Advertisement

One-hour Rule-in and Rule-out of Acute Myocardial Infarction Using High-sensitivity Cardiac Troponin I

      Abstract

      Objective

      We aimed to prospectively derive and validate a novel 1h-algorithm using high-sensitivity cardiac troponin I (hs-cTnI) for early rule-out and rule-in of acute myocardial infarction.

      Methods

      We performed a prospective multicenter diagnostic study enrolling 1811 patients with suspected acute myocardial infarction. The final diagnosis was centrally adjudicated by 2 independent cardiologists using all available information, including coronary angiography, echocardiography, follow-up data, and serial measurements of hs-cTnT (but not hs-cTnI). The hs-cTnI 1h-algorithm, incorporating measurements performed at baseline and absolute changes within 1 hour, was derived in a randomly selected sample of 906 patients (derivation cohort), and then validated in the remaining 905 patients (validation cohort).

      Results

      Acute myocardial infarction was the final diagnosis in 18% of patients. After applying the hs-cTnI 1h-algorithm developed in the derivation cohort to the validation cohort, 50.5% of patients could be classified as “rule-out,” 19% as “rule-in,” 30.5% as “observe.” In the validation cohort, the negative predictive value for acute myocardial infarction in the “rule-out” zone was 99.6% (95% confidence interval, 98.4%-100%), and the positive predictive value for acute myocardial infarction in the “rule-in” zone was 73.9% (95% confidence interval, 66.7%-80.2%). Negative predictive value of the 1h-algorithm was higher compared with the classical dichotomous interpretation of hs-cTnI and to the standard of care combining hs-cTnI with the electrocardiogram (both P < .001). Positive predictive value also was higher compared with the standard of care (P < .001).

      Conclusion

      Using a simple algorithm incorporating baseline hs-cTnI values and the absolute change within the first hour allows safe rule-out as well as accurate rule-in of acute myocardial infarction in 70% of patients presenting with suspected acute myocardial infarction.

      Keywords

      Clinical Significance
      • There were 1811 patients with suspected acute myocardial infarction enrolled in order to derive and validate a novel 1-hour algorithm using high-sensitivity cardiac troponin I.
      • More than 50% of patients could be ruled out safely, achieving a negative predictive value even higher compared with the current standard of care (hs-cTn and electrocardiogram).
      • Twenty percent of patients could be assigned to the acute myocardial infarction group achieving a positive predictive value higher as compared with the classical dichotomous interpretation of hs-cTn.
      Acute myocardial infarction is a major cause of death and disability worldwide. Patients with symptoms suggestive of acute myocardial infarction account for about 10% of all emergency department consultations. Only 15%-20% of them are diagnosed eventually as acute myocardial infarction.
      • Nawar E.W.
      • Niska R.W.
      • Xu J.
      National Hospital Ambulatory Medical Care Survey: 2005 emergency department summary.
      • Anderson J.L.
      • Adams C.D.
      • Antman E.M.
      • et al.
      2012 ACCF/AHA focused update incorporated into the ACCF/AHA 2007 guidelines for the management of patients with unstable angina/non-ST-elevation myocardial infarction: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines.
      • Hamm C.W.
      • Bassand J.P.
      • Agewall S.
      • et al.
      ESC Guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: The Task Force for the management of acute coronary syndromes (ACS) in patients presenting without persistent ST-segment elevation of the European Society of Cardiology (ESC).
      • Kontos M.C.
      • Diercks D.B.
      • Kirk J.D.
      Emergency department and office-based evaluation of patients with chest pain.
      Rapid identification of acute myocardial infarction is critical for the initiation of effective evidence-based treatment.
      • Anderson J.L.
      • Adams C.D.
      • Antman E.M.
      • et al.
      2012 ACCF/AHA focused update incorporated into the ACCF/AHA 2007 guidelines for the management of patients with unstable angina/non-ST-elevation myocardial infarction: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines.
      • Hamm C.W.
      • Bassand J.P.
      • Agewall S.
      • et al.
      ESC Guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: The Task Force for the management of acute coronary syndromes (ACS) in patients presenting without persistent ST-segment elevation of the European Society of Cardiology (ESC).
      • Thygesen K.
      • Alpert J.S.
      • Jaffe A.S.
      • et al.
      Third universal definition of myocardial infarction.
      Delays in “rule-in” of acute myocardial infarction may increase mortality and morbidity, whereas delays in “rule-out” may lead to prolonged assessments, unnecessary investigations, and patient anxiety, as welll as contribute to expensive overcrowding in the emergency department.
      • Forberg J.L.
      • Henriksen L.S.
      • Edenbrandt L.
      • Ekelund U.
      Direct hospital costs of chest pain patients attending the emergency department: a retrospective study.
      Clinical assessment, the 12-lead electrocardiogram (ECG), and cardiac troponin (cTn) form the 3 pillars for the early diagnosis of acute myocardial infarction in the emergency department. The recently developed high-sensitivity cardiac troponin (hs-cTn) assays, which allow measurement of even low cTn concentrations with high precision, have been shown to largely overcome the sensitivity deficit of conventional cTn within the first hours and provide higher overall diagnostic accuracy in the diagnosis of acute myocardial infarction.
      • Reichlin T.
      • Hochholzer W.
      • Bassetti S.
      • et al.
      Early diagnosis of myocardial infarction with sensitive cardiac troponin assays.
      • Keller T.
      • Zeller T.
      • Ojeda F.
      • et al.
      Serial changes in highly sensitive troponin I assay and early diagnosis of myocardial infarction.
      • Keller T.
      • Zeller T.
      • Peetz D.
      • et al.
      Sensitive troponin I assay in early diagnosis of acute myocardial infarction.
      These studies also revealed that the classical diagnostic interpretation of cTn as a dichotomous variable (“troponin-negative” and “troponin-positive”) no longer seems appropriate, as the positive predictive value (PPV) for acute myocardial infarction of being “troponin-positive” was only 50%-60%.
      • Reichlin T.
      • Hochholzer W.
      • Bassetti S.
      • et al.
      Early diagnosis of myocardial infarction with sensitive cardiac troponin assays.
      • Keller T.
      • Zeller T.
      • Ojeda F.
      • et al.
      Serial changes in highly sensitive troponin I assay and early diagnosis of myocardial infarction.
      • Keller T.
      • Zeller T.
      • Peetz D.
      • et al.
      Sensitive troponin I assay in early diagnosis of acute myocardial infarction.
      • Thygesen K.
      • Mair J.
      • Giannitsis E.
      • et al.
      How to use high-sensitivity cardiac troponins in acute cardiac care.
      • Safford M.M.
      • Parmar G.
      • Barasch C.S.
      • et al.
      Hospital laboratory reporting may be a barrier to detection of ‘microsize’ myocardial infarction in the US: an observational study.
      • Reichlin T.
      • Schindler C.
      • Drexler B.
      • et al.
      One-hour rule-out and rule-in of acute myocardial infarction using high-sensitivity cardiac troponin T.
      • Rubini Gimenez M.
      • Twerenbold R.
      • Reichlin T.
      • et al.
      Direct comparison of high-sensitivity-cardiac troponin I vs. T for the early diagnosis of acute myocardial infarction.
      • Reiter M.
      • Twerenbold R.
      • Reichlin T.
      • et al.
      Early diagnosis of acute myocardial infarction in the elderly using more sensitive cardiac troponin assays.
      • Reichlin T.
      • Irfan A.
      • Twerenbold R.
      • et al.
      Utility of absolute and relative changes in cardiac troponin concentrations in the early diagnosis of acute myocardial infarction.
      • Haaf P.
      • Drexler B.
      • Reichlin T.
      • et al.
      High-sensitivity cardiac troponin in the distinction of acute myocardial infarction from acute cardiac noncoronary artery disease.
      Unfortunately, the best possible way to interpret and clinically use hs-cTn levels in the early diagnosis of acute myocardial infarction is still debated.
      In a recent pilot study, a novel hs-cTnT 1h-algorithm has been shown to allow accurate rule-out and rule-in of acute myocardial infarction within 1 hour in up to 75% of patients.
      • Reichlin T.
      • Schindler C.
      • Drexler B.
      • et al.
      One-hour rule-out and rule-in of acute myocardial infarction using high-sensitivity cardiac troponin T.
      This algorithm is based on 2 concepts: First, the interpretation of hs-cTnT as a quantitative variable where the proportion of patients indeed suffering from acute myocardial infarction continuously increases with increasing hs-cTn values.
      • Reichlin T.
      • Schindler C.
      • Drexler B.
      • et al.
      One-hour rule-out and rule-in of acute myocardial infarction using high-sensitivity cardiac troponin T.
      • Haaf P.
      • Drexler B.
      • Reichlin T.
      • et al.
      High-sensitivity cardiac troponin in the distinction of acute myocardial infarction from acute cardiac noncoronary artery disease.
      Second, early absolute concentration changes within 1 hour provide incremental diagnostic information when added to baseline levels, with the combination acting as reliable surrogates for late concentrations at 3 hours or 6 hours.
      • Reichlin T.
      • Schindler C.
      • Drexler B.
      • et al.
      One-hour rule-out and rule-in of acute myocardial infarction using high-sensitivity cardiac troponin T.
      • Reichlin T.
      • Irfan A.
      • Twerenbold R.
      • et al.
      Utility of absolute and relative changes in cardiac troponin concentrations in the early diagnosis of acute myocardial infarction.
      • Haaf P.
      • Drexler B.
      • Reichlin T.
      • et al.
      High-sensitivity cardiac troponin in the distinction of acute myocardial infarction from acute cardiac noncoronary artery disease.
      Many experts remained skeptical about the safety of the hs-cTnT 1h-algorithm, particularly its possible extrapolation to other hs-cTn assays and its wider applicability.
      • Newby L.K.
      Myocardial infarction rule-out in the emergency department: are high-sensitivity troponins the answer?: comment on “One-hour rule-out and rule-in of acute myocardial infarction using high-sensitivity cardiac troponin T”.
      Accordingly, this novel triage concept has not been adopted clinically until now and the following clinically relevant questions have not been addressed: First, is it possible to derive and validate a similar 1h-algorithm for hs-cTnI in order to possibly extend the former finding to cTnI, which is the analyte most often used worldwide, and specifically to the most sensitive clinically available hs-cTn assays?
      • Thygesen K.
      • Mair J.
      • Giannitsis E.
      • et al.
      How to use high-sensitivity cardiac troponins in acute cardiac care.
      • Rubini Gimenez M.
      • Twerenbold R.
      • Reichlin T.
      • et al.
      Direct comparison of high-sensitivity-cardiac troponin I vs. T for the early diagnosis of acute myocardial infarction.
      • Cullen L.
      • Mueller C.
      • Parsonage W.A.
      • et al.
      Validation of high-sensitivity troponin I in a 2-hour diagnostic strategy to assess 30-day outcomes in emergency department patients with possible acute coronary syndrome.
      Second, what are the specific cutoff values for hs-cTnI that allow safe rule-out and accurate rule-in within 1 hour? Third, how many patients can be assigned rule-out within 1 hour? Fourth, how many patients can be assigned rule-in within 1 hour? Fifth, how does the 1h-algorithm compare vs the classical dichotomous interpretation of hs-cTnI or the current standard of care combining the 12-lead ECG with hs-cTn? Sixth, does a 2-hour hs-cTnI concentration add information for patients classified as “observe”?

      Methods

      Study Design and Population

      Advantageous Predictors of Acute Coronary Syndrome Evaluation (APACE) is an ongoing prospective international multicenter study designed to advance the early diagnosis of acute myocardial infarction.
      • Reichlin T.
      • Hochholzer W.
      • Bassetti S.
      • et al.
      Early diagnosis of myocardial infarction with sensitive cardiac troponin assays.
      • Reichlin T.
      • Schindler C.
      • Drexler B.
      • et al.
      One-hour rule-out and rule-in of acute myocardial infarction using high-sensitivity cardiac troponin T.
      • Reiter M.
      • Twerenbold R.
      • Reichlin T.
      • et al.
      Early diagnosis of acute myocardial infarction in the elderly using more sensitive cardiac troponin assays.
      • Reichlin T.
      • Irfan A.
      • Twerenbold R.
      • et al.
      Utility of absolute and relative changes in cardiac troponin concentrations in the early diagnosis of acute myocardial infarction.
      • Haaf P.
      • Drexler B.
      • Reichlin T.
      • et al.
      High-sensitivity cardiac troponin in the distinction of acute myocardial infarction from acute cardiac noncoronary artery disease.
      From April 2006 to September 2012, consecutive patients older than 18 years presenting to the emergency department with symptoms suggestive of acute myocardial infarction with an onset or peak within the last 12 hours were recruited at 9 sites in 3 countries (Switzerland, Spain, and Italy) after written informed consent was obtained.
      Enrollment was completely independent of renal function at presentation; only patients with terminal renal failure on chronic dialysis were excluded. For this analysis, patients were also excluded if 1) the final diagnosis remained unclear after adjudication (n = 69), or 2) ST-segment elevation myocardial infarction was the adjudicated final diagnosis, because biomarkers are considered to be of limited clinical value in these patients (n = 76). Among the remaining 2308 patients, samples at presentation as well as after 1 hour for measurement of hs-cTnI were available in 1811 patients. The most common reasons for missing values after 1 hour were early transfer to the catheterization laboratory or coronary care unit and diagnostic procedures around the 1-hour window (eg, computed tomography scan) that precluded blood draw at 1 hour. The study was carried out according to the principles of the Declaration of Helsinki and approved by the local ethics committees. The authors designed the study, gathered and analyzed the data, vouched for the data and analysis, wrote the paper, and decided to publish.

      Routine Clinical Assessment

      All patients underwent a clinical assessment that included medical history, physical examination, 12-lead ECG, continuous ECG monitoring, pulse oximetry, standard blood test, and chest radiography. Levels of cTn were measured at presentation and serially thereafter as long as clinically indicated. Timing and treatment of patients were left to discretion of the attending physician.

      Adjudicated Final Diagnosis

      Adjudication of the final diagnosis was performed centrally in a core laboratory (University Hospital Basel) and also included serial levels of Roche hs-cTnT (Roche Diagnostics, Mannheim, Germany) in order to take advantage of the higher sensitivity and higher overall diagnostic accuracy offered by hs-cTn assays
      • Rubini Gimenez M.
      • Twerenbold R.
      • Reichlin T.
      • et al.
      Direct comparison of high-sensitivity-cardiac troponin I vs. T for the early diagnosis of acute myocardial infarction.
      • Apple F.S.
      • Jesse R.L.
      • Newby L.K.
      • et al.
      National Academy of Clinical Biochemistry and IFCC Committee for Standardization of Markers of Cardiac Damage Laboratory Medicine Practice Guidelines: analytical issues for biochemical markers of acute coronary syndromes.
      (this allows the additional detection of small acute myocardial infarctions that were missed by the adjudication based on conventional cTn assays). Two independent cardiologists reviewed all available medical records–patient history, physical examination, results of laboratory testing (including hs-cTnT levels), radiologic testing, ECG, echocardiography, cardiac exercise stress test, lesion severity, and morphology in coronary angiography–pertaining to the patient from the time of emergency department presentation to 90-day follow-up. In situations of disagreement about the diagnosis, cases were reviewed and adjudicated in conjunction with a third cardiologist.
      Acute myocardial infarction was defined and cTn levels were interpreted as recommended in current guidelines.
      • Anderson J.L.
      • Adams C.D.
      • Antman E.M.
      • et al.
      2012 ACCF/AHA focused update incorporated into the ACCF/AHA 2007 guidelines for the management of patients with unstable angina/non-ST-elevation myocardial infarction: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines.
      • Hamm C.W.
      • Bassand J.P.
      • Agewall S.
      • et al.
      ESC Guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: The Task Force for the management of acute coronary syndromes (ACS) in patients presenting without persistent ST-segment elevation of the European Society of Cardiology (ESC).
      • Thygesen K.
      • Alpert J.S.
      • Jaffe A.S.
      • et al.
      Third universal definition of myocardial infarction.
      In brief, acute myocardial infarction was diagnosed when there was evidence of myocardial necrosis in association with a clinical setting consistent with myocardial ischemia. Myocardial necrosis was diagnosed by at least one cTn value above the 99th percentile together with a significant rise or fall.
      • Apple F.S.
      • Jesse R.L.
      • Newby L.K.
      • et al.
      National Academy of Clinical Biochemistry and IFCC Committee for Standardization of Markers of Cardiac Damage Laboratory Medicine Practice Guidelines: analytical issues for biochemical markers of acute coronary syndromes.
      • Apple F.S.
      • Wu A.H.
      • Jaffe A.S.
      European Society of Cardiology and American College of Cardiology guidelines for redefinition of myocardial infarction: how to use existing assays clinically and for clinical trials.
      Criteria to define rise or fall are described in the Methods section in the Appendix (available online).

      Measurement of hs-cTnI

      The measurement of hs-cTnI is described in the Appendix (available online).

      Follow-up and Clinical End Points

      After hospital discharge, patients were contacted after 3 and 12 months by telephone calls or in written form. Furthermore, information about death was obtained from the hospitals' electronic patient documentation, the family physicians' records, and national registries on mortality from each country. The primary prognostic end point was 30 days mortality.

      Algorithm Derivation and Validation

      The algorithm for use of hs-cTnI was developed in a randomly selected derivation sample of 906 patients (derivation cohort). The algorithm incorporates both baseline hs-cTnI levels and absolute hs-cTnI changes within 1 hour. Selection of these 2 parameters was based on the previously published, very high diagnostic accuracy of their combination.
      • Keller T.
      • Zeller T.
      • Ojeda F.
      • et al.
      Serial changes in highly sensitive troponin I assay and early diagnosis of myocardial infarction.
      • Reichlin T.
      • Schindler C.
      • Drexler B.
      • et al.
      One-hour rule-out and rule-in of acute myocardial infarction using high-sensitivity cardiac troponin T.
      • Reichlin T.
      • Irfan A.
      • Twerenbold R.
      • et al.
      Utility of absolute and relative changes in cardiac troponin concentrations in the early diagnosis of acute myocardial infarction.
      • Haaf P.
      • Drexler B.
      • Reichlin T.
      • et al.
      High-sensitivity cardiac troponin in the distinction of acute myocardial infarction from acute cardiac noncoronary artery disease.
      Optimal thresholds for rule-out were selected using receiver-operating characteristics curve to allow a minimal sensitivity of 99% (and 95% for an alternative option) for baseline values together with a negative predictive value (NPV) of at least 95% for the absolute cTnI change within 1 hour. Optimal thresholds for rule-in were obtained based on a classification and regression tree (CART) analysis.
      • Breiman L.
      Classification and Regression Trees.
      • Fonarow G.C.
      • Adams Jr., K.F.
      • Abraham W.T.
      • et al.
      Risk stratification for in-hospital mortality in acutely decompensated heart failure: classification and regression tree analysis.
      The CART algorithm provides a sequence of partitions of a given data set aimed at optimizing the prediction of a binary outcome variable. Each subsequent partition is obtained by splitting one of the preceding partition sets (nodes) into 2 parts. If quantitative predictor variables are used, a pair of new nodes is obtained by splitting an existing node at a given threshold value of one of these variables. The algorithm stops if no further improvement is possible or if any further split would violate a predefined criterion (eg, on the minimal node size).
      • Breiman L.
      Classification and Regression Trees.
      • Fonarow G.C.
      • Adams Jr., K.F.
      • Abraham W.T.
      • et al.
      Risk stratification for in-hospital mortality in acutely decompensated heart failure: classification and regression tree analysis.
      Nodes in the CART tree were constrained to have a minimal number of cases of 20 in parent and child nodes. In addition to baseline hs-cTnI levels and absolute hs-cTnI changes within 1 hour, age (as a continuous variable), relative hs-cTnI changes within 1 hour, sex, and ECG features (signs of ischemia or not) were included in the CART model as well. Then the algorithm developed in the derivation sample was tested for its diagnostic accuracy in a validation sample consisting of the remaining 906 subjects. The optimal decision values derived in the derivation cohort were rounded to give whole values in ng/L for the prospective testing in the validation cohort.

      Statistical Analysis

      Continuous variables are presented as mean (SD) or median (interquartile range); and categorical variables as numbers and percentages. Differences in baseline characteristics between patients with and without acute myocardial infarction and between patients in the derivation and validation cohort were assessed using the Mann-Whitney test for continuous variables and the Pearson chi-squared test for categorical variables. The primary outcome measures of this analysis were the NPV for acute myocardial infarction in the rule-out group, the PPV for acute myocardial infarction in the rule-in group, and the percentage of patients assigned the observational zone (all in the validation cohort). NPV and PPV of the hs-cTnI 1h-algorithm were compared with that of the classical dichotomous interpretation of hs-cTnI, as well as the current standard of care combining hs-cTnI with ischemic ECG findings. Mortality during 30 and 360 days of follow-up according to the classification provided by the hs-cTnI algorithm was plotted in Kaplan-Meier curves, and the log-rank test was used to assess differences in mortality between groups. Hazard ratios and 95% confidence intervals (CIs) were obtained from Cox proportional hazard models to quantify the magnitudes of group differences. All hypothesis testing was 2-tailed, and P < .05 was considered statistically significant. All statistical analyses were performed using SPSS for Windows 22.0 (SPSS Inc., Chicago, IL) and MedCalc 11.2.1.0 (MedCalc, Ostend, Belgium).

      Results

      Baseline characteristics of 1811 patients presenting to the emergency department with suspected acute myocardial infarction are shown in Table 1. The adjudicated final diagnosis was acute myocardial infarction in 329 patients (18%), unstable angina in 179 (10%), cardiac disease others than coronary artery disease in 258 (15%), noncardiac symptoms in 964 (53%), and unknown origin in 80 (4%).
      Table 1Baseline Characteristics of Patients
      CharacteristicAll

      N = 1811
      AMI

      n = 329
      No AMI

      n = 1482
      P-Value
      Age, y
       Median627260<.001
       IQR49-6260-8047-73
      Risk factors, n (%)
       Hypertension1115 (62)261 (79)854 (58)<.001
       Hypercholesterolemia912 (50)217 (66)695 (47)<.001
       Diabetes330 (18)92 (28)238 (16)<.001
       Current or previous smoking1129 (62)219 (66)910 (61).09
       Family history627 (35)129 (39)498 (34).06
      History, n (%)
       Coronary artery disease627 (35)164 (50)463 (31)<.001
       Previous AMI426 (24)117 (36)309 (21)<.001
       Previous revascularization495 (27)117 (36)378 (26)<.001
       Peripheral artery disease109 (6)43 (13)66 (5)<.001
       Previous stroke101 (6)33 (10)68 (5)<.001
      ECG findings, n (%)
       Left bundle branch block59 (3)19 (5)40 (3)<.001
       ST-segment elevation29 (2)6 (2)23 (2)<.001
       St-segment depression183 (10)98 (30)85 (6)<.001
       T-wave inversion233 (13)78 (24)155 (10)<.001
       No significant changes1344 (74)153 (46)1192 (80)<.001
      Body mass index (kg/m2)
       Median272626.490
       IQR24-3024-2924-30
      eGFR (mL/min/1.73 m2)
       Median857787<.001
       IQR69-10255-9455-94
      Medication at presentation, n (%)
       ASA656 (36)171 (52)485 (33)<.001
       Vitamin K antagonists144 (8)31 (9)113 (8).28
       B-blockers614 (34)141 (43)473 (32)<.001
       Statins635 (44)144 (44)491 (33)<.001
       ACEIs/ARBs685 (38)173 (52)512 (35)<.001
       Calcium antagonists256 (14)68 (21)188 (13)<.001
       Nitrates212 (12)70 (21)142 (10)<.001
      ACEI = angiotensin-converting enzyme inhibitor; AMI = acute myocardial infarction; ARB = angiotensin receptor blocker; ASA = acetyl salicylic acid; ECG = electrocardiogram; eGFR = estimated glomerular filtration rate; IQR = interquartile range.

      Derivation of the hs-cTnI 1h-Algorithm for the Diagnosis of Acute Myocardial Infarction

      Baseline characteristics of the derivation and validation cohort were similar (Supplemental Table, available online). For rule-out of acute myocardial infarction in the derivation cohort, the optimal threshold was defined as a baseline < 5.2 ng/L and an absolute change within 1 hour of < 1.9 ng/L. With these values, 56% of patients could be classified as “rule-out” (Figure 1A). The sensitivity and the NPV for acute myocardial infarction in the “rule-out” zone were 97.6% (95% CI, 93.8%-99.3%) and 99.2% (95% CI, 98.0%-99.8%), respectively. Details of patients with acute myocardial infarction that were missed by the hs-cTnI 1h-algorithm are described in Table 2. An alternative 1h-algorithm is described in the Appendix (available online).
      Figure thumbnail gr1
      Figure 1One-hour algorithm for the diagnosis of acute myocardial infarction (AMI) using hs-cTnI in the derivation (A) and validation cohort (B). Oh = hs-cTnI at presentation to the ED; Delta 1h = absolute change of hs-cTnI within 1 hour; Sens = sensitivity; Spec = specificity; NPV = negative predictive value; PPV = positive predictive value.
      Table 2Baseline Characteristics of the Patients with AMI Incorrectly Ruled Out by the hs-cTnI 1h-Algorithm
      GroupAge, ySexTime Since CPOHistory of CADhs-cTnT (ng/L; Peak Value italicized)hs-cTnI (ng/L; Peak Value italicized)Clinical Discharge DiagnosisCABG PerformedPTCA Performed
      0 h1 h2 h3 h0 h1 h2 h3 h
      Derivation76Male4 hYes20.220.417.717.74.34.64.54.3Other (unknown origin)NoNo
      Derivation75Male5 hYes39.035.04.55.3Rhythmogenic (atrial fibrillation)NoNo
      Derivation93Female9 hYes41.038.03.62.4Other (unknown origin)NoNo
      Derivation52Male4 hNo46.832.42.82.82.5Other (unknown origin)NoNo
      Validation77Male5 hNo55.052.944.94.74.85.1Other (unknown origin)NoNo
      Validation73Male4 hNo33.331.628.03.43.9Other (unknown origin)NoNo
      AMI = acute myocardial infarction; CABG = coronary artery bypass graft; CAD = coronary artery disease; CPO = chest pain onset; PTCA = percutaneous transluminal coronary angioplasty.
      For “rule-in” of acute myocardial infarction, the optimal threshold was an absolute change of ≥ 5.7 ng/L within 1 hour irrespective of the baseline value. All other variables in the CART analysis (baseline hs-cTnI, age, sex, ischemic ECG changes, and duration of symptoms) did not improve accuracy and did not emerge as contributors to the final decision tree. The specificity and PPV for acute myocardial infarction in the “rule-in” zone were 94.5% (95% CI, 92.6%-96.0%) and 74.9% (95% CI, 67.4-81.3%), respectively. The final adjudicated diagnosis of the ruled-in patients with diagnosis other than acute myocardial infarction (n = 41) were cardiac arrhythmias (n = 10), unstable angina (n = 7), heart failure (n = 7), noncardiac causes (n = 7), myocarditis (n = 5), hypertensive emergency (n = 3), pulmonary embolism (n = 1), and takotsubo cardiomyopathy (n = 1).
      Patients fulfilling neither the rule-out nor rule-in criteria (26%) were assigned to “observe.” The incidence of acute myocardial infarction was 15% in these patients.

      Validation of the hs-cTnI 1h-Algorithm for the Diagnosis of Acute Myocardial Infarction

      After applying the hs-cTnI 1h-algorithm developed in the derivation cohort (rounded to give whole values in ng/L) to the validation cohort, 50.5% of patients could be classified as “rule-out,” 19% as “rule-in,” and 30.5% as “observe” (Figure 1B). In the validation cohort, the sensitivity and the NPV for acute myocardial infarction in the “rule-out” zone were 98.8% (95% CI, 95.7%-99.9%) and 99.6% (95% CI, 98.4%-100%), respectively. The specificity and the PPV for acute myocardial infarction in the “rule-in” zone were 93.8% (95% CI, 91.8%-95.4%) and 73.9% (95% CI, 66.7%-80.2%), respectively. The final adjudicated diagnoses of the ruled-in patients with diagnoses other than acute myocardial infarction (n = 42) were cardiac arrhythmias (n = 14), noncardiac causes (n = 5), unstable angina (n = 4), myocarditis (n = 5), heart failure (n = 6), takotsubo cardiomyopathy (n = 2), pulmonary embolism (n = 3) and hypertensive emergency (n = 3).

      Comparison with the Classical Interpretation of Hs-cTnI

      A single cut-off value for hs-cTnI (99th percentile, 26.2 ng/L) at presentation resulted in a sensitivity and NPV of 69.6% (95% CI, 64.3-74.5) and 93.2% (95% CI, 91.8-94.5), and a specificity and PPV of 92.9% (95% CI, 91.5-94.2) and 68.6% (95% CI, 63.3-73.5), respectively. NPV of the 1h-algorithm was higher compared with the classical dichotomous interpretation of hs-cTnI (P < .001); PPV was not significantly higher (P = .895).

      Comparison with the Current Standard of Care (hs-cTnI + ECG)

      Combining the classical interpretation of hs-cTnI with ischemic ECG findings (ST-elevation, ST-depression, T-inversion, complete left bundle branch block not known to be old), normal hs-cTnI levels, and no ischemic ECG findings at presentation had a sensitivity and NPV of 79.0% (95% CI, 74.2-83.3) and 94.6% (95% CI, 93.2-95.8), respectively, while “rule-in” when either one (or both) were positive had a specificity and PPV of 79.0 (95% CI, 74.2-83.3) and 48.0% (95% CI, 43.7-52.3), respectively. The NPV and the PPV of 1h-algorithm were both significantly higher compared with that of the current standard of care (hs-cTnI + ECG; both P < .001).

      Diagnostic Performance of the 1h-Algorithm among Predefined Subgroups

      The NPV for acute myocardial infarction in the “rule-out” zone was similar among all predefined subgroups, whereas the PPV for acute myocardial infarction in the “rule-in” zone seemed to be higher in men and patients with preexisting coronary artery disease (Figure 2).
      Figure thumbnail gr2
      Figure 2Negative predictive value (NPV) and positive predictive value (PPV) of the 1h-algorithm in study subgroups in the validation group. Forest plots indicating (A) NPV and (B) PPV among different study subgroups including interaction P-value. CAD = coronary artery disease.

      Comparison of 1-Hour vs Other Time Points

      The area under the curve (AUC) of the combination of hs-cTnI at presentation with 1-hour absolute change (AUC 0.95; 95% CI, 0.93-0.96) for the diagnosis of acute myocardial infarction was significantly higher as compared with the AUC of hs-cTnI at presentation (AUC 0.93; 95% CI, 0.92-0.94; P < .001), and comparable with the combination of hs-cTnI at presentation with 2-hour absolute change (AUC 0.96; 95% CI, 0.95-0.97; P = ns).

      “Observe” Group

      Among patients classified as “observe” by the hs-cTnI 1h-algorithm, the AUC of the hs-cTnI concentration at 2 hours for the diagnosis of acute myocardial infarction was 0.81 (95% CI, 0.76-0.87).

      Undetectable Levels of hs-cTnI

      In the overall cohort, 14% of patients had undetectable levels of hs-cTnI. None of these patients were finally adjudicated to have an acute myocardial infarction, which resulted in a sensitivity and NPV of 100%. None of these patients had an increased hs-cTnI concentration above the 99th percentile at 2 hours.

      Mortality During Follow-up

      Cumulative 30-day mortality in the validation cohort was 0%, 1.4%, and 4.7% (P < .001, log-rank test) in patients classified as “rule-out,” “observe,” and “rule-in,” respectively (Figure 3A). Cumulative 360-day mortality was 1.3%, 5.1%, and 11% in patients classified as “rule-out,” “observe,” and “rule-in,” respectively (P < .001, log rank test; Figure 3B). The hazard ratio for the risk of death within 360 days was 3.9 (95% CI, 1.5-10.3; P = .005) for patients in the observational zone and 8.9 (95% CI, 3.6-22.3; P < .001) for patients in the rule-in group compared with patients in the rule-out group. Among patients dying during 360-day follow-up (n = 77), 36 patients (47%) had an adjudicated diagnosis of acute myocardial infarction. Causes of death during 360-day follow up were: 43% cardiac cause, 17% pulmonal cause, and 40% unknown cause.
      Figure thumbnail gr3
      Figure 3Kaplan-Meier curves for the cumulative mortality according to classification provided by the 1h hs-cTnI algorithm into “rule-out” (n = 457), “observational zone” (n = 276), and “rule-in” (n = 172) provided by the hs-cTnI 1-hour algorithm. Differences in mortality were assessed using the log-rank test. (A) mortality during first 30 days, (B) mortality during first 365 days.

      Discussion

      This international multicenter study was performed to prospectively develop and validate a 1h-algorithm for rapid rule-out and rule-in of acute myocardial infarction based on hs-cTnI. We found a similar performance of the 1h-hs-cTnI algorithm as recently described for the 1h-hs-cTnT algorithm in a pilot study,
      • Reichlin T.
      • Schindler C.
      • Drexler B.
      • et al.
      One-hour rule-out and rule-in of acute myocardial infarction using high-sensitivity cardiac troponin T.
      indicating that accurate rule-out and rule-in is feasible much more rapidly than suggested in current American Heart Association/American College of Cardiology
      • Anderson J.L.
      • Adams C.D.
      • Antman E.M.
      • et al.
      2012 ACCF/AHA focused update incorporated into the ACCF/AHA 2007 guidelines for the management of patients with unstable angina/non-ST-elevation myocardial infarction: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines.
      • Fonarow G.C.
      • Adams Jr., K.F.
      • Abraham W.T.
      • et al.
      Risk stratification for in-hospital mortality in acutely decompensated heart failure: classification and regression tree analysis.
      or European Society of Cardiology
      • Hamm C.W.
      • Bassand J.P.
      • Agewall S.
      • et al.
      ESC Guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: The Task Force for the management of acute coronary syndromes (ACS) in patients presenting without persistent ST-segment elevation of the European Society of Cardiology (ESC).
      guidelines in many patients. We report 7 major findings:
      First, the NPV for acute myocardial infarction in the “rule-out” zone defined only by hs-cTnI levels at presentation and the change within 1 hour was 99.6% (95% CI, 98.4%-100%) in the validation cohort. This algorithm assigned 50.5% of patients to the rule-out zone. As in clinical practice, the 1h-hs-cTnI algorithm would of course always be used in conjunction with full clinical assessment, including patient history and examination, and the 12-lead ECG; these additional clinical tools should allow clinicians to further increase the NPV and thereby approach 100%. Thereby, the use of the 1h-hs-cTn algorithm can be expected to help avoid unnecessary and costly imaging procedures in low-risk patients.
      • Redberg R.F.
      Coronary CT angiography for acute chest pain.
      Second, this innovative approach achieved a higher NPV as compared with the classical dichotomous interpretation of hs-cTnI (“cTn-negative”), and even higher compared with the current standard of care combining hs-cTnI with ECG findings.
      Third, the PPV for acute myocardial infarction in the “rule-in” zone was 74% in the validation cohort. Many of the patients in the “rule-in” zone with diagnosis other than acute myocardial infarction did have conditions that usually still require coronary angiography for accurate diagnosis, including takotsubo cardiomyopathy, myocarditis, and unstable angina.
      • Hamm C.W.
      • Bassand J.P.
      • Agewall S.
      • et al.
      ESC Guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: The Task Force for the management of acute coronary syndromes (ACS) in patients presenting without persistent ST-segment elevation of the European Society of Cardiology (ESC).
      Therefore, the immediate clinical consequence of being assigned to the “rule-in” zone would be urgent coronary angiography, unless clinical assessment would indicate another obvious condition associated with acute cardiomyocyte damage, for example, acute heart failure, tachyarrhythmia, or hypertensive crisis.
      • Thygesen K.
      • Alpert J.S.
      • Jaffe A.S.
      • et al.
      Third universal definition of myocardial infarction.
      The “rule-in” zone of this 1h-hs-cTnI algorithm is more precisely defined than, for example, in the 2011 European Society of Cardiology algorithm.
      • Anderson J.L.
      • Adams C.D.
      • Antman E.M.
      • et al.
      2012 ACCF/AHA focused update incorporated into the ACCF/AHA 2007 guidelines for the management of patients with unstable angina/non-ST-elevation myocardial infarction: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines.
      • Hamm C.W.
      • Bassand J.P.
      • Agewall S.
      • et al.
      ESC Guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: The Task Force for the management of acute coronary syndromes (ACS) in patients presenting without persistent ST-segment elevation of the European Society of Cardiology (ESC).
      As the “rule-in” of acute myocardial infarction in patients with mild elevations in hs-cTn often is challenging for clinicians,
      • Thygesen K.
      • Mair J.
      • Giannitsis E.
      • et al.
      How to use high-sensitivity cardiac troponins in acute cardiac care.
      • Mueller C.
      Biomarkers and acute coronary syndromes: an update.
      it is a key advantage of this 1h- hs-cTnI algorithm to provide more detailed guidance in this difficult setting.
      Fourth, the PPV achieved with the 1h-hs-cTnI algorithm was higher as compared with the classical dichotomous interpretation of hs-cTnI (“cTn-positive”), and even higher compared with the current standard of care combining hs-cTnI with ECG findings.
      Fifth, the 1h-hs-cTnI algorithm overall assigned 70% of patients a definite process (either rule-out or rule-in), with 30% of patients remaining in the observational zone. Thereby, the 1h-hs-cTnI algorithm was even more effective in the early triage of acute chest pain patients in comparison with, for example, the recently developed accelerated diagnostic protocol combining the Thrombolysis in Myocardial Infarction Score with cTn or hs-cTn levels at 0 hours and 2 hours,
      • Cullen L.
      • Mueller C.
      • Parsonage W.A.
      • et al.
      Validation of high-sensitivity troponin I in a 2-hour diagnostic strategy to assess 30-day outcomes in emergency department patients with possible acute coronary syndrome.
      • Than M.
      • Aldous S.
      • Lord S.J.
      • et al.
      A 2-hour diagnostic protocol for possible cardiac chest pain in the emergency department: a randomized clinical trial.
      • Than M.
      • Cullen L.
      • Aldous S.
      • et al.
      2-Hour accelerated diagnostic protocol to assess patients with chest pain symptoms using contemporary troponins as the only biomarker: the ADAPT trial.
      or the dual-marker approach combining hs-cTn with copeptin, which assign 20%-40% of patients for rapid “rule-out.”
      • Balmelli C.
      • Meune C.
      • Twerenbold R.
      • et al.
      Comparison of the performances of cardiac troponins, including sensitive assays, and copeptin in the diagnostic of acute myocardial infarction and long-term prognosis between women and men.
      • Keller T.
      • Tzikas S.
      • Zeller T.
      • et al.
      Copeptin improves early diagnosis of acute myocardial infarction.
      • Maisel A.
      • Mueller C.
      • Neath S.X.
      • et al.
      Copeptin helps in the early detection of patients with acute myocardial infarction: primary results of the CHOPIN trial (Copeptin Helps in the early detection Of Patients with acute myocardial INfarction).
      • Möckel M.
      • Searle J.
      • Hamm C.
      • et al.
      Early discharge using single cardiac troponin and copeptin testing in patients with suspected acute coronary syndrome (ACS): a randomized, controlled clinical process study.
      This difference is at least partly explained by the fact that the latter approaches exclusively select patients for “rule-out,” but do not provide guidance for “rule-in.”
      Sixth, among patients classified as “observe” by the hs-cTnI 1h-algorithm, a 2-hour hs-cTnI concentration provided diagnostic values for the diagnosis of AMI.
      Seventh, cumulative 30-day mortality was 0% in patients assigned the “rule-out” zone, further documenting the safety of this approach and the suitability of many of these patients for early discharge.
      Our findings extend and corroborate recent pilot data obtained for the 1h-hs-cTnT algorithm.
      • Reichlin T.
      • Schindler C.
      • Drexler B.
      • et al.
      One-hour rule-out and rule-in of acute myocardial infarction using high-sensitivity cardiac troponin T.
      • Body R.
      • Carley S.
      • McDowell G.
      • et al.
      Rapid exclusion of acute myocardial infarction in patients with undetectable troponin using a high-sensitivity assay.
      Overall, the performance of the hs-cTnI 1h-algorithm was similar to that for hs-cTnT (NPV 100%, PPV 80%, and 77% of patients assigned either rule-out or rule-in).
      • Reichlin T.
      • Schindler C.
      • Drexler B.
      • et al.
      One-hour rule-out and rule-in of acute myocardial infarction using high-sensitivity cardiac troponin T.
      The current finding relating to the 1h-hs-cTnI algorithm methodologically is even stronger, as the adjudication of the final diagnosis was based on a different hs-cTn assay in this analysis, while late samples of the same hs-cTnT assay were used in the 1h-hs-cTnT pilot study.
      • Reichlin T.
      • Schindler C.
      • Drexler B.
      • et al.
      One-hour rule-out and rule-in of acute myocardial infarction using high-sensitivity cardiac troponin T.
      The optimal management of patients assigned to the observational zone likely will be highly individualized. It may include coronary angiography in patients with a high clinical suspicion of acute myocardial infarction, coronary computed tomography angiography in patients with low-to-intermediate likelihood for acute myocardial infarction, a third hs-cTn sample at 2, 3, or 6 hours in many, or no further immediate diagnostic testing when complete clinical evaluation has established (eg, rapid atrial fibrillation or hypertensive crisis as the final diagnosis).
      • Anderson J.L.
      • Adams C.D.
      • Antman E.M.
      • et al.
      2012 ACCF/AHA focused update incorporated into the ACCF/AHA 2007 guidelines for the management of patients with unstable angina/non-ST-elevation myocardial infarction: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines.
      • Hamm C.W.
      • Bassand J.P.
      • Agewall S.
      • et al.
      ESC Guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: The Task Force for the management of acute coronary syndromes (ACS) in patients presenting without persistent ST-segment elevation of the European Society of Cardiology (ESC).
      • Thygesen K.
      • Mair J.
      • Giannitsis E.
      • et al.
      How to use high-sensitivity cardiac troponins in acute cardiac care.
      • Hoffmann U.
      • Truong Q.A.
      • Schoenfeld D.A.
      • et al.
      Coronary CT angiography versus standard evaluation in acute chest pain.
      • Litt H.I.
      • Gatsonis C.
      • Snyder B.
      • et al.
      CT angiography for safe discharge of patients with possible acute coronary syndromes.
      It might be possible to further simplify the “rule-out” in patients with very low (undetectable) hs-cTn levels.
      • Body R.
      • Carley S.
      • McDowell G.
      • et al.
      Rapid exclusion of acute myocardial infarction in patients with undetectable troponin using a high-sensitivity assay.
      • Bandstein N.
      • Ljung R.
      • Johansson M.
      • Holzmann M.J.
      Undetectable high sensitivity cardiac troponin T level in the emergency department and risk of myocardial infarction.
      • Rubini Giménez M.
      • Hoeller R.
      • Reichlin T.
      • et al.
      Rapid rule out of acute myocardial infarction using undetectable levels of high-sensitivity cardiac troponin.
      Recent evidence from several large studies indicated a very high NPV for acute myocardial infarction of very low (undetectable) hs-cTn levels even without any serial sampling. For example, using undetectable levels of hs-cTnI (Abbott), it was possible to safely rule out 14% of patients presenting to the emergency department achieving an NPV of 100%.
      • Body R.
      • Carley S.
      • McDowell G.
      • et al.
      Rapid exclusion of acute myocardial infarction in patients with undetectable troponin using a high-sensitivity assay.
      • Bandstein N.
      • Ljung R.
      • Johansson M.
      • Holzmann M.J.
      Undetectable high sensitivity cardiac troponin T level in the emergency department and risk of myocardial infarction.
      • Rubini Giménez M.
      • Hoeller R.
      • Reichlin T.
      • et al.
      Rapid rule out of acute myocardial infarction using undetectable levels of high-sensitivity cardiac troponin.
      • Christ M.
      • Popp S.
      • Pohlmann H.
      • et al.
      Implementation of high sensitivity cardiac troponin T measurement in the emergency department.
      Potential limitations of the present study merit consideration. First, our study was conducted in emergency department patients with symptoms suggestive of acute myocardial infarction. While the multicenter design ensures that our findings are widely applicable in this setting, additional studies are required to possibly extend our observation to, for example, patients presenting to a general practitioner, a setting with a much lower pretest probability for acute myocardial infarction, or, for example, patients admitted to a coronary care unit, a setting with a much higher pretest probability. Second, the data presented were obtained in a diagnostic study adjudicating the final diagnosis based on the universal definition of acute myocardial infarction. While this is the strongest methodology to quantify the accuracy of the 1h-hs-cTnI algorithm, additional intervention studies applying the 1h-hs-cTnI algorithm prospectively for clinical decision-making will provide important incremental insights. Third, we cannot comment on the performance of the hs-cTnI 1h-algorithm in patients with terminal kidney failure on chronic dialysis, because such patients were excluded from our study. Fourth, we developed this algorithm for the only clinically available hs-cTnI assay. Considering recent results indicating similar diagnostic accuracy for acute myocardial infarction among hs-cTnI assays,
      • Keller T.
      • Zeller T.
      • Ojeda F.
      • et al.
      Serial changes in highly sensitive troponin I assay and early diagnosis of myocardial infarction.
      • Rubini Giménez M.
      • Hoeller R.
      • Reichlin T.
      • et al.
      Rapid rule out of acute myocardial infarction using undetectable levels of high-sensitivity cardiac troponin.
      • Giannitsis E.
      • Katus H.A.
      Antiplatelet therapy – ticagrelor.
      it is likely that similar 1h-algorithms can be developed for other precommercial hs-cTn and possibly also for sensitive cTnI assays. Of course, each of them would require derivation and validation using stringent methodology, as done in this study.
      In conclusion, using a simple algorithm incorporating hs-cTnI baseline values and absolute changes within the first hour, a safe rule-out as well as an accurate rule-in of acute myocardial infarction could be performed within 1 hour in 70% of all patients with acute chest pain. This algorithm seems to be safe, significantly shortens the time needed for rule-out and rule-in of acute myocardial infarction, and may obviate the need for prolonged monitoring and serial blood sampling in many patients presenting to the emergency department with suspected acute myocardial infarction.

      Acknowledgment

      We thank the patients who participated in the study, the staff of the participating emergency departments, the research coordinators, and the laboratory technicians (particularly Michael Freese, Claudia Stelzig, Esther Garrido, Irina Klimmeck, Melanie Wieland, Janine Voegele, Beate Hartmann, and Fausta Chiaverio) for their most valuable efforts.

      Appendix.

      • Supplemental Methods
        • 1)
          Use of local conventional cTn values for adjudication of final diagnoses
        • 2)
          Use of hs-cTnT for adjudication of final diagnoses
        • 3)
          Assumption of linearity of absolute changes within the first hours
        • 4)
          Measurement of hs-cTnI
      • Supplemental Results
        • 1)
          Alternative algorithm
      • Supplemental Tables and Figures
        • 1)
          Supplemental Table
        • 2)
          Supplemental Figure 1
        • 3)
          Supplemental Figure 2
      • Supplemental References

      Supplemental Methods

      Use of Local Conventional cTn Values for Adjudication of Final Diagnoses

      For the Roche cardiac troponin T (cTnT) 4th generation assay, the 10% coefficient of variation (CV) level is 0.035 μg/L. The laboratories of the participating sites reported only 2 decimals; therefore, 0.04 μg/L was used as a cutoff for myocardial necrosis. In order to fulfill the criteria of a significant change (30% of 99th percentile or 10% CV level), a patient would, for example, need to have a level of < 0.01 ug/L at presentation and 0.04 μg/L at 6 hours. A patient would also qualify if the first level is 0.02 μg/L and the second 0.04 μg/L. A patient would not fulfill the criteria if the first level is 0.03 μg/L and the second is 0.04 μg/L. If the first level is 0.04 μg/L, the second level needs to be at least 0.06 μg/L.
      For the Abbott AxsymcTnI ADV (Abbott Laboratories, Abbot Park, IL), the 10% CV level is 0.16 μg/L. A patient having 0.16 μg/L at presentation would meet the criteria for significant change if the second were ≥ 0.21 μg/L. A patient having < 0.12 μg/L at presentation (limit of detection) would qualify if the second is > 0.16 μg/L.
      For the Beckmann Coulter AccucTnI (Beckmann Coulter, Brea, CA), the 10% CV level is 0.06 μg/L. A patient having 0.06 μg/L at presentation would qualify if the second is ≥ 0.08 μg/L. A patient having 0.05 at presentation would qualify if the second is 0.07 μg/L, but not 0.06 μg/L. A patient having undetectable cTnI (cTnI < 0.01 μg/L) at presentation would qualify if the second is ≥ 0.06 μg/L.

      Use of hs-cTnT for Adjudication of Final Diagnoses

      In order to identify additional patients with small acute myocardial infarctions that were missed by the adjudication using the less-sensitive conventional cTn assays, a second adjudication using hs-cTnT was performed in all nonacute myocardial infarction patients according to the first adjudication. For hs-cTnT, the 99th percentile (14 ng/L) was used as cutoff for myocardial necrosis.
      • Apple F.S.
      • Wu A.H.
      • Jaffe A.S.
      European Society of Cardiology and American College of Cardiology guidelines for redefinition of myocardial infarction: how to use existing assays clinically and for clinical trials.
      • Apple F.S.
      • Jesse R.L.
      • Newby L.K.
      • et al.
      National Academy of Clinical Biochemistry and IFCC Committee for Standardization of Markers of Cardiac Damage Laboratory Medicine Practice Guidelines: analytical issues for biochemical markers of acute coronary syndromes.
      Absolute changes in hs-cTnT were used to determine significant changes based on the diagnostic superiority of absolute over relative changes.
      • Reichlin T.
      • Irfan A.
      • Twerenbold R.
      • et al.
      Utility of absolute and relative changes in cardiac troponin concentrations in the early diagnosis of acute myocardial infarction.
      • Thygesen K.
      • Mair J.
      • Giannitsis E.
      • et al.
      How to use high-sensitivity cardiac troponins in acute cardiac care.
      Based on studies of the biological variation of cTn
      • Vasile V.C.
      • Saenger A.K.
      • Kroning J.M.
      • Jaffe A.S.
      Biological and analytical variability of a novel high-sensitivity cardiac troponin T assay.
      • Wu A.H.
      • Lu Q.A.
      • Todd J.
      • Moecks J.
      • Wians F.
      Short- and long-term biological variation in cardiac troponin I measured with a high-sensitivity assay: implications for clinical practice.
      as well as on data from previous chest pain cohort studies,
      • Keller T.
      • Zeller T.
      • Peetz D.
      • et al.
      Sensitive troponin I assay in early diagnosis of acute myocardial infarction.
      • Apple F.S.
      • Pearce L.A.
      • Smith S.W.
      • Kaczmarek J.M.
      • Murakami M.M.
      Role of monitoring changes in sensitive cardiac troponin I assay results for early diagnosis of myocardial infarction and prediction of risk of adverse events.
      a significant absolute change was defined as a rise or fall of at least 10 ng/L within 6 hours or 6 ng/L within 3 hours.

      Assumption of Linearity of Absolute Changes Within the First Hours

      The assumption of linearity of absolute changes within the first hours is based on unpublished internal data as well as recent data from Hammarsten et al
      • Hammarsten O.
      • Fu M.L.
      • Sigurjonsdottir R.
      • et al.
      Troponin T percentiles from a random population sample, emergency room patients and patients with myocardial infarction.
      showing a near-linear increase in levels of cTn, with increasing time from symptom onset in their non-ST-elevation myocardial infarction cohort.

      Measurement of hs-cTnI

      Blood samples for determination of hs-cTnI were collected at presentation to the emergency department and at 1 hour. After centrifugation, samples were frozen at −80°C until assayed in a blinded fashion in a core laboratory. The Abbott hs-cTnI assay used was the final precommercial release version of the ARCHITECT High Sensitive STAT Troponin I assay (Abbott Laboratories). Samples were thawed, mixed, and centrifuged (for 30 minutes at 3000 relative centrifugal force [RCF] and 4°C for serum samples or for 10 minutes, twice, at 3000 RCF for plasma samples) before analysis and according to manufacturer's instructions. The hs-cTnI assay has a 99th percentile concentration of 26.2 ng/L with a corresponding CV of < 5% and a limit of detection of 1.9 ng/L.
      • Koerbin G.
      • Tate J.
      • Potter J.M.
      • Cavanaugh J.
      • Glasgow N.
      • Hickman P.E.
      Characterisation of a highly sensitive troponin I assay and its application to a cardio-healthy population.
      Calculation of the glomerular filtration rate was performed using the abbreviated Modification of Diet in Renal disease formula.
      • Levey A.S.
      • Coresh J.
      • Greene T.
      • et al.
      Using standardized serum creatinine values in the modification of diet in renal disease study equation for estimating glomerular filtration rate.

      Supplemental Results

      Alternative Rule-out

      For the alternative rule-out of acute myocardial infarction in the derivation cohort, the optimal threshold was defined as a baseline < 2.8 ng/L and an absolute change within 1 hour of < 4.5 ng/L. With these values, 32% of patients could be classified as “rule-out” (Supplemental Figure 1). The sensitivity and the NPV for acute myocardial infarction in the “rule-out” zone were 100.0% (95% confidence interval [CI], 97.8-100.0) and 100.0% (95% CI, 98.7-100.0%) respectively.
      The optimal decision values derived in the derivation cohort were rounded to give whole values in ng/L for the prospective test in the validation cohort (rule out: baseline < 3 ng/L and an absolute change within 1 hour of < 5 ng/L). After applying the hs-cTnI alternative rule-out developed in the derivation cohort to the validation cohort, 33% of patients could be classified as “rule-out,” 19% as “rule-in,” and 48% as “observational zone” (Supplemental Figure 2). In the validation cohort the sensitivity and the NPV for acute myocardial infarction in the “rule-out” zone were 100.0% (95% CI, 97.8-100.0%) and 100.0% (95% CI, 98.8-100.0%), respectively.
      Supplemental TableBaseline Characteristics of Derivation and Validation Cohort
      CharacteristicDerivation Cohort

      n = 906
      Validation Cohort

      n = 905
      P-Value
      Age, y
       Median6262.66
       IQR49-7450-74
      Sex, n (%)
       Male620 (68)625 (69).77
      Risk factors, n (%)
       Hypertension545 (60)570 (63).22
       Hypercholesterolemia457 (50)455 (50).94
       Diabetes160 (18)170 (19).54
       Current or previous smoking559 (62)570 (63).57
       Family history310 (34)317 (35).72
      History, n (%)
       Coronary artery disease296 (33)331 (37).08
       Previous AMI201 (22)225 (25).18
       Previous revascularization231 (26)264 (29).08
       Peripheral artery disease50 (6)59 (7).37
       Previous stroke54 (6)47 (5).48
      eGFR, mL/min/1.73 m2
       Median8685.96
       IQR70-10269-105
      Final diagnoses, n (%)
       AMI162 (18)167 (19).75
       Unstable angina91 (10)87 (10).76
       Cardiac other122 (14)136 (15).56
       Noncardiac498 (55)466 (52).14
       Unknown32 (4)48 (5).07
      AMI = acute myocardial infarction; eGFR = estimated glomerular filtration rate; IQR = interquartile range.
      Figure thumbnail fx1
      Supplemental Figure 11h alternative algorithm for the diagnosis of AMI using hs-cTnI in the derivation cohort. 0h = hs-cTnI at presentation to the ED; AMI = acute myocardial infarction; Delta 1h = absolute change of hs-cTnI within 1 hour; NPV = negative predictive value; PPV = positive predictive value; Sens = sensitivity; Spec = specificity.
      Figure thumbnail fx2
      Supplemental Figure 21h alternative algorithm for the diagnosis of AMI using hs-cTnI in the validation cohort. 0h = hs-cTnI at presentation to the ED; AMI = acute myocardial infarction; Delta 1h = absolute change of hs-cTnI within 1 hour; NPV = negative predictive value; PPV = positive predictive value; Sens = sensitivity; Spec = specificity.

      References

        • Nawar E.W.
        • Niska R.W.
        • Xu J.
        National Hospital Ambulatory Medical Care Survey: 2005 emergency department summary.
        Adv Data. 2007; : 1-32
        • Anderson J.L.
        • Adams C.D.
        • Antman E.M.
        • et al.
        2012 ACCF/AHA focused update incorporated into the ACCF/AHA 2007 guidelines for the management of patients with unstable angina/non-ST-elevation myocardial infarction: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines.
        Circulation. 2013; 127: e663-e828
        • Hamm C.W.
        • Bassand J.P.
        • Agewall S.
        • et al.
        ESC Guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: The Task Force for the management of acute coronary syndromes (ACS) in patients presenting without persistent ST-segment elevation of the European Society of Cardiology (ESC).
        Eur Heart J. 2011; 32: 2999-3054
        • Kontos M.C.
        • Diercks D.B.
        • Kirk J.D.
        Emergency department and office-based evaluation of patients with chest pain.
        Mayo Clin Proc. 2010; 85: 284-299
        • Thygesen K.
        • Alpert J.S.
        • Jaffe A.S.
        • et al.
        Third universal definition of myocardial infarction.
        Circulation. 2012; 126: 2020-2035
        • Forberg J.L.
        • Henriksen L.S.
        • Edenbrandt L.
        • Ekelund U.
        Direct hospital costs of chest pain patients attending the emergency department: a retrospective study.
        BMC Emerg Med. 2006; 6: 6
        • Reichlin T.
        • Hochholzer W.
        • Bassetti S.
        • et al.
        Early diagnosis of myocardial infarction with sensitive cardiac troponin assays.
        N Engl J Med. 2009; 361: 858-867
        • Keller T.
        • Zeller T.
        • Ojeda F.
        • et al.
        Serial changes in highly sensitive troponin I assay and early diagnosis of myocardial infarction.
        JAMA. 2011; 306: 2684-2693
        • Keller T.
        • Zeller T.
        • Peetz D.
        • et al.
        Sensitive troponin I assay in early diagnosis of acute myocardial infarction.
        N Engl J Med. 2009; 361: 868-877
        • Thygesen K.
        • Mair J.
        • Giannitsis E.
        • et al.
        How to use high-sensitivity cardiac troponins in acute cardiac care.
        Eur Heart J. 2012; 33: 2252-2257
        • Safford M.M.
        • Parmar G.
        • Barasch C.S.
        • et al.
        Hospital laboratory reporting may be a barrier to detection of ‘microsize’ myocardial infarction in the US: an observational study.
        BMC Health Serv Res. 2013; 13: 162
        • Reichlin T.
        • Schindler C.
        • Drexler B.
        • et al.
        One-hour rule-out and rule-in of acute myocardial infarction using high-sensitivity cardiac troponin T.
        Arch Intern Med. 2012; 172: 1211-1218
        • Rubini Gimenez M.
        • Twerenbold R.
        • Reichlin T.
        • et al.
        Direct comparison of high-sensitivity-cardiac troponin I vs. T for the early diagnosis of acute myocardial infarction.
        Eur Heart J. 2014; 35: 2303-2311
        • Reiter M.
        • Twerenbold R.
        • Reichlin T.
        • et al.
        Early diagnosis of acute myocardial infarction in the elderly using more sensitive cardiac troponin assays.
        Eur Heart J. 2011; 32: 1379-1389
        • Reichlin T.
        • Irfan A.
        • Twerenbold R.
        • et al.
        Utility of absolute and relative changes in cardiac troponin concentrations in the early diagnosis of acute myocardial infarction.
        Circulation. 2011; 124: 136-145
        • Haaf P.
        • Drexler B.
        • Reichlin T.
        • et al.
        High-sensitivity cardiac troponin in the distinction of acute myocardial infarction from acute cardiac noncoronary artery disease.
        Circulation. 2012; 126: 31-40
        • Newby L.K.
        Myocardial infarction rule-out in the emergency department: are high-sensitivity troponins the answer?: comment on “One-hour rule-out and rule-in of acute myocardial infarction using high-sensitivity cardiac troponin T”.
        Arch Intern Med. 2012; 172: 1218-1219
        • Cullen L.
        • Mueller C.
        • Parsonage W.A.
        • et al.
        Validation of high-sensitivity troponin I in a 2-hour diagnostic strategy to assess 30-day outcomes in emergency department patients with possible acute coronary syndrome.
        J Am Coll Cardiol. 2013; 62: 1242-1249
        • Apple F.S.
        • Jesse R.L.
        • Newby L.K.
        • et al.
        National Academy of Clinical Biochemistry and IFCC Committee for Standardization of Markers of Cardiac Damage Laboratory Medicine Practice Guidelines: analytical issues for biochemical markers of acute coronary syndromes.
        Circulation. 2007; 115: e352-e355
        • Apple F.S.
        • Wu A.H.
        • Jaffe A.S.
        European Society of Cardiology and American College of Cardiology guidelines for redefinition of myocardial infarction: how to use existing assays clinically and for clinical trials.
        Am Heart J. 2002; 144: 981-986
        • Breiman L.
        Classification and Regression Trees.
        Wadsworth International Group, Belmont, CA1984
        • Fonarow G.C.
        • Adams Jr., K.F.
        • Abraham W.T.
        • et al.
        Risk stratification for in-hospital mortality in acutely decompensated heart failure: classification and regression tree analysis.
        JAMA. 2005; 293: 572-580
        • Redberg R.F.
        Coronary CT angiography for acute chest pain.
        N Engl J Med. 2012; 367: 375-376
        • Mueller C.
        Biomarkers and acute coronary syndromes: an update.
        Eur Heart J. 2014; 35: 552-556
        • Than M.
        • Aldous S.
        • Lord S.J.
        • et al.
        A 2-hour diagnostic protocol for possible cardiac chest pain in the emergency department: a randomized clinical trial.
        JAMA Intern Med. 2014; 174: 51-58
        • Than M.
        • Cullen L.
        • Aldous S.
        • et al.
        2-Hour accelerated diagnostic protocol to assess patients with chest pain symptoms using contemporary troponins as the only biomarker: the ADAPT trial.
        J Am Coll Cardiol. 2012; 59: 2091-2098
        • Balmelli C.
        • Meune C.
        • Twerenbold R.
        • et al.
        Comparison of the performances of cardiac troponins, including sensitive assays, and copeptin in the diagnostic of acute myocardial infarction and long-term prognosis between women and men.
        Am Heart J. 2013; 166: 30-37
        • Keller T.
        • Tzikas S.
        • Zeller T.
        • et al.
        Copeptin improves early diagnosis of acute myocardial infarction.
        J Am Coll Cardiol. 2010; 55: 2096-2106
        • Maisel A.
        • Mueller C.
        • Neath S.X.
        • et al.
        Copeptin helps in the early detection of patients with acute myocardial infarction: primary results of the CHOPIN trial (Copeptin Helps in the early detection Of Patients with acute myocardial INfarction).
        J Am Coll Cardiol. 2013; 62: 150-160
        • Möckel M.
        • Searle J.
        • Hamm C.
        • et al.
        Early discharge using single cardiac troponin and copeptin testing in patients with suspected acute coronary syndrome (ACS): a randomized, controlled clinical process study.
        Eur Heart J. 2015; 36: 369-376
        • Body R.
        • Carley S.
        • McDowell G.
        • et al.
        Rapid exclusion of acute myocardial infarction in patients with undetectable troponin using a high-sensitivity assay.
        J Am Coll Cardiol. 2011; 58: 1332-1339
        • Hoffmann U.
        • Truong Q.A.
        • Schoenfeld D.A.
        • et al.
        Coronary CT angiography versus standard evaluation in acute chest pain.
        N Engl J Med. 2012; 367: 299-308
        • Litt H.I.
        • Gatsonis C.
        • Snyder B.
        • et al.
        CT angiography for safe discharge of patients with possible acute coronary syndromes.
        N Engl J Med. 2012; 366: 1393-1403
        • Bandstein N.
        • Ljung R.
        • Johansson M.
        • Holzmann M.J.
        Undetectable high sensitivity cardiac troponin T level in the emergency department and risk of myocardial infarction.
        J Am Coll Cardiol. 2014; 63: 2569-2578
        • Rubini Giménez M.
        • Hoeller R.
        • Reichlin T.
        • et al.
        Rapid rule out of acute myocardial infarction using undetectable levels of high-sensitivity cardiac troponin.
        Int J Cardiol. 2013; 168: 3896-3901
        • Christ M.
        • Popp S.
        • Pohlmann H.
        • et al.
        Implementation of high sensitivity cardiac troponin T measurement in the emergency department.
        Am J Med. 2010; 123: 1134-1142
        • Giannitsis E.
        • Katus H.A.
        Antiplatelet therapy – ticagrelor.
        Hamostaseologie. 2012; 32: 177-185