Minimal MI Risk with Undetectable hs-cTnT and No ECG Ischemia in ED Patients with Chest Pain

Toni Rizzo

doi:10.1177/155989771404010

Summary

Approximately 15 to 20 million patients visit hospital emergency departments (EDs) in Europe and the United States for chest pain each year [Thygesen K et al. J Am Coll Cardiol 2012; Nawar EW et al. Adv Data 2007; Goodacre S et al. Heart 2005]. Traditionally several cardiac troponin measurements are required over hours in order to detect injury indicative of myocardial infarction. Recently developed high-sensitivity cardiac troponin assays, however, can detect increased troponin concentrations hours earlier than older generation assays potentially establishing a diagnosis with a single measurement. This article discusses the Undetectable High Sensitivity Cardiac Troponin T Level in the Emergency Department and Risk of Myocardial Infarction study [Bandstein N et al. J Am Coll Cardiol 2014].

Myocardial Infarction
Imaging Modalities
Emergency Radiology
Cardiac Imaging Techniques
Cardiology Clinical Trials

Myocardial Infarction
Imaging Modalities
Emergency Radiology
Cardiac Imaging Techniques
Cardiology Clinical Trials
Cardiology

Approximately 15 to 20 million patients visit hospital emergency departments (EDs) in Europe and the United States for chest pain each year [Thygesen K et al. J Am Coll Cardiol 2012; Nawar EW et al. Adv Data 2007; Goodacre S et al. Heart 2005]. However, only 10% to 20% of patients hospitalized for chest pain are diagnosed with myocardial infarction (MI) [Body R et al. J Am Coll Cardiol 2011; Than M et al. Lancet 2011; Pope JH et al. N Engl J Med 2000]. MI is characterized by cardiac troponin elevation in the presence of symptoms, ischemic electrocardiogram (ECG) changes or diagnostic imaging (eg, coronary angiography, echocardiogram) [Thygesen K et al. J Am Coll Cardiol 2012]. Traditionally several cardiac troponin measurements are required over hours in order to detect injury indicative of myocardial infarction. Recently developed high-sensitivity cardiac troponin assays, however, can detect increased troponin concentrations hours earlier than older generation assays potentially establishing a diagnosis with a single measurement.

To explore this hypothesis the investigators performed the Undetectable High Sensitivity Cardiac Troponin T Level in the Emergency Department and Risk of Myocardial Infarction study [Bandstein N et al. J Am Coll Cardiol 2014], which was presented by Nadia Bandstein, MD, Karolinska University Hospital, Stockholm, Sweden. The investigators hypothesized that all patients presenting with chest pain who have a first undetectable high-sensitivity cardiac troponin T (hs-cTnT), independent of symptom duration, and no signs of ischemia on ECG may be safely discharged from the ED.

The study population included all patients aged ≥25 years who visited the Karolinska University Hospital ED for chest pain and who had at least one hs-cTnT level analyzed between December 10, 2010, and December 31, 2012. The primary analysis group included all patients with an undetectable hs-cTnT (<5 ng/L) and no significant ST-elevation or depression on ECG. The primary endpoint was fatal or nonfatal MI within 30 days. Secondary endpoints included MI at 180 and 365 days and all-cause mortality at 30, 180, and 365 days. The hs-cTnT assay (Roche Diagnostics) has a detection limit of 2 ng/L, a 99th-percentile cutoff point of 14 ng/L, and a coefficient of variation of <10% at 13 ng/L.

A total of 330,821 patients visited the ED within the study period. Of these, 14,636 (4.4%) were aged ≥25 years with chest pain and had at least one hs-cTnT measured. The hs-cTnT was <5 ng/L (undetectable) in 61%, 5 to 14 ng/L in 21% (detectable but <99^th percentile), and >14 ng/L (>99^th percentile) in 18% of included patients. Patients in the group with undetectable hs-cTnT were younger, more likely to be female, and less likely to have diabetes, prior MI, stroke, or congestive heart failure (Table 1).

View this table:

Table 1.

Patient Characteristics

MI was defined according to recent guidelines [Thygesen K et al. J Am Coll Cardiol 2012]. Among the patients with undetectable hs-cTnT and no ischemia on the presenting ECG, there were 15 MIs within 30 days (negative predictive value [NPV], 99.8%; 95% CI, 99.7 to 99.9). In comparison, within 30 days there were 97 MIs in the group with hs-cTnT levels of 5 to 14 ng/L (NPV, 96.9%; 95% CI, 96.3 to 97.5) and 676 recurrent MIs in those with hs-cTnT levels >14 ng/L at the index presentation (NPV, 73.8%; 95% CI, 72.1 to 75.5; Table 2).

View this table:

Table 2.

Absolute Risk and Negative Predictive Value for MI at 30 and 365 Days After ED Discharge

There were two deaths within 30 days among patients with undetectable hs-cTnT (NPV, 100%; 95% CI, 99.9 to 100). At 1 year, there were 38 deaths in this group, of which 32 were caused by cancer and two were due to cardiovascular disease.

The admission rate for patients with an undetectable first hs-cTnT was 21%. Most patients (89%) had a second hs-cTnT measured and of these, 90% remained undetectable. The authors acknowledge that there may have been patients discharged without a second hs-cTnT measured who would have had an elevated hs-cTnT consistent with an MI if it had been checked; however, they imply that the equivalent 1-year mortality rate in the admitted and discharged populations argues against significant, unrecognized MIs.

Overall, the results of this study demonstrated that a first undetectable hs-cTnT level (<5 ng/L) and no signs of ischemia on ECG in patients presenting to the ED with chest pain ruled out MI with a high degree of accuracy, regardless of duration of symptoms, timing of hs-cTnT measurement, prior disease, age, sex, or other risk factors for MI. Dr. Bandstein concluded that use of a single hs-cTnT may prevent unnecessary hospital admissions and shorten ED stays leading to a reduction in ED overcrowding.

View Summary