High-Sensitivity Troponin T Test Rules Out AMI in ED

Ricki Lewis, PhD

April 17, 2017

A single high-sensitivity cardiac troponin T (hs-cTnT) test with findings below the level of detection, teamed with a nonischemic electrocardiogram (ECG), can identify patients with chest pain in the emergency department (ED) who are at sufficiently low risk for acute myocardial infarction (AMI) to be safely released to outpatient care, a new study has found.

John W. Pickering, PhD, from the University of Otago in Christchurch, New Zealand, and colleagues, report the findings of their meta-analysis online April 17 in Annals of Internal Medicine.

The researchers and an independent commentator caution that a second test should be repeated 2 to 3 hours after symptom onset. "Because troponin may not be detectable in the circulation immediately after myocardial injury, some patients with AMI who present very early after onset of pain may not have detectable troponin," the researchers explain.

"For this reason, guidelines recommend a second sample approximately 3 hours after symptom onset in these patients. We observed that 50% of patients with false-negative results had blood sampling within 3 hours of symptom onset and therefore recommend a cautious approach to implementation to exclude patients presenting soon after symptom onset. Data are currently insufficient to establish a minimum safe duration below 3 hours," the authors write

A previous study of the use of hs-cTnT to rule out AMI concluded that only 0.2% of 14,636 patients sustain an AMI within 30 days of discharge after a determination of low risk with hs-cTnT and ECG. However, the data were collected from just one urban university hospital. Other studies have considered patients from EDs that use different cutoffs and sometimes measured troponin I as well as troponin T.

To re-evaluate the relationship between use of hs-cTnT and ECG to identify ED patients at low risk for AMI within 30 days, the new meta-analysis included 11 geographically diverse cohorts (from France, England, Germany, the Netherlands, Spain, Italy, Switzerland, New Zealand, and Australia) that used a level of detection less than 0.005 μg/L on the Roche Diagnostics hs-cTnT assay.

The researchers searched MEDLINE and EMBASE for prospective studies published in peer-reviewed journals from 2008 to 2016 that followed evaluation of acute coronary syndrome in the ED using ECG and hs-cTnT. The two tests establish the primary endpoint of biochemical evidence of myocardial damage (hs-cTnT) and clinical evidence of ischemia (ECG). Secondary endpoints were death and major adverse cardiac events (cardiac arrest, AMI, cardiogenic shock, emergency revascularization procedure, ventricular arrhythmia, or high-degree atrioventricular block requiring intervention) within 30 days.

The 11 studies included 9241 patients, ranging from 166 to 2831 at individual institutions. Of the patients, 15.4% actually had an AMI. Use of the two tests showed that on average 30.6% of patients were at low risk for AMI, although this figure ranged from 3.8% to 73.5%. The pooled estimate of sensitivity was 98.7% (95% confidence interval [CI], 96.6% - 99.5%), and sensitivities of the cohorts ranged from 87.5% to 100% (P < .001).

Fourteen patients had AMI despite classification of low risk with hs-cTnT and ECG. However, 7 of the 14 took the blood test before 3 hours had elapsed since symptoms began, and 4 of them did so before 2 hours had elapsed.

With the high prevalence of chest pain, it is not surprising that a few patients will sustain an AMI despite designation of low risk after hs-cTnT and ECG, said Ryan Stanton, MD, an emergency medicine physician practicing at Baptist Health Richmond and Harrison Memorial Hospital in Lexington, Kentucky. "With something this large, 9200 people, 14 falling out is a pretty small number," he told Medscape Medical News. "Some of it is timing, but also people don't fit the mold. In medicine, with anything approaching 100% specificity, someone is going to fall out of the curve. Maybe they weren't having ischemia enough to call it cell death and spillage of troponin, but maybe the pain was still cardiac."

Secondary outcomes were assessed for eight of the studies. Of the 8059 patients, 21 had major adverse cardiac events. The pooled estimate of sensitivity for these events was 98.0% (95% CI, 94.7% - 99.3%), and sensitivities of the cohorts ranged from 87.9% to 100%.

 "Acute myocardial infarction may be ruled out in a substantial proportion of patients after only 1 blood draw," the researchers conclude.

Dr Stanton suggested an improvement over the single blood draw that his hospitals practice: two tests, 2 hours apart, to assess a quantitative result rather than a positive or negative result. "With high-sensitivity tests, we're looking more for change. Maybe a patient is still completely normal and underneath the threshold to be called MI, but if there's an upward change, that's worrisome, because the level of troponin shouldn't be changing. Conversely, people may have renal disease or a chronic condition in which they always have some troponin in the system, so looking at a second test to see if there is change is important for them," he said.

Limitations of the study include possible exclusion of some patients due to the chaos that can erupt in an ED, as well as variation in follow-up care after discharge and in the setup, calibration, and operation of analyzers in different laboratories. "In practice, expecting these assays to universally perform well and consistently at such low values is optimistic," the researchers write.

Dr Pickering has reported nonfinancial support from Roche Diagnostics. Several coauthors reported various financial and nonfinancial relationships with Abbott Diagnostics, Roche Diagnostics, Beckman-Coulter, Novartis, and other companies. Dr Stanton has disclosed no relevant financial relationships.

Ann Intern Med. Published online April 17, 2017. Abstract

For more news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.