If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Minneapolis Medical Research Foundation, MinnDepartment of Laboratory Medicine and Pathology, Hennepin County Medical Center and University of Minnesota, Minneapolis
Requests for reprints should be addressed to Fred S. Apple, PhD, Hennepin County Medical Center, Clinical Laboratories P4, 701 Park Avenue, Minneapolis, MN 55415.
Minneapolis Medical Research Foundation, MinnDepartment of Laboratory Medicine and Pathology, Hennepin County Medical Center and University of Minnesota, Minneapolis
This study examined the performance of single high-sensitivity cardiac troponin I (hs-cTnI) measurement strategies to rule out acute myocardial infarction.
Methods
This was a prospective, observational study of consecutive patients presenting to the emergency department (n = 1631) in whom cTnI measurements were obtained using an investigational hs-cTnI assay. The goals of the study were to determine 1) negative predictive value (NPV) and sensitivity for the diagnosis of acute myocardial infarction, type 1 myocardial infarction, and type 2 myocardial infarction; and 2) safety outcome of acute myocardial infarction or cardiac death at 30 days using hs-cTnI less than the limit of detection (LoD) (<1.9 ng/L) or the High-STEACS threshold (<5 ng/L) alone and in combination with normal electrocardiogram (ECG).
Results
Acute myocardial infarction occurred in 170 patients (10.4%), including 68 (4.2%) type 1 myocardial infarction and 102 (6.3%) type 2 myocardial infarction. For hs-cTnI<LoD (27%), the NPV and sensitivity for acute myocardial infarction were 99.6% (95% confidence interval 98.9%-100%) and 98.8 (97.2%-100%). For hs-cTnI<5 ng/L (50%), the NPV and sensitivity for acute myocardial infarction were 98.9% (98.2%-99.6%) and 94.7% (91.3%-98.1%). In combination with a normal ECG, 1) hs-cTnI<LoD had an NPV of 99.6% (98.9%-100%) and sensitivity of 99.4% (98.3%-100%); and 2) hs-cTnI<5 ng/L had an NPV of 99.5% (98.8%-100%) and sensitivity of 98.8% (97.2%-100%). The NPV and sensitivity for the safety outcome were excellent for hs-cTnI<LoD alone or in combination with a normal ECG, and for hs-cTnI<5 ng/L in combination with a normal ECG.
Conclusion
Strategies using a single hs-cTnI alone or in combination with a normal ECG allow the immediate identification of patients unlikely to have acute myocardial infarction and who are at very low risk for adverse events at 30 days.
Strategies using a single high-sensitivity cardiac troponin I measurement at presentation in combination with a normal electrocardiogram allow the immediate identification of patients unlikely to have acute myocardial infarction and who are at very low risk for adverse events at 30 days.
•
The implementation of these approaches may reduce overcrowding, facilitate early discharge in selected patients, expedite triaging, and reduce costs.
High-sensitivity (hs) cardiac troponin (cTn) I and T assays are analytically superior assays compared with contemporary cTn assays and are able to measure cTn at very low concentrations with excellent precision.
for the IFCC Task Force on Clinical Applications of Cardiac Bio-Markers Cardiac troponin assays: guide to understanding analytical characteristics and their impact on clinical care.
Both hs-cTnI and hs-cTnT assays are available and clinically used worldwide, with only the hs-cTnT assay recently cleared for use in the United States by the US Food and Drug Administration.
The ability to measure very low hs-cTn concentrations with clinically acceptable imprecision has allowed the development of new rule-out strategies, which have suggested that both acute myocardial infarction and/or myocardial injury can be safely excluded with a single measurement at presentation.
Two particular strategies have gained attention. The first strategy is based on the use of an assay's limit of detection (LoD), an analytical threshold below the 99th percentile.
The second strategy, the High-STEACS approach, consists of using a hs-cTnI concentration (assay dependent) threshold selected on the basis of a clinical need, rather than an analytical threshold.
This approach was derived and validated in the High-STEACS cohort study, in which a single hs-cTnI concentration <5 ng/L was shown to identify patients at very low risk for cardiac events.
Studies examining the use of single measurements to rule out acute myocardial infarction have primarily been performed outside the United States, in select cohorts of patients with chest pain, with the intent to exclude type 1 myocardial infarction.
No large study has tested and compared the rule-out of acute myocardial infarction, including type 1 and 2 myocardial infarction, using the 1) LoD and 2) High-STEACS approaches using an hs-cTnI assay in a US population. The goals of the present study were to examine the diagnostic performance of these two approaches for 1) ruling out acute myocardial infarction, including type 1 and 2 myocardial infarction; and 2) examine the safety outcomes for acute myocardial infarction or cardiac death at 30 days.
Methods
Study Design and Population
Following institutional review board approval, we prospectively included consecutive, unselected patients presenting from February 4, 2014 through May 9, 2014 in whom initial pre-set serial cTnI measurements at 0, 3, 6, and 9 hours were ordered on clinical indication at Hennepin County Medical Center (Minneapolis, MN) to rule in and rule out acute myocardial infarction (Use of TROPonin In Acute coronary syndromes [UTROPIA]; NCT02060760). For inclusion, patients needed a baseline cTnI measurement at presentation and at least one additional cTnI measured within 24 hours of presentation before discharge and at least one 12-lead electrocardiogram (ECG) performed. Exclusion criteria were age <18 years, ST-segment elevation myocardial infarction, pregnancy, trauma, declined to participate on research as documented on information disclosure, did not present through the emergency department, or were transferred from an outside hospital. For patients with more than one presentation during the study period, we included only the first.
Cardiac Troponin I Assays
Fresh ethylenediaminetetraacetic acid plasma samples were simultaneously measured with both the contemporary cTnI (clinically used) and hs-cTnI (investigational) assays on the ARCHITECT i1000SR or i2000SR analyzers (Abbott Diagnostics, Abbott Park, IL). Only the hs-cTnI assay data were used for the present study. Sex-specific 99th percentile upper reference limits (URL) for the hs-cTnI assay were 16 ng/L for females and 34 ng/L for males; % coefficients of variation were 5.3% at 15 ng/L and <20% at the LoD of 1.9 ng/L.
Diagnosis of type 1 and type 2 myocardial infarction using a high-sensitivity cardiac troponin I assay with sex-specific 99th percentiles based on the third universal definition of myocardial infarction classification system.
Incidence of undetectable, measurable, and increased cardiac troponin I concentrations above the 99th percentile using a high-sensitivity vs. a contemporary assay in patients presenting to the emergency department.
All cases with at least one hs-cTnI concentration >99th percentile were adjudicated according to the Third Universal Definition of Myocardial Infarction consensus recommendations by two clinicians after review of all available medical records, including 12-lead ECG, echocardiography, angiography, hs-cTnI values, and clinical presentation.
Cases with an adjudication discrepancy were reviewed and adjudicated by a third senior clinician.
To guide the adjudication of acute myocardial infarction in relation to the presence or absence of a significant rise and/or fall of cTnI, an algorithm was developed for the hs-cTnI assay on the basis of biological variation,
Diagnosis of type 1 and type 2 myocardial infarction using a high-sensitivity cardiac troponin I assay with sex-specific 99th percentiles based on the third universal definition of myocardial infarction classification system.
with the primary purpose of ensuring that changes within biological variation were not deemed abnormal. If the initial hs-cTnI value was below the sex-specific 99th percentile cutoff, then a rise of >69% and/or fall of >41% on serial sampling were used to suggest a significant dynamic rise and/or fall. Conversely, if the initial hs-cTnI value was above the 99th percentile, then a change of at least >20% was used.
For the diagnosis of acute myocardial infarction, a rise and/or fall with at least one value above the 99th percentile occurring in appropriate clinical circumstances consistent with acute myocardial ischemia was required, plus at least one additional myocardial infarction criteria: 1) ischemic symptoms, 2) development of pathologic Q waves in the 12-lead ECG, 3) ECG changes indicative of new ischemia, 4) imaging evidence of new loss of viable myocardium or new regional wall motion abnormality, or 5) identification of an intracoronary thrombus by angiography or autopsy.
Type 1 myocardial infarction was defined as myocardial infarction related to atherosclerotic plaque rupture, ulceration, fissuring, erosion, or dissection with resulting intraluminal thrombus.
Type 2 myocardial infarction was defined as myocardial infarction secondary to an ischemic imbalance between myocardial oxygen supply and/or demand not due to atherothrombosis.
For type 2 myocardial infarction to be adjudicated, cases were required to have a rise and/or fall of cTnI with at least 1 value above the 99th percentile plus at least 1 additional myocardial infarction criteria according to the Universal Definition of Myocardial Infarction, including the objective evidence or documentation of supply/demand imbalance.
The diagnostic outcomes examined were 1) acute myocardial infarction, 2) type 1 myocardial infarction, and 3) type 2 myocardial infarction during the index hospitalization. The safety outcome was a composite of acute myocardial infarction or cardiac death at 30 days, including events occurring during the index hospitalization.
Statistical Analyses
Categorical variables are shown as percentages. Continuous variables are shown as mean values ± standard deviation. The diagnostic and safety outcomes were examined for 1) LoD (<1.9 ng/L) and 2) High-STEACS (<5 ng/L) threshold based on a single hs-cTnI at presentation alone and in combination with a normal ECG. The ECGs were categorized as normal according to previously described criteria
(Supplementary Methods, available online). Diagnostic performance statistics were sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV); 95% confidence intervals (CIs) were ascertained using binomial proportions. Subgroup analyses were performed on early presenters, defined as individuals who had their first cTnI sample obtained ≤2 hours after symptom onset. All analysis was done using SAS version 9.4 (SAS Institute, Cary, NC).
Results
Baseline characteristics are shown in Table 1. Among the 1631 patients who met inclusion criteria, 444 patients (27%) had hs-cTnI<LoD at presentation. Using the High-STEACS threshold, 812 patients (50%) had hs-cTnI<5 ng/L at presentation. A total of 601 patients (37%) had a normal ECG. During the index hospitalization, acute myocardial infarction occurred in 170 patients (10.4%), including 68 (4.2%) type 1 and 102 (6.3%) type 2 myocardial infarctions.
Table 1Baseline Characteristics
Characteristic
Value
Study cohort (n)
1631
Age (y), mean (SD)
57 (15)
Female gender
720 (44)
Hypertension
1074 (66)
Diabetes mellitus
496 (43)
Dyslipidemia
696 (43)
Coronary artery disease
371 (23)
Prior myocardial infarction
190 (12)
Prior percutaneous coronary intervention
150 (9)
Prior coronary artery bypass graft
73 (4)
Congestive heart failure
231 (14)
Atrial fibrillation
129 (8)
Peripheral vascular disease
42 (3)
Cerebrovascular disease
153 (9)
Renal insufficiency, nondialysis
161 (10)
End-stage renal disease on hemodialysis
80 (5)
History of tobacco use
969 (59)
Chest discomfort
835 (51)
Dyspnea
680 (42)
Arm and/or shoulder discomfort
250 (15)
Jaw and/or neck discomfort
98 (6)
Epigastric discomfort
93 (6)
Nausea and/or vomiting
381 (23)
Fatigue
444 (27)
Baseline hs-cTnI concentrations <1.9 ng/L
444 (27)
Baseline hs-cTnI concentrations <5 ng/L
812 (50)
Normal 12-lead ECG
601 (37)
Values are number (percentage) unless otherwise noted.
Rule-Out Using the LoD Alone and in Combination with a Normal ECG
In patients with hs-cTnI<LoD at presentation (27% of patients), independent of ECG findings, the NPV and sensitivity for acute myocardial infarction were 99.6% (95% CI, 98.9%-100%) and 98.8% (95% CI, 97.2%-100%), respectively (Table 2). Using hs-cTnI<LoD alone, 2 of 170 patients with acute myocardial infarction were missed, corresponding to a miss rate of 1.2% (or 2 of 444 patients with an hs-cTnI<LoD, 0.5%). In comparison with hs-cTnI<LoD alone, the addition of a normal ECG (16% of patients) offered a NPV of 99.6% (95% CI, 98.9%-100%) and sensitivity of 99.4% (95% CI, 98.3%-100%) for acute myocardial infarction. Using hs-cTnI<LoD with a normal ECG, only 1 of 170 patients with acute myocardial infarction was missed, corresponding to a miss rate of 0.6% (or 1 of 254 patients with an hs-cTnI<LoD and a normal ECG, 0.4%).
Table 2Use of a Single hs-cTnI at Presentation Alone and in Combination with a Normal 12-lead ECG for the Diagnosis of Acute Myocardial Infarction (Type 1 and 2 Myocardial Infarction), Type 1 Myocardial Infarction Alone, and Type 2 Myocardial Infarction Alone
Parameter
LoD (1.9 ng/L)
High-STEACS (<5 ng/L)
Baseline hs-cTnI<LoD
Baseline hs-cTnI<LoD and Normal ECG
Baseline hs-cTnI<5 ng/L
Baseline hs-cTnI<5 ng/L and Normal ECG
Acute myocardial infarction
Proportion qualifying
444/1631 (27)
254/1631 (16)
812/1631 (50)
406/1631 (25)
Proportion of missed MIs
2/170 (1.2)
1/170 (0.6)
9/170 (5.3)
2/170 (1.2)
NPV
99.6 (98.9-100)
99.6 (98.8-100)
98.9 (98.2-99.6)
99.5 (98.8-100)
Sensitivity
98.8 (97.2-100)
99.4 (98.3-100)
94.7 (91.3-98.1)
98.8 (97.2-100)
PPV
14.2 (12.2-16.1)
12.3 (10.5-14.0)
19.7 (16.9-22.4)
13.7 (11.8-15.6)
Specificity
30.3 (27.9-32.6)
17.3 (15.4-19.3)
55.0 (52.4-57.5)
27.7 (25.4-30.0)
Type 1 myocardial infarction
Proportion qualifying
443/1529 (29)
254/1529 (17)
807/1529 (53)
405/1529 (27)
Proportion of missed MIs
1/68 (1.5)
1/68 (1.5)
4/68 (5.9)
1/68 (1.5)
NPV
99.8 (99.3-100)
99.6 (98.8-100)
99.5 (99.0-100)
99.8 (99.3-100)
Sensitivity
98.5 (95.7-100)
98.5 (95.7-100)
94.1 (88.5-99.7)
98.5 (95.7-100)
PPV
6.2 (4.7-7.6)
5.3 (4.0-6.5)
8.9 (6.8-10.9)
6.0 (4.6-7.3)
Specificity
30.3 (27.9-32.6)
17.3 (15.4-19.3)
5.5 (5.2-5.8)
27.7 (25.4-30.0)
Type 2 myocardial infarction
Proportion qualifying
443/1563 (28)
253/1563 (16)
808/1563 (52)
405/1563 (26)
Proportion of missed MIs
1/102 (0.98)
0/102 (0)
5/102 (4.9)
1/102 (0.98)
NPV
99.8 (99.3-100)
100 (100-100)
99.4 (98.8-99.9)
99.8 (99.3-100)
Sensitivity
99.0 (97.1-100)
100 (100-100)
95.1 (90.9-99.3)
99.0 (97.1-100)
PPV
9.0 (7.3-10.7)
7.8 (6.3-9.2)
12.9 (10.5-15.2)
8.7 (7.1-10.4)
Specificity
30.3 (27.8-32.6)
17.3 (15.4-19.3)
55.0 (52.4-57.5)
27.7 (25.4-30.0)
Values are number (percentage) or percentage (95% confidence interval).
At 30 days, the NPV and sensitivity for acute myocardial infarction or cardiac death were 99.6% (95% CI, 98.9%-100%) and 98.8% (95% CI, 97.2%-100%) for hs-cTnI<LoD alone, and 99.6% (95% CI, 98.8%-100%) and 99.4% (95% CI, 98.3%-100%) for hs-cTnI<LoD with a normal ECG (Figure, Table 3). Using hs-cTnI<LoD alone, 2 of 171 events (1.2%) were missed (or 2 of 444 patients with an hs-cTnI<LoD, 0.5%); whereas using hs-cTnI<LoD with a normal ECG only 1 of 171 events (0.6%) was missed (or 1 of 254 patients with a hs-cTnI<LoD and a normal ECG, 0.4%).
FigureSafety outcome: risk stratification at 30 days for acute myocardial infarction and cardiac death. Columns for proportion of patients qualifying for each approach (limit of detection, <1.9 ng/L, and High-STEACS, <5 ng/L) with and without a normal result on electrocardiogram (ECG), and corresponding sensitivities for the safety outcome.
Table 3Safety Outcome: Risk Stratification at 30 Days for Acute Myocardial Infarction and Cardiac Death (Including Events During Index Hospitalization) Using a Single hs-cTnI at Presentation Alone and in Combination with a Normal 12-Lead ECG
Parameter
LoD (1.9 ng/L)
High-STEACS (<5 ng/L)
Baseline hs-cTnI<LoD
Baseline hs-cTnI<LoD and Normal ECG
Baseline hs-cTnI<5 ng/L
Baseline hs-cTnI<5 ng/L and Normal ECG
Acute myocardial infarction
Proportion of missed events
2/171 (1.2)
1/171 (0.6)
9/171 (5.3)
2/171 (1.2)
NPV
99.6 (98.9-100)
99.6 (98.8-100)
98.9 (98.2-99.6)
99.5 (98.8-100)
Sensitivity
98.8 (97.2-100)
99.4 (98.3-100)
94.7 (91.4-98.1)
98.8 (97.2-100)
Type 1 myocardial infarction
Proportion of missed events
1/70 (1.4)
1/70 (1.4)
4/70 (5.7)
1/70 (1.4)
NPV
99.8 (99.3-100)
99.6 (98.8-100)
99.5 (99.0-100)
99.8 (99.3-100)
Sensitivity
98.6 (95.8-100)
98.6 (95.8-100)
94.3 (88.9-99.7)
98.6 (95.8-100)
Type 2 myocardial infarction
Proportion of missed events
1/108 (0.9)
0/108 (0)
5/103 (4.6)
1/108 (0.9)
NPV
99.8 (99.3-100)
100 (100-100)
99.4 (98.8-99.9)
99.8 (99.3-100)
Sensitivity
99.1 (97.3-100)
100 (100-100)
95.4 (91.4-99.3)
99.1 (97.3-100)
Values are number (percentage) or percentage (95% confidence interval).
For ruling out type 1 myocardial infarction alone, baseline hs-cTnI<LoD alone resulted in a NPV of 99.8% (95% CI, 99.3%-100%) and sensitivity of 98.5% (95% CI, 95.7%-100%). In combination with a normal ECG, hs-cTnI<LoD resulted in an NPV of 99.6% (95% CI, 98.8%-100%) and sensitivity of 98.5% (95% CI, 95.7%-100%). For type 1 myocardial infarction, the sensitivity for the safety outcome was 98.6% (95% CI, 95.8%-100%) using either hs-cTnI<LoD alone or in combination with a normal ECG.
For ruling out type 2 myocardial infarction alone, baseline hs-cTnI<LoD alone resulted in an NPV of 99.8 (95% CI, 99.3%-100%) and sensitivity of 99.0% (95% CI, 97.1%-100%). In combination with a normal ECG, hs-cTnI<LoD resulted in an NPV and sensitivity of 100% (95% CI, 100%-100%). For type 2 myocardial infarction, the sensitivity for the safety outcome was 99.1% (95% CI, 97.3%-100%) using hs-cTnI<LoD alone and 100% (95% CI, 100%-100%) in combination with a normal ECG.
In early presenters the NPV and sensitivity for the diagnostic and safety outcomes was 100% (95% CI, 100%-100%) using hs-cTnI<LoD alone or in combination with a normal ECG (Table 4).
Table 4Use of a Single hs-cTnI at Presentation Alone and in Combination with a Normal 12-Lead ECG for 1) Diagnosis of Acute Myocardial Infarction and 2) 30-Day Risk Stratification for Acute Myocardial Infarction or Cardiac Death in Early Presenters
Early Presenters (n = 262)
LoD (1.9 ng/L)
High-STEACS (<5 ng/L)
Baseline hs-cTnI<LoD
Baseline hs-cTnI<LoD and Normal ECG
Baseline hs-cTnI<5 ng/L
Baseline hs-cTnI<5 ng/L and Normal ECG
Diagnostic outcome, acute myocardial infarction
Proportion qualifying
78/262 (30)
41/262 (16)
137/262 (52)
63/262 (24)
Proportion of missed MIs
0/39 (0)
0/39 (0)
2/39 (5.1)
0/39 (0)
NPV
100 (100-100)
100 (100-100)
98.5 (96.5-100)
100 (100-100)
Sensitivity
100 (100-100)
100 (100-100)
94.9 (88.0-100)
100 (100-100)
PPV
21.2 (15.3-27.1)
17.7 (12.6-22.7)
29.6 (21.6-37.6)
19.6 (14.1-25.1)
Specificity
35.0 (28.7-41.2)
18.4 (13.3-23.5)
60.5 (54.1-67.0)
28.3 (22.3-34.2)
Safety outcome, 30-day acute myocardial infarction or cardiac death
Proportion of missed events
0/40 (0)
0/40 (0)
2/40 (5)
0/40 (0)
NPV
100 (100-100)
100 (100-100)
98.5 (96.5-100)
100 (100-100)
Sensitivity
100 (100-100)
100 (100-100)
95.0 (88.3-100)
100 (100-100)
Values are number (percentage) or percentage (95% confidence interval).
Rule-Out Using the High-STEACS Threshold Alone and in Combination with a Normal ECG
In patients with hs-cTnI<5 ng/L at presentation (50% of patients), independent of ECG findings, the NPV and sensitivity for acute myocardial infarction were 98.9% (95% CI, 98.2%-99.6%) and 94.7% (95% CI, 91.3%-98.1%), respectively. Using hs-cTnI<5 ng/L alone, 9 of 170 patients with acute myocardial infarction were missed, corresponding to a miss rate of 5.3% (or 9 of 812 patients with hs-cTnI<5 ng/L, 1.1%). The addition of a normal ECG to a hs-cTnI<5 ng/L (25% of patients) showed a NPV of 99.5% (95% CI, 98.8%-100%) and a sensitivity of 98.8% (95% CI, 97.2%-100%) for acute myocardial infarction (Table 2). Using hs-cTnI<5 ng/L with a normal ECG, 2 of 170 patients with acute myocardial infarction were missed, corresponding to a miss rate of 1.2% (or 2 of 406 patients with hs-cTnI<5 ng/L and a normal ECG, 0.5%).
At 30 days, the NPV and sensitivity for acute myocardial infarction or cardiac death was 98.9% (95% CI, 98.2%-99.6%) and 94.7% (95% CI, 91.4%-98.1%) for hs-cTnI<5 ng/L alone, and 99.5% (95% CI, 98.8%-100%) and 98.8% (95% CI, 97.2%-100%) for hs-cTnI<5 ng/L with a normal ECG (Table 3). Using hs-cTnI<5 ng/L alone, 9 of 171 events (5.3%) were missed (or 9 of 812 patients with hs-cTnI<LoD, 1.1%), whereas when using hs-cTnI<5 ng/L with a normal ECG, only 2 of 171 (1.2%) were missed (or 2 of 406 patients with hs-cTnI<5 ng/L and a normal ECG, 0.5%).
For ruling out type 1 myocardial infarction alone, baseline hs-cTnI<5 ng/L alone resulted in an NPV of 99.5% (95% CI, 99.0%-100%) and sensitivity of 94.1% (95% CI, 88.5%-99.7%). In combination with a normal ECG, hs-cTnI<5 ng/L resulted in a NPV of 99.8% (95% CI, 99.3%-100%) and sensitivity of 98.5% (95% CI, 95.7%-100%). For type 1 myocardial infarction, the sensitivities for the safety outcome were 94.3% (95% CI, 88.9%-99.7%) using hs-cTnI<5 ng/L alone and 98.6% (95% CI, 95.8%-100%) in combination with a normal ECG.
For ruling out type 2 myocardial infarction alone, baseline hs-cTnI<5 ng/L alone resulted in a NPV of 99.4% (95% CI, 98.8%-99.9%) and sensitivity of 95.1% (95% CI, 90.9%-99.3%). In combination with a normal ECG, hs-cTnI<5 ng/L alone resulted in a NPV of 99.8% (95% CI, 99.3%-100%) and sensitivity of 99.0% (95% CI, 97.1%-100%). For type 2 myocardial infarction, the sensitivities for the safety outcome were 95.4% (95% CI, 91.4%-99.3%) using hs-cTnI<5 ng/L alone and 99.1% (95% CI, 97.3%-100%) in combination with a normal ECG.
In early presenters, hs-cTnI<5 ng/L alone resulted in an NPV of 98.5% (95% CI, 96.5%-100%) and sensitivity of 94.9% (95% CI, 88.0%-100%) for acute myocardial infarction, whereas in combination with a normal ECG, hs-cTnI<5 ng/L had a NPV and sensitivity of 100% (95% CI, 100%-100%) (Table 4). Similarly, hs-cTnI<5 ng/L alone had an NPV 98.5% (95% CI, 96.5%-100%) and sensitivity of 95.0% (95% CI, 88.3%-100%) for the safety outcome of acute myocardial infarction or cardiac death at 30 days, whereas in combination with a normal ECG, both the NPV and sensitivity were 100% (95% CI, 100%-100%).
Discussion
Several findings are unique to our study evaluating the LoD and High-STEACS threshold rule-out strategies using a single hs-cTnI at presentation, alone and in combination with a normal ECG. First, we demonstrate that both strategies are excellent in safely ruling out acute myocardial infarction when used in combination with a normal ECG, as demonstrated by the very high NPV and sensitivity achieved for both the diagnostic and safety outcomes, including an excellent performance in early presenters. The use of these strategies allows the immediate identification of patients in whom the clinical presentation is unlikely to be due to an acute myocardial infarction (type 1 and 2 myocardial infarction) and who are at very low risk for adverse events at 30 days. The implementation of these approaches may reduce overcrowding, facilitate early discharge in selected patients, expedite triaging and reduce costs.
Second, our study provides novel insights into the performance of single measurement rule-out strategies across myocardial infarction subtypes, including both type 1 and 2 myocardial infarctions. Our findings suggest that both the LoD and High-STEACS approaches in combination with a normal ECG are excellent in safely ruling out both type 1 and 2 myocardial infarction. Our study uniquely demonstrates that these rule-out approaches have an excellent clinical performance in a heterogeneous, all-comers cohort of patients undergoing hs-cTnI measurements on clinical indication, regardless of the presence or absence of chest pain, reflective of US practice. In contrast, most studies assessing rule-out strategies (outside the United States) often based their findings on select cohorts of patients with chest pain (5-7, 12) without providing detailed insights as to whether the rule-out strategies are applicable across the spectrum of patients with acute myocardial infarction, including both type 1 and 2 myocardial infarction.
Third, among patients with a normal ECG, the High-STEACS approach seems more efficient because it applies to a larger proportion of patients than the LoD. In our study, we demonstrate that a single hs-cTnI<LoD, regardless of ECG findings, offers an excellent NPV and sensitivity for both the diagnostic and safety outcome, an approach applying to a similar proportion of patients than the one seen when combining a baseline hs-cTnI<5 ng/L and a normal ECG. The highest proportion of patients qualifying for rule-out was seen with the High-STEACS approach using hs-cTnI alone, which applied to 50% of patients. However, although the High-STEACS approach using hs-cTnI alone offered a very high NPV, the achieved sensitivity (approximately 95%) for both the diagnostic and safety outcome may not meet the desired acceptable event miss rate (approximately ≤1% miss rate or ≥99% sensitivity),
What is an acceptable risk of major adverse cardiac event in chest pain patients soon after discharge from the Emergency Department? A clinical survey.
However, when combined with a normal ECG, the High-STEACS approach offered an excellent NPV and sensitivity for the diagnostic and safety outcomes.
Prior non-US studies examining single measurement rule-out strategies have mostly examined hs-cTnT, with few non-US studies assessing the Abbott hs-cTnI assay. Similar to our findings, the Advantageous Predictors of Acute Coronary Syndrome Evaluation (APACE) investigators (Switzerland) also examined the Abbott hs-cTnI assay using the <1.9 ng/L threshold for their calculations in 1567 patients and reported both a sensitivity and NPV of 100%.
reported a sensitivity of 94.5% and an NPV of 99.2%. Whether differences in the ECG adjudication alone explain the difference in the achieved sensitivity is uncertain. These observations highlight that other factors not related to hs-cTnI concentrations alone may influence the diagnostic performance.
Last, the present study complements our recent work using the LoD to rule out acute myocardial injury. An hs-cTnI<LoD demonstrates excellent a) sensitivity and NPV for ruling out acute myocardial injury
and for ruling out type 1 and type 2 myocardial infarction, and b) risk stratification at 30 days for acute myocardial infarction or cardiac death. We note that our findings are limited to one hs-cTnI assay (Abbott Diagnostics) and emphasize that independent studies need to be carried out for other hs-cTn assays.
for the IFCC Task Force on Clinical Applications of Cardiac Bio-Markers Cardiac troponin assays: guide to understanding analytical characteristics and their impact on clinical care.
Single measurement rule-out strategies using very low hs-cTnI concentrations such as the LoD and the High-STEACS approaches are excellent in safely ruling out acute myocardial infarction, including type 1 and 2, particularly when combined with a normal ECG. Both rule-out strategies quickly identify patients at low risk for acute myocardial infarction or cardiac death at 30 days, representing a potential opportunity to improve care and reduce costs.
Supplementary Methods
A normal 12-lead electrocardiogram (ECG) was defined as an entirely normal ECG (including those with normal variation ST elevation) or where there were nondiagnostic ST-T wave abnormalities. Sinus bradycardia, prolonged PR interval, low voltage, right or left atrial hypertrophy, right ventricular conduction delay, and occasional premature atrial beats were all within normal for the purpose of this study. All ECGs with atrial fibrillation, sinus tachycardia, high-degree atrioventricular block, premature ventricular contractions, bundle branch block, intraventricular conduction delay (>120 ms), paced rhythm, left ventricular hypertrophy, pathologic Q waves, ST segment depression ≥0.05 mV in 2 contiguous leads, T wave inversion (≥0.15 mV in 2 contiguous leads with prominent R wave or R/S ratio >1), or ST elevation were considered abnormal. Nonspecific ST-T wave abnormalities were slight variations in ST or T that were <1.5 mm of abnormal T wave inversion in 2 consecutive leads or up to 0.5 mm ST depression in 2 consecutive leads or both, or T wave flattening.
References
Apple F.S.
Sandoval Y.
Jaffe A.S.
Ordonez-Llanos J.
for the IFCC Task Force on Clinical Applications of Cardiac Bio-Markers
Cardiac troponin assays: guide to understanding analytical characteristics and their impact on clinical care.
Diagnosis of type 1 and type 2 myocardial infarction using a high-sensitivity cardiac troponin I assay with sex-specific 99th percentiles based on the third universal definition of myocardial infarction classification system.
Incidence of undetectable, measurable, and increased cardiac troponin I concentrations above the 99th percentile using a high-sensitivity vs. a contemporary assay in patients presenting to the emergency department.
What is an acceptable risk of major adverse cardiac event in chest pain patients soon after discharge from the Emergency Department? A clinical survey.
Funding: The UTROPIA study (NCT02060760) is partially funded through a grant from 1) Abbott Diagnostics, who had no role in the design and conduction of the study; including data collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the final manuscript; and 2) the Minneapolis Medical Research Foundation.
Conflict of Interest: SWS is a consultant for Alere and advisor for Roche Clinical Diagnostics. SAL is a research principal investigator through the Minneapolis Medical Research Foundation (MMRF), not salaried, for Biokit, Hytest Ltd, and Instrumentation Laboratory; and is on the editorial board of the Journal of Applied Laboratory Medicine. FSA is a consultant for Philips Healthcare Incubator and Metanomics Healthcare; is on the Board of Directors for HyTest Ltd; has received honoraria from Instrumentation Laboratory and Abbott POC; is a research principal investigator through the MMRF, not salaried, for Abbott Diagnostics, Roche Diagnostics, Siemens Healthcare, Alere, Ortho-Clinical Diagnostics, Nanomix, Becton Dickinson, and Singulex; and is Associate Editor for Clinical Chemistry.
Authorship: All authors had access to the data and a role in writing the manuscript.
30257-7&id=gr1.jpg){kind=link}