view as a second best option.

Lehmacher et al. present findings from a validation study assessing the utility of the Access 2 hs-cTnI assay (Beckman Coulter) in a rapid 0/1 h protocol for diagnosis of MI (4). Bio-banked serum specimens were analyzed from the Biomarkers in Acute Cardiac Care study, an ongoing observational prospective trial enrolling patients with suspected acute MI. This single-center study in Germany enrolled 1879 patients with suspected acute MI without ST-elevation (13.7% prevalence of MI). Three percent and 26.1% of patients had a final diagnosis of acute or chronic myocardial injury, respectively. In this cohort, 44.5% of patients had either a baseline hs-cTnI concentration <4 ng/L with chest pain symptom onset greater than 3 hours previously or a baseline hs-cTnI <5 ng/L and a change at 1 hour of <4 ng/L and were triaged into the rule-out group. This approach achieved a sensitivity of 97.7% and a negative predictive value of 99.3%, thus missing 2.3% (n = 6) of MI patients. Within the group of missed MI patients, 5 out of 6 were diagnosed with type 2 MI and 4 were female. Patients were ruled in if the baseline hs-cTnI was >50 ng/L or the 1 h delta was >15 ng/L; 20.3% of individuals were ruled in with an overall specificity of 88%.

The proportion of patients triaged into the observation zone differs depending upon a number of variables, including the timing protocol utilized for blood draws and the time from symptom onset. Validation of the ESC 0/2 h algorithm with the Beckman Access hs-cTnI assay, with identical cutoffs but wider delta criteria compared to the 0/1 h algorithm, yielded a substantially different percentage of patients in the rule-out arm (78%) and observation arm (16%) compared to the 0/1 h algorithm, with 44.5% and 35.1% of patients in the rule-out and observation groups, respectively (5). The observational group in the ESC 0/1 h algorithm remains challenging regardless of the hs-cTn assay used to classify patients in that group, with known heterogeneity and incidence of non-ST elevation MI up to approximately 20%. Additional serial hs-cTn testing along with echocardiography is recommended for those in the observation group, as opposed to only additional invasive cardiac testing.

The specificity and positive predictive value (PPV) of hs-cTnI to identify MI has been reported in other studies to decrease with age and renal function, which was particularly notable in the present study with age (Spec_Age ≥ 65 of 83.6% vs Spec_Age < 65 of 91.7%). There was also significant variation with sex (PPV_male 57.3% vs PPV_female 38.0%) (6). It is hypothesized that older patients may have more frequent hs-cTn increases due to acute or chronic myocardial injury or type 2 MI. Although these elevated hs-cTn results may not be as predictive or specific for MI in these populations, they do portend poor outcomes. The future risk of incident MI or mortality at 3 years in the present study was higher for patients classified as observe and rule-in compared to rule-out (HR_rule-in: 7.11, HR_observe: 5.94) using the 0/1 h algorithm, which was also consistent with observations using the Access assay in the ESC 0/2 h algorithm.

How realistic is a 0/1 h strategy? For real-world performance of the 0/1 h algorithm to mimic the clinical sensitivity and specificity reported by Lehmacher et al., specimens must be collected within a strict window of time, which often presents a challenge in busy EDs. A CDP involving sample collection at presentation and 1 h post-presentation requires a clearly defined decision-making pathway and dedicated staff able to deliver appropriate care. Requirements for the windows of specimen collection were not defined in the present study but likely explain why the 0/2 h approach yielded a significantly greater number of patients in the rule-out group compared to the 0/1 h algorithm with the same hs-cTnI assay. Thus, although similar criteria exist for either the 0/1 or 0/2 h strategy, for many hospitals a 0/2 h approach may be more logistically feasible despite a minimal longer length of stay in the ED. However, there are no studies that have demonstrated true financial benefit of a 1 h vs 2 h strategy.

Clinical performance of a CDP is also highly dependent upon several other factors including geographical location, prevalence or rate of MI, access to care, and accurate assessment of symptom onset. The proportion of patients with a baseline hs-cTn below the limit of quantitation (the lowest concentration with a CV <20%) or limit of detection varies greatly with demographics and physical location of the ED. The negative predictive value will also depend highly on the MI rate in the region (for both type 1 and type 2 MI) and will be misleading if applied to high-risk patients. Observational studies with CDPs, although important for potential assessment and impact on patient care, are also not congruent with changes in patient care or outcomes. Without having the Access hs-cTnI results available to clinicians at the time of care, it is difficult to know the true efficacy or safety of using the 0/1 h protocol with complete certainty.

One nuance that has historically limited integration and validation of various CDPs in the United States are regulatory restrictions by the Food and Drug Administration that limit reporting hs-cTn to the limit of quantitation, as opposed to the limit of detection, which is widely used internationally as a cutoff using the same hs-cTn assays. Fortunately, the study by Lehmacher et al. utilized hs-cTnI cutoffs of 4 and 5 ng/L for the rule-out cohort, concentrations above the published limit of quantitation of 2 ng/L, which allows the Access hs-cTnI assay and ESC 0/1 h algorithm to be globally utilized. Although there are different analytical characteristics and 99th percentiles between serum and lithium heparin plasma for the Access hs-cTnI assays, both within and outside the United States, these differences are irrelevant when utilizing the assay holistically within the ESC algorithms (7).

In low-risk patients, CDPs provide valuable opportunity to implement protocols to rapidly exclude MI and safely discharge patients. There are caveats for these CDPs, particularly for hs-cTnI assays, which each have their own unique cutoffs and delta criteria due to the lack of hs-cTnI standardization. Which CDP is best is highly dependent upon the patient population and resource allocation within an individual hospital or ED and needs to be determined after careful review of the evidence. Implementation of a 0/2 h protocol is more realistic for many hospitals that do not have dedicated specimen collection staff in the ED. This issue, coupled with significant time constraints and staffing challenges, makes execution of true 0/1 h CDPs nearly impossible to achieve (8).

Nonstandard Abbreviations

cTn, cardiac troponin; MI, myocardial infarction; hs-cTn, high-sensitivity cTn; ED, emergency department; CDP, clinical decision protocol; ESC, European Society of Cardiology; PPV, positive predictive value.

Author Contributions

The corresponding author takes full responsibility that all authors on this publication have met the following required criteria of eligibility for authorship: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; (c) final approval of the published article; and (d) agreement to be accountable for all aspects of the article thus ensuring that questions related to the accuracy or integrity of any part of the article are appropriately investigated and resolved. Nobody who qualifies for authorship has been omitted from the list.

Authors’ Disclosures or Potential Conflicts of Interest

Upon manuscript submission, all authors completed the author disclosure form. Disclosures and/or potential conflicts of interest.

Employment or Leadership

None declared.

Consultant or Advisory Role

A.K. Saenger, Radiometer, LumiraDX, Quidel.

Stock Ownership

None declared.

Honoraria

None declared.

Research Funding

None declared.

Expert Testimony

None declared.

Patents

None declared.

References