P4575UA and NSTEMI in the era of high-sensitivity Troponin: impact on patient risk profile and management

Abstract

Abstract Aim High-sensitivity troponin assays (HST) entered the clinical arena to facilitate exclusion of ACS in the emergency department (ER) in patients presenting with chest pain. Due to its higher sensitivity there is the potential for an overestimation of the diagnosis of NSTEMI, and possibly ACS overall. We assessed the impact of HST in the classification of ACS (NSTEMI vs UA) and its ability to predict obstructive coronary disease (CAD), in a population of pts referred to coronary angiography (ICA). Methods Retrospective analysis of 1844 pts with suspected NSTEMI or UA referred for ICA from a single ER between Feb 2013 and Nov 2018. Standard Troponin-I was used until Feb 2016 and HST thereafter. The characteristics of UA and NSTEMI pts before and after the introduction of HST were compared. Multivariate binary logistic regression models were used to access the association of different troponin assays with CAD (>50% for LM and >70% for the remaining). Sensitivity, specificity, NPV and PPV for angiographic CAD were also determined. Results The relative proportion of patients with UA and NSTEMI was similar between study periods: 31% vs 29% and 69% vs 71%, respectively (p=0.3). Clinical and angiographic characteristics did not differ in UA pts before and after the use of HST. NSTEMI patients in the HST era were less frequently women (39% vs 32%, p=0.026), had higher creatinine (0.93 IQR 0.75–1.3 vs 1.0 IQR 0.82–1.38), higher number of lesions (2 IQR 1–4 vs 3 IQR 1–4) and a lower rate of normal coronary arteries (10.5% vs 3.9%, p<0.001). The prevalence of significant CAD in this population, before and after HST, was 65% and 73%, respectively (p=0.001). However, when clinically relevant characteristics and judgement were accounted for, both standard troponin (OR 0.99, 95% CI 0.99–1.01) and HST (1.0, 95% CI 1.0–1.0) were poor predictors of significant CAD. Sensitivity was 69% vs 72%, specificity 30% vs 30%, PPV 65% vs 73% and NNP 34% vs 28%, respectively. Finally, rates of percutaneous intervention did not differ between the two periods (30% vs 33.5%, p=0.157), nor between ACS types within each period. CAD/PCI in ACS within each period Conclusion The introduction of HST did not result in an increase of the diagnosis of NSTEM vs UA, suggesting that clinical judgment remains an important determinant of the diagnosis of ACS. Also there was no evidence of an increase in PCI rates, despite worse CAD severity in NSTEMI patients.