Skip to main content
Skip main navigationClose Drawer MenuOpen Drawer Menu

MAINTENANCE ALERT: Our website management system will undergo essential maintenance on Monday and Tuesday, May 27-28. All articles and features will remain accessible during this period. Thank you for your patience as we work to enhance your user experience!

Implementation of High-Sensitivity Cardiac Troponin Assays in the United StatesFree Access

Original Investigation

J Am Coll Cardiol, 81 (3) 207–219
Sections

Central Illustration

Abstract

Background

Few data exist regarding the implementation of high-sensitivity cardiac troponin (hs-cTn) assays in the United States since their approval.

Objectives

This study sought to explore trends in hs-cTn assay implementation over time and assess the association of their use with in-hospital cardiac testing and outcomes.

Methods

The study examined trends in implementation of hs-cTn assays among participating hospitals in the National Cardiovascular Data Registry Chest Pain-MI [Myocardial Infarction] Registry from January 1, 2019 through September 30, 2021. Associations among hs-cTn use, use of in-hospital diagnostic imaging, and patient outcomes were assessed using generalized estimating equation models with logistic or gamma distributions.

Results

Among 550 participating hospitals (N = 251,000), implementation of hs-cTn assays increased from 3.3% in the first quarter of 2019 to 32.6% in the third quarter of 2021 (Ptrend < 0.001). Use of hs-cTn was associated with more echocardiography among persons with non–ST-segment elevation acute coronary syndrome (NSTE-ACS; 82.4% vs 75.0%; adjusted odds ratio: 1.43; 95% CI: 1.19-1.73) but not among low-risk chest pain individuals. Use of hs-cTn was associated with less invasive coronary angiography among low-risk patients (3.7% vs 4.5%; adjusted odds ratio: 0.73; 95% CI: 0.58-0.92) but similar use for patients with NSTE-ACS. There was no association between hs-cTn use and noninvasive stress testing or coronary computed tomography angiography testing. Among individuals with NSTE-ACS, hs-cTn use was not associated with revascularization or in-hospital mortality. Use of hs-cTn was associated with a shorter length of stay (median 47.6 hours vs 48.0 hours; ratio: 0.94; 95% CI: 0.90-0.98).

Conclusions

Implementation of hs-cTn among U.S. hospitals is increasing, but most U.S. hospitals continue to use less sensitive assays. The use of hs-cTn was associated with modestly shorter length of stay, greater use of echocardiography for NSTE-ACS, and less use of invasive angiography among low-risk patients.

Introduction

The role of biomarker testing for the evaluation and management of patients with suspected acute coronary syndrome (ACS) has evolved. In the early 1990s, cardiac troponin (cTn) subtypes (I and T) demonstrated improved sensitivity and efficiency to diagnose myocardial infarction (MI) compared with creatinine kinase-MB, thus leading to adoption of cTn as the preferred biomarker for evaluation of chest pain and diagnosis of MI.1-3 Additional refinements to analytical technology in the late 2000s led to more sensitive assays, further improving diagnostic sensitivity and accuracy for MI.4-6

Given their superiority to less sensitive conventional cTn assays, high-sensitivity cTn (hs-cTn) was recommended as the preferred biomarker by the Universal Definition of MI Task Force for the diagnosis of MI.7 Following regulatory approvals, hs-cTn assays were first implemented in clinical practice in many international regions beginning in 2010. Clinical practice guidelines in these regions quickly supported use of hs-cTn assays in preference to conventional cTn assays.8,9 Capitalizing on the enhanced sensitivity and precision of these assays, clinical decision pathways were developed to facilitate more efficient rule-out of MI with high negative predictive value for adverse cardiac events.10-12 In European cohorts, implementation of hs-cTn in clinical practice was associated with reductions in length of stay (LOS), less stress testing, and similar use of invasive coronary angiography.13

In contrast to Europe, regulatory approval of hs-cTn assays in the United States by the U.S. Food and Drug Administration did not occur until 2017. Since then, several more hs-cTn assays have been approved. Currently, it is unknown how many hospitals in the United States have transitioned from conventional cTn assays to hs-cTn assays. Furthermore, although international observational studies and randomized controlled trials demonstrate that hs-cTn may improve the efficiency of chest pain evaluation while maintaining safety, it is unclear whether these benefits translate to clinical practice in the United States, where population characteristics, hospital characteristics, and practice patterns regarding testing differ considerably, including a range of sizes, nonprofit status, teaching practices, and geographic locations.14-17 This study explored: 1) the prevalence of hs-cTn assay use among participating institutions in the National Cardiovascular Data Registry (NCDR) Chest Pain-MI Registry; and 2) the association of hs-cTn assay use with cardiovascular testing and outcomes among patients included in the registry.

Methods

Data source

The NCDR Chest Pain-MI Registry (previously known as the Acute Coronary Treatment and Intervention Network Registry-Get With The Guidelines) is an ongoing, nationwide, voluntary, quality improvement registry sponsored by the American College of Cardiology (ACC).18 The registry includes individuals aged 18 years or older with type 1 MI (both ST-segment elevation MI [STEMI] and non-ST-segment elevation MI [NSTEMI]), unstable angina, or low-risk chest pain. The registry excludes: 1) patients transferred from an outside facility with a transfer time of >24 hours; 2) patients with type 2 MI; and 3) patients arriving for scheduled procedures unless an in-hospital STEMI occurs. Patients with MI must meet the Universal Definition of MI criteria.7 Unstable angina patients are given a clinical diagnosis and must not meet the definition of STEMI or NSTEMI. A low-risk chest pain diagnosis is assigned to patients presenting with chest pain who may represent myocardial ischemia, who have a minimum of 1 electrocardiogram and cTn measurement, but who do not meet the definition of unstable angina, NSTEMI, or STEMI. The entry of low-risk chest pain and unstable angina patients into the registry is optional for participating hospitals and, if included, may comprise either all eligible patients or a sample. If a sample is input, a minimum of 10 consecutive unstable angina patients and/or 30 consecutive low-risk chest pain patients must be entered per month. Waiver of written informed consent and authorization for this NCDR study was granted by the Advarra Institutional Review Board. The NCDR uses a data quality program that incorporates data abstraction training, quality feedback reports, internal quality assurance protocols, and a data audit program.19 A 2019 audit of the registry revealed a 90% mean agreement, 95% interrater reliability, and 94% registry completeness.19 The Duke Clinical Research Institute served as the data analytic center for this analysis to aggregate deidentified data for research purposes.

Hospital implementation of high-sensitivity cardiac troponin assays

We identified 806 hospitals with available data on cTn assay use that participated in the NCDR Chest Pain-MI Registry between January 1, 2019 and September 30, 2021 (Supplemental Figure 1). Hospitals that did not enter data (at least 1 patient) in the registry in consecutive quarters over our study period were excluded, as were hospitals that had fewer than 40 patients in the registry annually. This resulted in a final cohort of 550 hospitals for analysis.

Implementation of hs-cTn at each hospital was defined when ≥25% of patients in the Chest Pain-MI registry at that facility were evaluated with an hs-cTn assay during a quarter. This prespecified, ≥25% cutoff was chosen considering that hospitals may not make an immediate uniform transition to hs-cTn, that non–hs-cTn assays could still be used within a hospital by clinicians in certain instances, and that patients transferred from outside institutions may have been evaluated by other assays that appear in the electronic medical record. The hs-cTn assays included were the Roche Elecsys Troponin T Gen 5 STAT hs-cTnT assay, the Abbott ARCHITECT STAT hs-cTnI assay, the Beckman Coulter Access hs-cTnI assay, and the Siemens Atellica IM, Dimension Vista, Dimension EXL, and ADVIA Centaur hs-cTnI assays. The proportion of hospitals that implemented hs-cTn assays in each quarter was calculated and described. For the final quarter of the study period (July 2021 to Sept 2021), the distribution of the most commonly used hs-cTn assays across hospitals was determined. Similarly, for the final quarter of the study, characteristics of hospitals that had or had not implemented hs-cTn assays were presented, including region, community description, membership in the Council of Teaching Hospitals, hospital bed size, coronary angiography and cardiac surgery capabilities, nonprofit status, annual MI volumes (STEMI and NSTEMI combined), and overall patient characteristics of those evaluated at participating hospitals.

High-sensitivity cardiac troponin assay use, patient testing, and outcomes

There were a total of 598,327 patients from 806 hospitals participating in the NCDR Chest Pain-MI registry between January 1, 2019 and September 30, 2021 (Supplemental Figure 2). After sequentially excluding patients without available cTn data during the study period, patients in the hospitals who did not have data in every quarter or hospitals with <40 eligible patients annually, patients who were transferred from outside institutions, STEMI patients, patients with short data collection forms, patients evaluated by both conventional cTn assays and hs-cTn assays during their index admission, and patients with nonindex admission within a single site, the final study cohort for our patient-level analyses consisted of 251,000 patients. Baseline demographics, past medical history, and discharge medications were presented.

Outcomes of interest for this analysis at the patient level were hospital length of stay (LOS), in-hospital mortality, and in-hospital cardiovascular diagnostic testing, including invasive coronary angiography, echocardiography, cardiac magnetic resonance (CMR), a composite of any stress testing (exercise or pharmacologic) or computed tomography (CT) coronary angiography, and coronary revascularization with percutaneous coronary intervention (PCI) or coronary artery bypass grafting (CABG).

Statistical analysis

First, the hospital-level percentage of patients who used hs-cTn was calculated in each quarter during the study period. Hospitals with at least 25% of patients evaluated with hs-cTn in each quarter were defined as hospitals having implemented hs-cTn in that quarter. A sensitivity analysis, using implementation thresholds of at least 50% and 75% of patients evaluated with hs-cTn, was also performed. To assess the trend in the proportion of hospitals that implemented hs-cTn from quarter (Q) 1, 2019 to Q3, 2021, the Cochran-Armitage trend test was used. Characteristics of hospitals that had or had not implemented hs-cTn in Q3, 2021 were compared using the Wilcoxon rank sum test for continuous variables, and the chi-square test for categorical variables.

Next, we compared the patients who were or were not evaluated by hs-cTn assays. The baseline characteristics and past medical history of patients were summarized and presented using median and IQR (presented with first quartile, third quartile) for continuous variables and frequencies (and percentages) for categorical variables. The Wilcoxon rank sum test was used to compare continuous variables, and the chi-square test was used to compare categorical variables between the 2 patient groups.

Finally, we compared in-hospital testing, procedures, in-hospital-mortality, and LOS in hours at the patient level between patients who were or were not evaluated by hs-cTn assays. For binary outcomes, including in-hospital mortality, invasive coronary angiography, PCI, CABG, echocardiography, CMR, and a composite of any stress testing or CT coronary angiography, multivariable models were used to assess the association between hs-cTn use and each outcome, respectively. Generalized estimating equations (GEE) logistic regression models with adjustment for clustering (ie, statistical dependence) of observations from the same hospital were used to compare binary outcomes between the 2 groups of patients after adjusting for case mix. The GEE method was implemented with a compound symmetrical working correlation matrix and empirical (sandwich) SE estimates. The independent (exposure) variable in the multivariable models was whether the patient was evaluated by an hs-cTn assay or not during the admission. The association of the variables of interest with the binary endpoints are presented as odds ratios (OR; effect size) and 95% CIs. Because there were few PCI, CABG, and deaths among low-risk chest pain patients, logistic models were not applied to this group. A GEE model with a gamma distribution and log-link function was used for LOS in hours (right skewed continuous variable). Ratio with 95% CI for average LOS in hours was presented. Use of non-hs-cTn assays served as the reference level for all comparisons. For all multivariable analyses (except PCI, CABG, and in-hospital mortality that were excluded because of low event rates in the low-risk chest patients), interaction of hs-cTn used and patients’ subgroup (NSTEMI or unstable angina vs low-risk chest pain) were tested to determine whether the association between hs-cTn use and the outcome was the same for NSTE-ACS (ie, NSTEMI or unstable angina) vs low-risk chest pain. Generalized score tests were used for testing the interaction terms. If the interaction term was statistically significant, we presented the effect size and 95% CI of the outcome variable for hs-cTn (vs conventional cTn) within each patient subgroup. All multivariable models discussed earlier were adjusted for age, sex, race, insurance, history of hypertension, diabetes mellitus, peripheral artery disease, prior MI, prior PCI, prior CABG, smoking status, heart failure, dialysis, admission vital signs (heart rate and systolic blood pressure), initial estimated glomerular filtration rate, initial cTn (as a ratio of the upper reference limit), cardiogenic shock, and cardiac arrest. Continuous covariates such as age, systolic blood pressure, heart rate, initial estimated glomerular filtration rate, and initial cTn ratio were fitted with restricted cubic spline with 3 knots at 10%, 50%, and 90% of their distribution of the study group. The missing percentages were small (<3%) for all the covariates in the multivariable models, missing values of continuous variables were imputed to sex and NSTEMI, unstable angina or low-risk chest pain patients’ specific median of the nonmissing values, and missing values of categorical variables were imputed to mode. We excluded patients with missing outcomes variables in each outcome model, respectively; missing percentages of outcomes were rare (<0.4%).

A P value of <0.05 was considered significant for all analyses unless otherwise stated. All statistical analyses were performed using SAS software version 9.4 software (SAS Institute).

Results

High-sensitivity cardiac troponin assay implementation at the hospital level

Among the 550 hospitals included in this analysis, implementation of hs-cTn assays increased from 3.3% in Q1, 2019 to 32.6% in Q3, 2021 (Ptrend < 0.001) (Central Illustration). There were no significant differences in geographic region, community region, membership in the Council of Teaching Hospitals, hospital bed size, level of service provided, nonprofit status, annual MI volume, or differences in patient demographic (median age, percentage of female patients, percentage of Hispanic patients, percentage of Black patients), and percentage of patients uninsured among hospitals that did (n = 179) or did not (n = 371) implement hs-cTn assays in Q3, 2021 (Table 1). The distribution of hs-cTn assays used by participating hospitals in Q3, 2021 is illustrated in Figure 1. In a sensitivity analysis, using higher implementation thresholds of at least 50% and 75% of patients evaluated by hs-cTn, the prevalence of implementation of hs-cTn assays was slightly lower among participating hospitals (Supplemental Table 1).

Central Illustration
Central Illustration

Association Between High-Sensitivity Cardiac Troponin Use and Patient-Level In-Hospital Testing and Outcomes

(Left) Increase in implementation of high-sensitivity cardiac troponin (hs-cTn) assays over time among hospitals. (Right) Association between high-sensitivity cardiac troponin use and patient-level testing and outcomes among patients with non–ST-segment elevation acute coronary syndrome (NSTE-ACS) and low-risk chest pain patients. ∗A multivariable regression analysis examining the association between high-sensitivity cardiac troponin use and in-patient mortality was not performed among low-risk chest pain individuals due to a low number of events.

Table 1 Characteristics of Hospitals That Did or Did Not Implement High-Sensitivity Cardiac Troponin Assaysa

Overall (N = 550)Did Not Implement hs-cTn (n = 371)Implemented hs-cTn (n = 179)P Value
Hospital region
 West90 (16.4)59 (15.9)31 (17.3)0.42
 Northeast60 (10.9)44 (11.9)16 (8.9)
 Midwest140 (25.5)88 (23.7)52 (29.1)
 South260 (47.3)180 (48.5)80 (44.7)
Hospital community description
 Rural97 (17.6)71 (19.1)26 (14.5)0.06
 Suburban186 (33.8)133 (35.9)53 (29.6)
 Urban267 (48.6)167 (45.0)100 (55.9)
Hospital profit type
 Government10 (1.8)10 (2.7)0 (0.0)0.09
 Private or community496 (90.2)332 (89.5)164 (91.6)
 University44 (8.0)29 (7.8)15 (8.4)
Member of Council of Teaching Hospitals73 (13.3)47 (12.7)26 (14.5)0.47
Hospital level of service
 No catheterization laboratory services1 (0.2)1 (0.3)0 (0.0)0.73
 Diagnostic angiography (only)6 (1.1)4 (1.1)2 (1.1)
 Diagnostic angiography and PCI142 (25.8)100 (27.0)42 (23.5)
 Diagnostic angiography, PCI, and cardiac surgery401 (72.9)266 (71.7)135 (75.4)
Hospital total beds294.0 (183.0-442.0)285.0 (180.0-432.0)320.0 (187.0-464.0)0.09
Annual MI volume201.6 (127.3-297.3)200.0 (127.3-287.3)208.7 (128.4-329.3)0.26
Characteristics of patients treated at hospitals
 Patient age at hospitals64.0 (62.5-66.0)64.0 (62.0-66.0)65.0 (63.0-66.0)0.15
 Percent of female patients admitted at every hospital34.8 (31.6-39.2)34.8 (31.6-39.2)34.8 (31.7-39.5)0.92
 Percent of Hispanic patients admitted at every hospital3.3 (1.2-11.2)3.5 (1.2-11.0)3.0 (1.2-11.6)0.98
 Percent of Black patients admitted at every hospital7.7 (2.6-18.1)7.8 (2.6-19.4)7.5 (2.1-15.9)0.26
 Percent of patients uninsured admitted at every hospital7.4 (4.1-11.7)7.1 (3.9-12.2)7.5 (4.5-11.1)0.90
 Percent of patients with commercial insurance admitted at every hospital60.9 (47.6-72.2)60.6 (45.7-72.3)61.7 (50.7-72.0)0.35

Values are n (%) or median (IQR [first quartile, third quartile]).

hs-cTn = high-sensitivity cardiac troponin assays; MI = myocardial infarction; PCI = percutaneous coronary intervention.

a Missing percentages were small (<5%) for all variables.

Figure 1
Figure 1

Distribution of High-Sensitivity Cardiac Troponin Assays Used by Participating Hospitals

Pie chart illustrating distribution of high-sensitivity cardiac troponin assays used among hospitals that had implemented high-sensitivity cardiac troponin by quarter 3, 2021.

Baseline patient characteristics and initial evaluation

Of the 251,000 patients who met eligibility criteria over the study period, 155,049 had an NSTEMI, 15,989 had unstable angina, and 79,962 had low-risk chest pain. Among 621 hospitals that contributed patients to the patient-level analyses of this study, 221 (36%) hospitals included low-risk chest pain patients. There was no difference in the characteristics of hospitals that did not contribute low-risk chest pain patients (Supplemental Table 2). Among 251,000 patients included in the registry during the study period, 28,967 (11.5%) were evaluated with hs-cTn assays.

Patients evaluated with hs-cTn were slightly older (65.0 years [IQR: 54.0-74.0 years] vs 64.0 years [IQR: 53.0-74.0 years]; P < 0.001) and more commonly White (83.1% vs 79.9%; P < 0.001). Additionally, those evaluated with hs-cTn were less likely to be of Hispanic or Latino ethnicity (8.9% vs 10.0%; P < 0.001) and less likely to be uninsured (8.3% vs 6.8%; P < 0.001). Differences in medical comorbidities (Table 2) at baseline were small, and although occasionally statistically significant, they were not commonly clinically significant. Persons evaluated with hs-cTn had a slightly higher prevalence of dyslipidemia (60.9% vs 57.5%; P < 0.001) and cancer (12.9% vs 11.1%; P < 0.001) than patients evaluated with conventional cTn assays (Table 2).

Table 2 Baseline Patient Characteristicsa

Overall (N = 251,000)Not Evaluated With hs-cTn (n = 222,033)Evaluated With hs-cTn (n = 28,967)P Value
Demographics
 Age, y64.0 (53.0-74.0)64.0 (53.0-74.0)65.0 (54.0-74.0)<0.001
 Female104,261 (41.5)92,343 (41.6)11,918 (41.1)<0.001
 Race
  White201,542 (80.3)177,464 (79.9)24,078 (83.1)<0.001
  Black35,774 (14.3)32,342 (14.6)3,432 (11.9)
  Asian5,094 (2.0)4,667 (2.1)427 (1.5)
  American Indian/Alaskan1,440 (0.6)1,272 (0.6)168 (0.6)
  Native Hawaiian/Pacific Islander496 (0.2)444 (0.2)52 (0.2)
 Hispanic or Latino ethnicity24,877 (9.9)22,301 (10.0)2,576 (8.9)<0.001
Health insurance
 Commercial146,734 (58.5)129,620 (58.4)17,114 (59.1)<0.001
 Government83,679 (33.3)73,823 (33.3)9,856 (34.0)
 Uninsured20,436 (8.1)18,473 (8.3)1,963 (6.8)
Past medical history
 Never smoker114,329 (45.6)101,704 (45.8)12,625 (43.6)<0.001
 Hypertension184,799 (73.6)163,398 (73.6)21,401 (73.9)0.30
 Dyslipidemia145,293 (57.9)127,664 (57.5)17,629 (60.9)<0.001
 Requiring dialysis7,298 (2.9)6,637 (3.0)661 (2.3)<0.001
 Cancer28,310 (11.3)24,582 (11.1)3,728 (12.9)<0.001
 Diabetes mellitus89,531 (35.7)79,375 (35.8)10,156 (35.1)0.02
 Prior MI50,323 (20.1)44,530 (20.1)5,793 (20.0)0.75
 Prior PCI62,501 (24.9)55,442 (25.0)7,059 (24.4)0.02
 Prior CABG30,296 (12.1)27,122 (12.2)3,174 (11.0)<0.001
 Prior heart failure36,698 (14.6)32,369 (14.6)4,329 (14.9)0.10
 Atrial fibrillation28,579 (11.4)25,102 (11.3)3,477 (12.0)<0.001
 Prior cerebrovascular disease30,974 (12.3)27,280 (12.3)3,694 (12.8)0.02
 Peripheral arterial disease17,202 (6.9)15,043 (6.8)2,159 (7.5)<0.001

Values are median (IQR) or n (%).

CABG = coronary artery bypass grafting; other abbreviations as in Table 1.

a Missing percentages were small (<5%) for all variables.

Individuals evaluated by hs-cTn assays included a slightly higher proportion of persons with a diagnosis of unstable angina (7.1% vs 6.3%), a lower proportion of patients with NSTEMI (61.1% vs 61.9%), and a similar proportion of patients with low-risk chest pain (31.8% vs 31.9%) when compared with our cohort of patients evaluated by conventional cTn assays (Table 3); collectively, the proportion of patients with NSTE-ACS (NSTEMI and unstable angina) were similar between the groups (68.2% vs 68.1%). The majority of patients did not have a documented risk score; individuals evaluated with hs-cTn more commonly had a score documented compared with persons evaluated by conventional cTn assays (43.2% vs 40.7% respectively; P < 0.001). Thrombolysis In Myocardial Infarction (TIMI; 11.6% vs 7.4%;P <0.001) and Global Registry of Acute Coronary Events (GRACE; 3.8% vs 2.4%; P < 0.001) risk scores were used more frequently among individuals evaluated with hs-cTn (Table 3). In contrast, the HEART (history, electrocardiogram, age, risk factors, and initial troponin; 32% vs 27.2%; P < 0.001) and Emergency Department Assessment of Chest Pain Score (EDACS; 0.08% vs 0.02%; P < 0.001) scores were used more frequently among individuals evaluated with conventional cTn assays, respectively.

Table 3 Initial Patient Evaluationa

Overall (N = 251,000)Not Evaluated With hs-cTn (n = 222,033)Evaluated With hs-cTn (n = 28,967)P Value
Hospital characteristics
 Hospital region
  West46,433 (18.5)43,182 (19.5)3,251 (11.2)<0.001
  Northeast15,899 (6.3)13,387 (6.0)2,512 (8.7)
  Midwest56,553 (22.5)46,635 (21.0)9,918 (34.2)
  South132,115 (52.6)118,829 (53.5)13,286 (45.9)
 Hospital community description
  Urban123,227 (49.1)107,876 (48.6)15,351 (53.0)<0.001
  Rural44,159 (17.6)39,003 (17.6)5,156 (17.8)
  Suburban83,614 (33.3)75,154 (33.9)8,460 (29.2)
 Hospital profit type
  Private227,653 (90.7)201,552 (90.8)26,101 (90.1)<0.001
  Government8,747 (3.5)8,018 (3.6)729 (2.5)
  University14,600 (5.8)12,463 (5.6)2,137 (7.4)
Member of Council of Teaching Hospitals30,042 (12.0)25,626 (11.5)4,416 (15.2)<0.001
 Hospital level of service
  No catheterization services350 (0.1)350 (0.2)0 (0.0)<0.001
  Diagnostic catheterization only1,151 (0.5)1,090 (0.5)61 (0.2)
  Diagnostic catheterization and PCI57,458 (22.9)51,476 (23.2)5,982 (20.7)
  Diagnostic catheterization, PCI, and cardiac surgery192,041 (76.5)169,117 (76.2)22,924 (79.1)
 Hospital total beds315 (209-475)313 (208-457)341 (217-558)<0.001
Cardiac status
 Time from arrival to initial ECG (among direct arrivals), min8 (5-15)8 (8-15)9 (5-16)0.11
 Symptom onset to arrival among direct presenters, h2.9 (1.3-8.0)2.9 (1.3-8.0)3.1 (1.3-8.6)<0.001
 Heart rate on admission, beats/min83.0 (71.0-97.0)83.0 (71.0-97.0)82.0 (71.0-96.0)<0.001
 Systolic BP on admission, mm Hg149.0 (131.0-170.0)149.0 (131.0-170.0)149.0 (131.0-169.0)0.25
 Acute heart failure24,480 (9.8)21,771 (9.8)2,709 (9.4)0.01
 Cardiogenic shock2,400 (1.0)2,177 (1.0)223 (0.8)<0.001
 Out-of-hospital cardiac arrest1,728 (0.7)1,533 (0.7)195 (0.7)0.74
Risk scores
 Risk score documented102,754 (40.9)90,254 (40.7)12,500 (43.2)<0.001
 TIMI risk score performed19,878 (7.9)16,527 (7.4)3,351 (11.6)<0.001
 GRACE risk score performed6,458 (2.6)5,366 (2.4)1,092 (3.8)<0.001
 HEART risk score performed79,013 (31.5)71,129 (32.0)7,884 (27.2)<0.001
 EDACS risk score performed177 (0.07)172 (0.08)5 (0.02)<0.001
 Other risk score performed3,830 (1.5)3,029 (1.4)801 (2.8)<0.001
Laboratory results
 Initial creatinine among nondialysis patients, mg/dL1.0 (0.8-1.2)1.0 (0.8-1.2)1.0 (0.8-1.2)<0.001
 Initial eGFR (MDRD) among nondialysis patients70 (54-86)70 (54-86)71 (55-87)<0.001
 Initial troponin ratio (to URL)1.5 (0.4-12.5)1.4 (0.4-11.8)2.4 (0.5-20.4)<0.001
 Peak troponin ratio (to URL)9.1 (0.7-84.9)9.0 (0.7-84.4)10.4 (0.7-88.5)<0.001
Patient final diagnosis
 Low-risk chest pain79,962 (31.9)70,742 (31.9)9,220 (31.8)<0.001
 Unstable angina15,989 (6.4)13,934 (6.3)2,055 (7.1)
 NSTEMI155,049 (61.8)137,357(61.9)17,692 (61.1)

Values are n (%) or median (IQR).

BP = blood pressure; ECG = electrocardiogram; EDACS = Emergency Department Assessment of Chest Pain Score; eGFR = estimated glomerular filtration rate; GRACE = Global Registry of Acute Coronary Events; HEART = history, electrocardiogram, age, risk factors, and initial troponin; MDRD = Modification of Diet in Renal Disease; NSTEMI = non–ST-segment myocardial infarction; TIMI = Thrombolysis In Myocardial Infarction; URL = upper reference limit; other abbreviations as in Table 1.

a Missing percentages were small (<7%) for all variables.

In-hospital testing and revascularization

Differences in testing patterns among patients evaluated with hs-cTn as compared with conventional cTn assays are shown in Table 4 and in the Central Illustration.

Table 4 In-Hospital Cardiac Testing and Revascularization

SubgroupAmong Patients Evaluated With hs-cTnAmong Patients Not Evaluated With hs-cTnUnadjusted OR (95% CI)Adjusted OR (95% CI)Adjusted P Value
TTENSTEMI/UA15,169 (82.4)105,239 (75.0)1.41 (1.19-1.66)1.43 (1.19-1.73)<0.001
Low-risk chest pain1,625 (19.7)12,816 (19.4)0.99 (0.74-1.33)0.93 (0.71-1.22)0.59
Stress testing or coronary CTANSTEMI/UA1,453 (7.9)10,310 (7.3)0.93 (0.79-1.10)0.98 (0.72-1.35)0.93
Low-risk chest pain1,661 (20.1)16,327 (24.6)
CMRNSTEMI/UA72 (0.4)348 (0.3)1.71 (1.16-2.50)1.73 (1.12-2.69)0.01
Low-risk chest pain14 (0.2)191 (0.3)
Invasive coronary angiographyNSTEMI/UA16,992 (96.3)128,158 (95.8)1.02 (0.81-1.30)0.99 (0.82-1.19)0.90
Low-risk chest pain296 (3.7)2,898 (4.5)0.82 (0.66-1.30)0.73 (0.58-0.92)0.008
PCINSTEMI/UA10,394 (52.7)79,009 (52.3)0.99 (0.95-1.04)0.99 (0.94-1.04)0.77
Low-risk chest pain8 (0.09)123 (0.2)
CABGNSTEMI/UA1,859 (9.4)13,707 (9.1)1.06 (0.96-1.18)1.06 (0.94-1.18)0.34
Low-risk chest pain8 (0.09)38 (0.05)

Values are n (%) unless otherwise indicated.

CMR = cardiac magnetic resonance; CTA = computed tomography angiography; OR = odds ratio; PCI = percutaneous coronary intervention; TTE = transthoracic echocardiogram; UA = unstable angina; other abbreviations as in Tables 1, 2, and 3.

A significant interaction was present (P = 0.009) between use of hs-cTn and subsequent use of echocardiography for certain subgroups: among patients with NSTE-ACS, use of hs-cTn was associated with greater use of echocardiography (82.4% vs 75.0%; adjusted OR: 1.43; 95% CI: 1.19-1.73) compared with patients evaluated with conventional cTn assays. In contrast, among low-risk chest pain patients, there was no significant association between hs-cTn evaluation and echocardiography compared with conventional assays (19.7% vs 19.4%, respectively; adjusted OR: 0.93; 95% CI: 0.71-1.22).

Among all individuals in the analysis (NSTE-ACS and low-risk chest pain), hs-cTn evaluation was associated with slightly greater CMR use as compared with persons evaluated with conventional cTn assays; however, the incident rates were low (0.3% vs 0.3%; adjusted OR: 1.73; 95% CI: 1.12-2.69). The interaction for CMR and hs-cTn evaluation by NSTE-ACS status (vs low-risk chest pain) presentation was not significant (P = 0.15).

No association between hs-cTn evaluation and increased use of stress testing or CT coronary angiography was observed compared with conventional cTn use (11.7% vs 12.9%, respectively; adjusted OR: 0.98; 95% CI: 0.72-1.35). No significant interaction was detected for the composite of stress testing or CT coronary angiography and hs-cTn evaluation by diagnosis (NSTE-ACS vs unstable angina; P = 0.08).

A significant interaction was observed (P = 0.047) between invasive coronary angiography and hs-cTn evaluation among the subgroups; among patients with NSTE-ACS, most patients without contraindications underwent invasive coronary angiography, with no difference among those tested with hs-cTn as compared with those evaluated with conventional cTn assays (96.3% vs 95.8%, respectively; adjusted OR: 0.99; 95% CI: 0.82-1.19). However, among low-risk chest pain patients, those evaluated with hs-cTn assays were less likely to undergo invasive coronary angiography (3.7% vs 4.5%; adjusted OR: 0.73; 95% CI: 0.58-0.92) compared with patients evaluated with conventional assays.

Coronary revascularization with PCI (0.1% vs 0.2%; P = 0.05) and CABG (0.1% vs 0.1%; P = 0.21) were uncommon among low-risk chest pain patients evaluated using hs-cTn and conventional cTn assays, respectively. Among patients with NSTE-ACS, there was no significant difference in coronary revascularization with PCI (52.7% vs 52.3%; adjusted OR: 0.99; 95% CI: 0.94-1.04) or CABG (9.4% vs 9.1%; adjusted OR: 1.06; 95% CI: 0.94-1.18) among patients evaluated by hs-cTn compared with conventional cTn, respectively.

In-hospital mortality and length of stay in hours

In-hospital mortality was very low for low-risk chest pain patients and was similar among individuals evaluated by hs-cTn and conventional cTn assays (0% vs 0.02%, respectively; P = 0.16). Among patients with NSTE-ACS, no difference in in-hospital mortality was observed between patients evaluated with hs-cTn compared with patients evaluated with conventional cTn (2.8% vs 3.2%; adjusted OR: 0.98; 95% CI: 0.87-1.11).

LOS in hours was slightly but significantly shorter among patients with NSTE-ACS evaluated by hs-cTn as compared with conventional cTn (median LOS, 66.9 hours [IQR: 45.1-120.6 hours] vs 67.8 hours [IQR: 45.5-122.3 hours]; P = 0.01). LOS was also shorter among low-risk chest pain patients evaluated by hs-cTn as compared with conventional cTn (median LOS, 5.8 hours [IQR: 3.4-24.9 hours] vs 6.2 hours [IQR: 3.5-26.6 hours]; P < 0.001). There was no interaction between hospital LOS and hs-cTn evaluation by patient presentation (P = 0.39). Among all patients (Table 5), those evaluated by hs-cTn assays had 0.94 times the average LOS in hours compared with patients evaluated with conventional assays after adjustment for confounding variables (median LOS, 47.6 hours [IQR: 21.8-87.5 hours] vs 48.0 hours [IQR: 23.1-89.3 hours]; ratio: 0.94; 95% CI: 0.90-0.98).

Table 5 In-Hospital Mortality and Length of Stay

OutcomeSubgroupAmong Patients Evaluated With hs-cTnAmong Patients Not Evaluated With hs-cTnUnadjusted OR (95% CI)Adjusted OR (95% CI)Adjusted P Value
MortalityNSTEMI/UA517 (2.8)4,469 (3.2)0.94 (0.83-1.06)0.98 (0.87-1.11)0.77
Low-risk chest pain0 (0.0)15 (0.02)
LOS Among Patients Evaluated With hs-cTn, hLOS Among Patients Not Evaluated With hs-cTn, hUnadjusted Ratio (95% CI)Adjusted Ratio (95% CI)
LOSNSTEMI/UA66.9 (45.5-120.6)67.8 (45.5-122.3)0.95 (0.92-0.98)0.94 (0.90-0.98)0.009
Low-risk chest pain5.8 (3.4-24.9)6.2 (3.5-26.6)

Values are n (%) or median (IQR) unless otherwise indicated.

LOS = length of stay; other abbreviations as in Tables 1, 3, and 4.

Discussion

In this national examination of hs-cTn implementation in the United States that included 550 hospitals of diverse size, geographic location, and nonprofit status, we report several important findings. First, more than two-thirds of U.S. hospitals included in the NCDR registry had not implemented hs-cTn testing by the end of the study period in September 2021. Second, we observed greater use of echocardiography among patients with NSTE-ACS and a lower use of invasive coronary angiography among low-risk chest pain patients when they were evaluated with hs-cTn assays as compared with conventional assays. Third, patients evaluated by hs-cTn had a modestly shorter LOS and no difference in in-hospital mortality compared with patients evaluated by conventional cTn assays. Contrary to concerns that hs-cTn implementation could be associated with excess admissions and increased use of costly testing in the United States, these results (which are similar to those reported in studies performed outside the United States) provide reassurance that transitioning to hs-cTn methods does not appear to increase resource use.

Comparative effectiveness trials suggest that hs-cTn assays may reduce LOS without compromising safety compared with conventional cTn assays.20-22 However, despite approval in the United States in 2017, the majority of hospitals in this large national Chest Pain-MI Registry had not implemented hs-cTn testing by the fall of 2021. There are several potential explanations for this slow uptake. First, the handful of prospective comparative effectiveness trials of hs-cTn assays has predominantly occurred in international populations.20-22 Thus, few real-world U.S. data on implementation of hs-cTn have been published, with most limited to results from integrated health networks that are based at academic institutions.14-17,23 Accordingly, reservations may persist regarding whether hs-cTn implementation could increase LOS and cascade testing. Furthermore, practice patterns in the United States, including greater testing of lower-risk patients in the setting of more prevalent cardiovascular comorbidities combined with fears of litigation resulting from missed MI, may render observations in international populations less relevant in the United States. Our findings in a large spectrum of geographic locations and hospital types should provide reassurance regarding these concerns. A second reason for slow uptake may be the delayed Food and Drug Administration approval of several assays until more recently. Finally, the 2021 American Heart Association (AHA) and ACC chest pain guideline, which provided a Class I recommendation for hs-cTn as the preferred biomarker for evaluating acute chest pain, was not published until October 2021; this may provide the impetus to increase hs-cTn assay uptake.24

At a hospital-level, no statistical differences were present in hospital characteristics according to hs-cTn implementation at a hospital-level in Q3, 2021. Similarly, the characteristics of patients evaluated at hospitals who had or had not implemented hs-cTn were similar in the final quarter of our study.

Prior single hospital network implementation studies from the United States have observed small declines (or no change) in stress testing and CT coronary angiography and no significant change in echocardiography use after implementation of hs-cTn assays.14-17 The present study’s findings, derived from 550 institutions (including 179 that had implemented hs-cTn assays) are generally concordant with these smaller studies. Although in the present analysis evaluation of chest pain with hs-cTn was associated with more echocardiography use among patients with NSTE-ACS, this should be considered an appropriate test use in these patients.24,25 However, reassuringly, we did not find an increase in echocardiography use among low-risk chest pain patients evaluated with hs-cTn. Importantly, hs-cTn was not associated with a difference in stress testing or CT coronary angiography use. Our data are consistent with most (but not all) single-center studies in the United States regarding the impact of hs-cTn on use of invasive coronary angiography and revascularization.14-17 We found that hs-cTn use was associated with lower use of invasive coronary angiography in low-risk chest pain patients and observed no difference in invasive angiography, PCI, or CABG among individuals with NSTE-ACS. Collectively, these data suggest favorable changes in the appropriateness of subsequent testing. We anticipate that following the 2021 AHA/ACC chest pain guideline,24 unnecessary testing in low-risk chest pain patients may be further reduced when these patients are assessed with hs-cTn assays.

No difference was observed in in-hospital mortality among patients evaluated with hs-cTn assays compared with conventional cTn assays. However, consistent with international data,13 clinical trial results,21,22 and smaller U.S. implementation studies,14,16 evaluation with hs-cTn was associated with a modestly shorter hospital LOS. Importantly, LOS was not increased as a result of the potential increase in hs-cTn elevations. We were unable to assess how frequently institutions were using hs-cTn–based accelerated diagnostic pathways. Widespread use of such pathways could be expected to reduce LOS to durations substantially lower than those observed in this study, particularly among low-risk chest pain patients. Implementation of hs-cTn without use of an accelerated diagnostic pathway may provide only marginal reductions in LOS.

Study limitations

Given that the entry of low-risk chest pain and unstable angina patients into the registry is optional for participating sites, and if entered, institutions have the option of including all their eligible patients or a sample, there is a possibility of selection bias. Indeed, the percentage of patients in this analysis with NSTEMI is higher than typical chest pain analyses; this higher pretest probability for MI may thus affect post-test accuracy for a true positive result. That stated, this is the exact scenario where higher sensitivity may be associated with a favorable impact on use. Arguably the strongest value of hs-cTn in this analysis was the impact of the negative result, with shorter LOS and no increase in use or poor outcomes. Impact of hs-cTn on lower prevalence populations requires further prospective study in patients presenting with chest pain. Furthermore, testing and outcome analyses were performed using multivariable models to account for known or suspected confounding variables. However, there is a possibility of unmeasured confounders contributing to the differences in testing and outcomes observed in our analyses. Furthermore, this limitation also precludes us from determining the magnitude of the influence of hs-cTn implementation on the diagnosis of NSTEMI vs unstable angina, as well as and the resulting absolute changes in echocardiogram, stress test, or angiography volumes. Additionally, assignment of diagnosis (NSTEMI, unstable angina, or low-risk chest pain) occurs at the individual sites; therefore, we were not able to confirm the accuracy of these classifications. However, the NCDR uses a data quality program, and agreement during audits of the Chest Pain-MI registry is high (>90%).19 Type 2 MI patients are excluded from the registry, and it is unknown whether their inclusion would have altered our observations. Given the increase in hs-cTn use over time, it is unclear whether observed associations of hs-cTn use with hospital processes of care and LOS reflect the impact of hs-cTn or underlying temporal trends in test use or LOS that are independent of hs-cTn use. Because postdischarge outcomes such as 30-day mortality or recurrent MI were not available in the registry, we could not assess postdischarge safety, which may be particularly relevant for low-risk chest pain patients, nor could we assess postdischarge invasive or noninvasive tests that are not included in the registry.

Conclusions

Although implementation of hs-cTn among U.S. hospitals is increasing, most U.S. hospitals continue to use less sensitive cTn assays. Chest pain evaluation using hs-cTn assays was associated with shorter LOS overall, fewer invasive angiograms in low-risk chest pain patients, and no difference in mortality. Use of hs-cTn was associated with more use of echocardiography among individuals with NSTE-ACS. These data indicate that further opportunities to implement hs-cTn more widely and effectively in U.S. hospitals persist that could optimize care for patients with possible or definitive ACS.

Perspectives

COMPETENCY IN SYSTEMS-BASED PRACTICE: hs-cTn assays are the preferred biomarker for evaluating patients with acute chest pain, yet most hospitals in the United States still employ less-sensitive troponin assays.

TRANSLATIONAL OUTLOOK: Further efforts are needed to identify and overcome barriers to wider availability and implementation of hs-cTn assays.

Funding Support and Author Disclosures

This research was supported by the American College of Cardiology’s National Cardiovascular Data Registry (NCDR). The views expressed in this manuscript represent those of the author(s) and do not necessarily represent the official views of the NCDR or its associated professional societies identified at CVQuality.ACC.org/NCDR. Dr McCarthy has received support from the National Heart, Lung, and Blood Institute (NHLBI) T32 postdoctoral training grant (5T32HL094301-12); and has received consulting income from Abbott Laboratories. Dr Wang has received research grants to the Duke Clinical Research Institute from Abbott, AstraZeneca, Bristol Myers Squibb, Boston Scientific, Cryolife, Chiesi, Merck, Portola, and Regeneron; and has received consulting honoraria from AstraZeneca, Bristol Myers Squibb, Cryolife, and Novartis. Dr Sandoval has served on advisory boards for Roche Diagnostics and Abbott Diagnostics without personal financial compensation; and has also been a speaker without personal financial compensation for Abbott Diagnostics. Dr Smilowitz has received partial support from a Career Development Award from NHLBI (K23HL150315); and has received consulting honoraria from Abbott Vascular. Dr Wasfy has received support from the American Heart Association (18 CDA 34110215). Dr de Lemos has received grant support from Abbott Diagnostics and Roche Diagnostics; and has received consulting income from Ortho Clinical Diagnostics, Quidel, Beckman Coulter, and Siemens’s Health Care Diagnostics. Dr Kontos has served as chair of the Chest Pain-MI Steering committee. Dr Apple has served as an Associate Editor of Clinical Chemistry; has participated in the advisory boards of Werfen, Siemens Healthineers, and Qorvo Biotechnology; has received consulting income from AWE Medical and Hytest; and has received cardiac biomarker grant support (nonsalaried) through his institutional research institute from Abbott Diagnostics, Abbott POC, Siemens Healthcare, Ortho-Clinical Diagnostics, Roche Diagnostics, Beckman Coulter, BD, and Quidel. Dr Daniels has received consulting income from Quidel, Roche, and Siemens; and has participated in clinical endpoint committees or data safety monitoring boards for Abbott, Applied Therapeutics, and Quidel. Dr Newby has received research grant support from Roche Diagnostics and BioKier; and has received consulting honoraria from Beckman-Coulter, CSL, and Medtronic. Dr Jaffe has served as a consultant for Abbott, Beckman-Coulter, Siemens, Roche, Ortho Diagnostics, Radiometer, ET Healthcare, Sphingotec, Astellas, RCE Technologies, Amgen, and Novartis. Dr Januzzi has served as a Trustee of the American College of Cardiology; has received grant support from Abbott, Applied Therapeutics, HeartFlow Inc, Innolife, and Roche Diagnostics; has received consulting income from Abbott, Janssen, Novartis, Merck, and Roche Diagnostics; and has participated in clinical endpoint committees or data safety monitoring boards for Abbott, AbbVie, Bayer, CVRx, Pfizer, and Takeda. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.

Abbreviations and Acronyms

ACC

American College of Cardiology

ACS

acute coronary syndrome

CABG

coronary artery bypass grafting

cTn

cardiac troponin

GEE

generalized estimating equations

hs-cTn

high-sensitivity cardiac troponin

LOS

length of stay

MI

myocardial infarction

NCDR

National Cardiovascular Data Registry

NSTE-ACS

non–ST-segment elevation acute coronary syndrome

NSTEMI

non–ST-segment elevation myocardial infarction

PCI

percutaneous coronary intervention

STEMI

ST-segment elevation myocardial infarction

References

  • 1. Katus H.A., Remppis A., Neumann F.J., et al. "Diagnostic efficiency of troponin T measurements in acute myocardial infarction". Circulation 1991;83:902-912.

    CrossrefMedlineGoogle Scholar
  • 2. Apple F.S., Falahati A., Paulsen P.R., et al. "Improved detection of minor ischemic myocardial injury with measurement of serum cardiac troponin I". Clin Chem 1997;43:2047-2051.

    CrossrefMedlineGoogle Scholar
  • 3. Brogan G.X., Hollander J.E., McCuskey C.F., et al. "Evaluation of a new assay for cardiac troponin I vs creatine kinase-MB for the diagnosis of acute myocardial infarction. Biochemical Markers for Acute Myocardial Ischemia (BAMI) Study Group". Acad Emerg Med 1997;4:6-12.

    CrossrefMedlineGoogle Scholar
  • 4. 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.

    CrossrefMedlineGoogle Scholar
  • 5. 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.

    CrossrefMedlineGoogle Scholar
  • 6. 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.

    CrossrefMedlineGoogle Scholar
  • 7. Thygesen K., Alpert J.S., Jaffe A.S., et al. "Fourth universal definition of myocardial infarction (2018)". J Am Coll Cardiol 2018;72:2231-2264.

    View ArticleGoogle Scholar
  • 8. Chew D.P., Aroney C.N., Aylward P.E., et al. "2011 Addendum to the National Heart Foundation of Australia/Cardiac Society of Australia and New Zealand guidelines for the management of acute coronary syndromes (ACS) 2006". Heart Lung Circ 2011;20:487-502.

    CrossrefMedlineGoogle Scholar
  • 9. Roffi M., Patrono C., Collet J.P., et al. "2015 ESC guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: Task Force for the Management of Acute Coronary Syndromes in Patients Presenting without Persistent ST-Segment Elevation of the European Society of Cardiology (ESC)". Eur Heart J 2016;37:267-315.

    CrossrefMedlineGoogle Scholar
  • 10. 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.

    View ArticleGoogle Scholar
  • 11. Twerenbold R., Neumann Johannes T., et al. "Prospective validation of the 0/1-h algorithm for early diagnosis of myocardial infarction". J Am Coll Cardiol 2018;72:620-632.

    View ArticleGoogle Scholar
  • 12. Chapman A.R., Anand A., Boeddinghaus J., et al. "Comparison of the efficacy and safety of early rule-out pathways for acute myocardial infarction". Circulation 2017;135:1586-1596.

    CrossrefMedlineGoogle Scholar
  • 13. Twerenbold R., Jaeger C., Rubini Gimenez M., et al. "Impact of high-sensitivity cardiac troponin on use of coronary angiography, cardiac stress testing, and time to discharge in suspected acute myocardial infarction". Eur Heart J 2016;37:3324-3332.

    CrossrefMedlineGoogle Scholar
  • 14. Ganguli I., Cui J., Thakore N., Orav E.J., et al. "Downstream cascades of care following high-sensitivity troponin test implementation". J Am Coll Cardiol 2021;77:3171-3179.

    View ArticleGoogle Scholar
  • 15. Sandoval Y., Askew J.W., Newman J.S., et al. "Implementing high-sensitivity cardiac troponin T in a US regional healthcare system". Circulation 2020;141:1937-1939.

    CrossrefMedlineGoogle Scholar
  • 16. Ola O., Akula A., Michieli L.D., et al. "Clinical impact of high-sensitivity cardiac troponin-T assays in the community". J Am Coll Cardiol 2021;77:3160-3170.

    View ArticleGoogle Scholar
  • 17. Ford J.S., Chaco E., Tancredi D.J., et al. "Impact of high-sensitivity cardiac troponin implementation on emergency department length of stay, testing, admissions, and diagnoses". Am J Emerg Med 2021;45:54-60.

    CrossrefMedlineGoogle Scholar
  • 18. Peterson E.D., Roe M.T., Rumsfeld J.S., et al. "A call to ACTION (Acute Coronary Treatment and Intervention Outcomes Network): a national effort to promote timely clinical feedback and support continuous quality improvement for acute myocardial infarction". Circ Cardiovasc Qual Outcomes 2009;2:491-499.

    CrossrefMedlineGoogle Scholar
  • 19. Malenka D.J., Bhatt D.L., Bradley S.M., et al. "The National Cardiovascular Data Registry Data Quality Program 2020". J Am Coll Cardiol 2022;79:1704-1712.

    View ArticleGoogle Scholar
  • 20. Anand A., Lee K.K., Chapman A.R., et al. "High-sensitivity cardiac troponin on presentation to rule out myocardial infarction: a stepped-wedge cluster randomized controlled trial". Circulation 2021;143:2214-2224.

    CrossrefMedlineGoogle Scholar
  • 21. Chew D.P., Lambrakis K., Blyth A., et al. "A randomized trial of a 1-hour troponin T protocol in suspected acute coronary syndromes". Circulation 2019;140:1543-1556.

    CrossrefMedlineGoogle Scholar
  • 22. Shah A.S.V., Anand A., Strachan F.E., et al. "High-sensitivity troponin in the evaluation of patients with suspected acute coronary syndrome: a stepped-wedge, cluster-randomised controlled trial". Lancet 2018;392:919-928.

    CrossrefMedlineGoogle Scholar
  • 23. Bevins N.J., Chae H., Hubbard J.A., et al. "Emergency department management of chest pain with a high-sensitivity troponin-enabled 0/1-hour rule-out algorithm". Am J Clin Pathol 2022;157:774-780.

    CrossrefGoogle Scholar
  • 24. Gulati M., Levy P.D., Mukherjee D., et al. "2021 AHA/ACC/ASE/CHEST/SAEM/SCCT/SCMR guideline for the evaluation and diagnosis of chest pain: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines". J Am Coll Cardiol 2021;78:22: e187-e285.

    View ArticleGoogle Scholar
  • 25. Amsterdam E.A., Wenger N.K., Brindis R.G., et al. "2014 AHA/ACC guideline for the management of patients with non-ST-elevation acute coronary syndromes: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines". J Am Coll Cardiol 2014;64:24: e139-e228. https://doi.org/10.1016/j.jacc.2014.09.017.

    View ArticleGoogle Scholar

Footnotes

Listen to this manuscript's audio summary by Editor-in-Chief Dr Valentin Fuster on www.jacc.org/journal/jacc.

Martha Gulati, MD, MS, served as Guest Associate Editor for this paper. Athena Poppas, MD, served as Guest Editor-in-Chief for this paper.

The authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the Author Center.