The European-derived high-sensitivity cardiac troponin I assay (hs-cTnI) was found to have high sensitivity and specificity for the ruling out and ruling in, respectively, of acute myocardial infarction (MI) in patients visiting emergency departments (EDs) in the United States, according to study results published in the Annals of Emergency Medicine.
Although the use of rapid hs-cTnI assays for the diagnosis of MI have been validated in other populations globally, their diagnostic accuracy remains uncertain in the United States, with ED clinicians wary of leveraging these tests for this purpose. The investigators hypothesized that implementing these algorithms in EDs in the United States would be associated with excellent negative predictive values (NPVs) and acceptable positive predictive values (PPVs) within 1 hour.
In this multicenter, prospective High-Sensitivity Cardiac Troponin I in the United States trial, 2346 patients (median age, 56 years; all patients aged ≥22 years; 56.0% men) presenting to the ED with suspected acute MI were enrolled. Study participants were tested with the hs-cTnI assay between April 2015 and April 2016. Core laboratories performed the Siemens Healthineers Atellica hs-cTnI immunoassay on plasma samples at baseline, 40 to 90 minutes, and 2 to 3 hours (0/1-hour algorithm and 0/2- to 3-hour algorithm, respectively). Patients were divided into rule-out, rule-in, and continued observation groups, based on results of the hs-cTnI assay. The diagnosis of acute MI was independently adjudicated by assessors blinded to the assay results.
There were 2113 participants who were eligible for analysis with the 0/1-hour algorithm, of whom 1065 (50.4%) were ruled out within 1 hour with a 98.7% sensitivity (95% CI, 96.3%-99.7%) and a 99.7% negative predictive value (NPV; 95% CI, 99.2%-99.9%), and 265 (12.5%) were ruled in, with a 95.7% specificity (95% CI, 94.7%-96.6%) and a 69.4% positive predictive value (PPV; 95% CI, 63.5%-74.9%). The remainder of the cohort (n=783; 37.1%) was categorized in the “under continued evaluation” group. The incidence of acute MI was 5.6% (95% CI, 4.2%-7.5%).
Of the 1916 participants eligible for analysis of the 0/2- to 3-hour algorithm, 1066 (55.6%) were ruled out within 2 to 3 hours, with 99.1% sensitivity (95% CI, 96.8%-99.9%) and a 99.8% NPV (95% CI, 99.3%-100.0%). There were 254 patients (13.3%) who were ruled in during this period, with 95.5% specificity (95% CI, 94.4%-96.4%) and a 69.7% PPV (95% CI, 63.6%-75.3%). A total of 596 patients (31.1%) remained in the continued observation group, not having met the rule-in or rule-out cutoffs. There was no significant difference in terms of sensitivity, specificity, NPV, and PPV between the algorithm groups.
Of the 2335 patients (99.5%) with available data on adverse events at 30 days, 24 (1.0%) had all-cause mortality, 12 (0.5%) had acute MI after discharge, 32 (1.4%) had postdischarge coronary revascularization, and 50 (2.1%) were hospitalized for heart failure. The risks for death or acute MI postdischarge were 0.2%, 2.1%, and 4.8% in the rule-out, continued evaluation, and rule-in groups, respectively.
Study limitations include a possible limited applicability to nonwhite or nonblack populations, possible underestimations of PPV and rule-in rates, a lack of serial troponin measurements in some patients, and the lack of an adequate baseline blood draw for some patients.
“The rapid rule-in of acute myocardial infarction with hs-cTnI assays requires more studies in the United States to improve the positive predictive values and specificities,” cautioned the authors.
Nowak RM, Christenson RH, Jacobsen G, et al. Performance of novel high-sensitivity cardiac troponin I assays for 0/1-hour and 0/2- to 3-hour evaluations for acute myocardial infarction: results from the HIGH-US study [published online February 8, 2020]. Ann Emerg Med. doi:10.1016/j.annemergmed.2019.12.008