by Dr Geoffrey Modest
Two articles recently came out in the Annals of Internal Medicine which looked at simple and efficient ways to rule out acute myocardial infarction in patients with chest pain who go to the emergency room. This blog will deal with an article looking at a single measurement of high-sensitivity troponin (see doi:10.7326/M16-2562). Tomorrow, I will review an article looking at a somewhat more complex algorithm.
Details:
— 9241 patients who presented to the emergency dept with possible acute coronary syndrome were evaluated in this collaborative meta-analysis from 11 prospective cohort studies in Europe, New Zealand, and Australia. A publicly-funded study,
— 64% male, mean age 61
— Prevalence of acute MI range from 7-23% with an overall prevalence of 15%
— Study exclusion criteria were pretty consistent across the studies, but those with renal failure requiring dialysis were excluded in 3 of the studies and atypical presentations in one study
— 2 studies did not perform a 2nd troponin measurement on some low risk patients, but in general 2nd troponin levels were drawn for clinical care purposes and later outcome adjudication, at least 6 hours after symptom onset. 9 of the included studies were classified as having a high risk of bias due to reported nonconsecutive nonrandom patient selection (e.g. not recruiting patients 24 hours a day, 7 days a week, esp in some of the smaller hospitals where the resources were lacking) or exclusions due to missing data
–Overall 2825 patients (30.6%) were classified as low risk, defined as no new ischemia on ECG and high sensitivity cardiac troponin T (hs-cTnT) measurement below the limits of detection, <0.005 mcg/L
Results:
— 14 (0.5%) of the low-risk patients had an acute myocardial infarction during hospitalization (primary outcome), with the test performing at a sensitivity of 98.7% (96.6%-99.5%). In 7 of these 14 cases the time between symptom onset and blood sampling was < 3 hours (< 2 hours in 4 cases). The pooled negative predictive value was 99.3% (96.5%-100%)
— major adverse cardiac events (MACE) or death within 30 days (secondary outcome) occurred 21 times (including index admission for acute MI) after a negative index test result. Overall test sensitivity was 98.0% (94.7%-99.3%).
— A total of 126 (1.3%) of patients died within the 30 day follow-up. But no low-risk patients died.
Commentary:
— 10 to 20% of patients who present to the ED with suspected cardiac-related chest pain have an acute MI.
— Prior studies have shown that the hs-cTnT below the limits of detection reliably detects those patients who may be safely discharged from the ED for outpatient management, and this is been incorporated into European guidelines (eg NICE guidelines, updated 2016, state: “consider performing a single high-sensitivity troponin test only at presentation to rule out NSTEMI”). A few retrospective analyses have suggested the utility of a single hs-cTnT below detectable level in ruling out an acute MI, however the studies were considered to be methodologically flawed
— One advantage of the current study is that it was carried out in several different countries and with several different baseline patient cardiac risk and comorbidities, as well as differing prevalence of acute MI and the proportion of patients identified as low risk. This makes the conclusion more potentially generalizable.
— 50% of those in the low-risk group who actually had an acute MI after the negative troponin had their troponin level checked prior to 3 hours after symptom onset. This reinforces the recommendation of checking a 2nd troponin level in this group (ie, if checked <3 hours after symptom onset) if the first one were negative, as per the European guidelines.
— the specificity of hs-cTnT was poor, as expected, since increased troponin levels are not specific to an MI
— unfortunately, these researchers when unable to access patient-specific data to further elucidate the specifics of those patients who had false negative rates (beyond the too-early troponin testing).
–this hs-cTnT assay has been used in Europe for years, but was just approved for use in the US this year (I am not sure how available it is or how often it is used at this point).
So, this study and the one tomorrow raise the potential that in the not-so-distant future, low-risk patients may be able to be efficiently evaluated and discharged from the ED. And this hopefully would also apply to community-based settings as well. their email addresses, i can add them to the list