Prognostication in emergency room patients: comparing ultrasensitive and contemporary quantification of cardiac troponin levels below the 99th percentile.

Abstract

Introduction: Cardiac troponin levels below the 99th percentile improve the predictive efficacy for cardiovascular events when associated with relevant clinical variables. However, whether ultra-sensitive analytical methods improve this predictive efficacy over less sensitive or contemporary analytical methods remains unknown.

Methods: This retrospective observational study involved consecutive patients who presented to the emergency department for suspected acute coronary syndrome and underwent measurement of ultra-sensitive cardiac troponin I (Singulex) and contemporary cardiac troponin I (Siemens) with levels below the 99th percentile. The clinical characteristics of these patients were analysed, and the efficacy of both analytical methods for predicting cardiovascular events over a 4-year follow-up period was compared.

Results: In total, 838 patients were analysed (mean age, 62.9 ± 16.6 years; 42.2% women). Their cumulative incidence of the composite cardiovascular event (death, readmission for myocardial infarction, and readmission for heart failure) was 25.9% over the 4-year follow-up. Both Singulex cardiac troponin I (analysed by quartiles) and Siemens cardiac troponin (analysed as detectable/undetectable) improved the predictive efficacy for the combined event over clinical variables [Harrell's C-index (95% confidence interval): 0.77 (0.74-0.80) vs. 0.79 (0.76-0.81) and 0.77 (0.74-0.80) vs. 0.78 (0.75-0.81), respectively; p = 0.018]. However, there were no statistically significant difference between the two predictive models that included the aforementioned troponin assays.

Conclusions: Detectable levels of cardiac troponin using a contemporary analytical method or those near the 99th percentile using an ultra-sensitive analytical method improve the predictive efficacy for cardiovascular events, with no differences between the two methods.