Early, sensitive troponin I assessment safe, effective for acute MI diagnosis

LONDON — A new algorithm incorporating a more sensitive cutoff for troponin I values to identify non-STEMI would result in lower mortality and allow more rapid treatment or discharge of patients with acute chest pain, according to results from the BACC trial presented at the European Society of Cardiology Congress.

Dirk Westermann, MD, PhD, and colleagues evaluated 1,045 patients with acute chest pain indicative of MI who presented at a hospital in Germany. Troponin T levels were assessed at admission and again 3 hours later for diagnostic purposes, and troponin I levels were measured at 0, 1 and 3 hours after admission for the purposes of the study.

Patients were diagnosed with non-STEMI using the standard approach (n = 184), and a troponin I value higher than the 99th percentile, or 27 ng/L for the assay used (STAT high sensitive Troponin I, Abbott Diagnostics), served as the study cutoff. Follow-up was conducted for 6 months among those with non-STEMI and 793 patients discharged without an acute MI diagnosis. Patients with STEMI were excluded from analysis.

Westermann, of the University Heart Centre Hamburg, Germany, calculated the optimal troponin I cutoff as 6 ng/L or less at admission and after 1 hour, with a negative predictive value (NPV) of 99.7% for non-STEMI type 1. Using this cutoff to rule out all non-STEMI, they observed similar NPVs for troponin I measured 1 hour and 3 hours after admission (99% vs. 99.5%), with a sensitivity of 97.6% at 1 hour and 98.8% at 3 hours. Both NPVs were significantly higher than those calculated for a 27-ng/L cutoff (94.8% at 1 hour and 96.8% at 3 hours). The researchers noted that use of this algorithm would result in discharge of 402 of the 1,045 patients within 1 hour of admission.

A troponin I value greater than 6 ng/L and a change of 12 ng/L at 1 hour after admission were used to rule in non-STEMI for the study algorithm. Similar positive predictive values (PPVs) for non-STEMI were for measurement at 1 and 3 hours (87.1% vs. 84.6%, respectively), with specificity of 98% at 1 hour and 96.8% at 3 hours.

Westermann and colleagues validated the rule-out and rule-in algorithms for non-STEMI by applying them to two independent cohorts of 4,009 patients: APACE, which had evaluable troponin I data at 1 hour; and ADAPT, which had evaluable troponin I at 2 hours. They calculated an NPV of 99.2% in the APACE cohort and 99.7% in the ADAPT cohort for the rule-out algorithm, and a PPV of 80.4% and 81.5%, respectively, for the rule-in algorithm.

Mortality at 6 months was very low among patients in the BACC cohort who were ruled out for non-STEMI due to troponin I values of 6 ng/L or lower, Westermann said, whereas patients ruled in for non-STEMI according to the study algorithm had higher mortality rates. However, the highest mortality rates were observed for “grey zone” patients, or those with elevated but stable troponin I levels who were not identified by either the rule-in or rule-out algorithm.

Westermann and colleagues compared the two troponin I cutoffs in 74,738 participants without prevalent CVD enrolled in the BiomarCare study. Participants with troponin I values of 6 ng/L or lower had lower mortality rates than those with troponin I values of 27 ng/L or lower at both 1 and 5 years (P < .001 for both).

“A 1-hour algorithm is safe to rule out acute MI; there’s no need to wait for 3 hours,” Westermann concluded here. “A sensitive troponin cutoff of 6 ng/L performed better compared to the 99th percentile in view of lower follow-up mortality, [and] low troponin I values predict mortality in the general population. Further studies are needed to test the best cutoff for each troponin assay, and to validate a 1-hour algorithm prospectively.” – by Adam Taliercio

Reference:

Neumann JT, et al. Hot Line I – Acute MI. Presented at: European Society of Cardiology Congress; Aug. 29-Sept. 2, 2015; London.

Disclosure: Abbott Diagnostics provided funding for the study. Westermann reports receiving research support from Bayer and Novartis, as well as speaker/consultant honoraria from AstraZeneca, Bayer, Berlin Chemie, Biotronic, Orion and Novartis.