Optimizing Use of High-Sensitivity Troponin for Risk-Stratification of Acute Pulmonary Embolism.

Abstract

High-sensitivity troponin T (HS-TnT) may improve risk-stratification in hemodynamically stable acute pulmonary embolism (PE), but an optimal strategy for combining this biomarker with clinical risk-stratification tools has not been determined. We hypothesized that different HS-TnT cutoff values may be optimal for identifying (1) low-risk patients who may be eligible for outpatient management and (2) patients at increased risk of clinical deterioration who might benefit from advanced PE therapies. Retrospective analysis of hemodynamically stable patients in the University of Michigan acute ED-PE registry with available HS-TnT values. Primary and secondary outcomes were 30-day mortality and need for intensive care unit-level care. Receiver operating characteristic curves were used to determine optimal HS-TnT cutoffs in the entire cohort, and for those at higher risk based on the simplified Pulmonary Embolism Severity Index (PESI) or imaging findings. The optimal HS-TnT cutoff in the full cohort, 12 pg/mL, was significantly associated with 30-day mortality (odds ratio [OR]: 3.94, 95% confidence interval [CI]: 1.48-10.50) and remained a significant predictor after adjusting for the simplified PESI (sPESI) score and serum creatinine (adjusted OR: 3.05, 95% CI: 1.11-8.38). A HS-TnT cutoff of 87 pg/mL was associated with 30-day mortality (OR: 5.01, 95% CI: 2.08-12.06) in patients with sPESI ≥1 or right ventricular dysfunction. In this retrospective, single-center study of acute PE patients, we identified distinct optimal HS-TnT values for different clinical uses-a lower cutoff, which identified low-risk patients even in the absence of other risk-stratification methods, and a higher cutoff, which was strongly associated with adverse outcomes in patients at increased risk.