Mokhtari A, Lindahl B, Smith JG, Holzmann MJ, Khoshnood A, Ekelund U. Diagnostic Accuracy of High-Sensitivity Cardiac Troponin T at Presentation Combined With History and ECG for Ruling Out Major Adverse Cardiac Events. Ann Emerg Med. 2016;68(6):649-658.e3. [paper]
Why I chose this article
High-sensitivity cardiac troponin T (hs-cTnT) is a relatively new biomarker assay and the clinical application is just beginning to be explored. It has potential to not only have significant diagnostic value, but prognostic value as well. Even though we’re not ordering hs-cTnT right now, it seems reasonably expected that we’ll be incorporating this marker into our clinical practice in the near future.
Acute coronary syndrome (ACS) is a chief concern in the ED among chest pain patients and it’s typically evaluated with history, ECG, and troponin levels. Studies suggest that a single hs-cTnT below the limit of detection (<5 ng/L) at ED presentation can accurately rule out acute MI, but the test alone does not safely rule out unstable angina. One study suggests a cutoff of <14ng/L could be safely used but this has not be validated. It has been suggested but not validated that a combination of hs-cTnT, ECG, and physician gestalt can safely rule out ACS and major adverse cardiac events (MACE).
- limit of blank of 3 ng/L
- limit of detection of 5 ng/L
- coefficient of variation <10% at 99th percentile cutoff point of 14 ng/L
Can non high-risk clinical gestalt, non-ischemic ECG result, and a single hs-cTnT <5 ng/mL at presentation identify patients at very low risk for 30-day MACE?
Prospective observational study of patients presenting between Feb 2013 and Apr 2014 at a Swedish tertiary care hospital with annual ED census of 65,000. The ED was staffed with emergency physicians including interns, residents, and attendings. Patient history was stratified as high, intermediate, low, very low risk and all patients were managed per usual care. Investigators evaluated hs-cTnT alone or in combination with ECG (ischemic/nonischemic) and/or clinical history (high/non high-risk).
- 18 years or older
- Primary complaint of nontraumatic chest pain
- hs-cTnT had to have been ordered at presentation (0 hours, some mild caveats)
- Severe communication barriers, dementia, unable to provide consent
- Incomplete history, ECG result, or hemolyzed 0-hour sample
Primary outcome: MACE within 30 days, including index visit.
- Acute MI
- Unstable angina
- Cardiac arrest
- Cardiogenic shock
- Ventricular arrhythmia requiring intervention
- High-degree AV block requiring intervention
- Death from cardiac or unknown cause
Secondary outcome: 30-day MACE w/o unstable angina
Study Subject Characteristics
- 1167 patients enrolled, 1138 included in final analysis
- Median age 53.2 (18-98 years old)
- 54.6% male
- 19.9% w/ previous acute MI
- 96.1% had two hs-cTnTs during index visit
- 11% had 30-day MACE
- 7.6% had 30-MACE w/o UA
- 7% (80 of 1138 patients) had an acute MI at the index visit
- Of these, 14 had a 0-hour hs-cTnT measured < or = 2 hours from symptom onset
- Of these, 3 had an initial hs-cTnT between 5 and 14 ng/L
- Using only hs-cTnT <5ng/L
- 30.1% of all patients identified for rule-out
- Sensitivity of 96.8%, LR- 0.10, and NPV 98.8% for 30-day MACE
- Post-test probability = 1.2%
- 4 patients missed (2 acute MI, 2 UA)
- hs-cTnT <5 ng/L + negative ECG
- Sensitivity 97.6%, LR- 0.07, NPV 99.1%
- 3 patients missed (1 acute MI, 2 UA)
- hs-cTnT < 5 ng/L + negative ECG + non high-risk history
- 29.2% of patients ruled out
- Sensitivity of 99.2%, LR- 0.02, and NPV 99.7%
- 1 patient missed (UA)
- hs-cTnT < 14 ng/L had lower sensitivity (low 90s%) both for 30-day MACE and 30-day MACE w/o UA
This was a single-site study in Sweden which makes generalizability difficult, even though the authors state that they have a similar acute MI prevalence to other studies. They had “office hours” enrollment instead of continuous enrollment which can introduce bias. This study is a good start, but there needs to be additional studies or a large multi-national study to further validate.
This study showed that a hs-cTnT <5 ng/L at presentation with nonischemic ECG and non high-risk history identified patients with very low risk of MACE within 30 days for the first time. They further clarified that hs-cTnT <5 ng/L alone does not rule out ACS as four patients, two with acute MIs, would have been missed using this assay alone.
A strategy incorporating hs-cTnT as above could potentially allow safe discharge of ~30% of ED chest pain patients after a single hs-cTnT at presentation. The authors accepted a test threshold/equipoise at a 30-day MACE risk of 2% as the point at which the risks/benefits of further testing (stress test, admission) are nearing balance, although they reference a study where EM physicians tend to seek a lower MACE risk of 0.5%. Their study data are well within these thresholds.
If implemented, I think this could have some major implications on ED length of stay and resource utilization. Patients could potentially be going home after a single test, optimizing ED flow, while maintaining similar outcomes in ACS rule out. All in all, I think it would be a beneficial strategy for the reasons above.