Dayana Flores, Joan Walter, Desiree Wussler, Nikola Kozhuharov, Albina Nowak, Julia Dinort, Patrick Badertscher, Jasmin Martin, Zaid Sabti, Jeanne du Fay de Lavallaz, Thomas Nestelberger, Jasper Boeddinghaus, Tobias Zimmermann, Luca Koechlin, Bettina Glatz, Rafal Czmok, Eleni Michou, Danielle M Gualandro, Tobias Breidthardt and Christian Mueller
Objectives: The primary objective was to directly compare the prognostic accuracy of hs-cTnT versus hs-cTnI for the prediction of death in patient with pneumonia.
Methods: The prognostic accuracy of high sensitivity (hs)-cTnT and hs-cTnI was directly compared among patients presenting with dyspnea to the emergency department and centrally adjudicated by two independent experts to have pneumonia. Blood samples for the blinded measurement of hs-cTnT and hs-cTnI, as well as NTproBNP were obtained at ED presentation. CURB-65 was calculated as the multivariate risk score recommended in current guidelines. Primary endpoints were all-cause and cardiovascular (CV) mortality at 1 year.
Results: Among 306 patients, median age was 75 years, 38% were women, with extensive comorbidities including chronic obstructive pulmonary disease (COPD) in 41% and chronic heart failure (HF) in 26%, Cumulative 1-year all-cause mortality was 26.8% (82 deaths) and cumulative 1-year CV mortality was 9.5% (29 CV-deaths). While both hs-cTnT and hs-cTnI were independent predictors of death, the prognostic accuracy of hs-cTnT as quantified by the area under the curve (AUC) was significantly higher than hs-cTnI for 1-year all-cause mortality (AUC 0.73, 95%CI 0.66-0.779 vs. AUC 0.66, 95%CI 0.59-0.72; p=0.003) and CV-death (AUC 0.82, 95%CI 0.76-0.88 vs. 0.72, 95%CI 0.64-0.80; p=0.006), and comparable to NT-proBNP (AUC 0.72, 95%CI 0.59-0.72 and AUC 0.84, 95%CI 0.78-0.90 respectively, both p=ns). Compared to CURB-65 (AUC 0.60), the prognostic accuracy of hs-cTn was similar (hs-cTnI, p=0.463) or even higher (hs-cTnT, p=0.003).
Conclusions: Hs-cTnT has high prognostic accuracy and is superior to hs-cTnI in the prediction of all-cause and CV-mortality in patients with pneumonia.