Time-pattern of adverse outcomes after an infection-triggered acute heart failure decompensation and the influence of early antibiotic administration and hospitalisation: results of the PAPRICA-3 Study

Miró, Òscar; Aguirre Tejedo, Alfons; Mebazaa, Alexandre

doi:http://dx.doi.org/10.1007/s00392-019-01481-3

Time-pattern of adverse outcomes after an infection-triggered acute heart failure decompensation and the influence of early antibiotic administration and hospitalisation: results of the PAPRICA-3 Study

Citation

Miró Ò, Takagi K, Gayat É, Gil V, Llorens P, Martín-Sánchez FJ. Et al. Time-pattern of adverse outcomes after an infection-triggered acute heart failure decompensation and the influence of early antibiotic administration and hospitalisation: results of the PAPRICA-3 Study. Clin Res Cardiol. 2020 Jan; 109(1):34-45. DOI: 10.1007/s00392-019-01481-3

Abstract

Objective: To investigate whether patients with an acute heart failure (AHF) episode triggered by infection present different outcomes compared to patients with no trigger and the effects of early antibiotic administration (EAA) and hospitalisation. Methods: Two groups were made according to the AHF trigger: infection (G1) or none identified (G2). The primary outcome was 13-week (91-days) all-cause mortality, and secondary outcomes were 13-week post-discharge mortality, readmission or combined endpoint. Comparisons are presented as unadjusted and adjusted (MEESSI risk score) hazard ratios (uHR/aHR) for G1 compared to G2 patients, also estimated by weeks. Stratified analysis by EAA (provided/not provided) and patient disposition (discharged/hospitalised) was performed. Results: We included 6727 patients (G1 = 3973; G2 = 2754). The 13-week mortality uHR was 1.11 (0.99-1.25; p = 0.06; with significant increases in the first 3 weeks), and the aHR was 0.91 (0.81-1.02; p = 0.11). There were no differences in unadjusted secondary post-discharge outcomes; however, G1 outcomes significantly improved after adjustment: aHR 0.83 (0.71-0.96; p = 0.01) for mortality, 0.92 (0.84-0.99; p = 0.04) for readmission, and 0.92 (0.85-0.99; p = 0.04) for the combined endpoint. We found a differentiated effect of hospitalisation (p < 0.05 for interaction; better post-discharge readmission and combined outcomes in G1), and a trend (p = 0.06) to lower mortality in G1 patients with EAA. Additionally, there were some differences between groups in baseline and acute episode characteristics. Conclusion: AHF triggered by infection is not associated with a higher mid-term mortality and has better post-discharge outcomes; however, the first 3 weeks are an extremely vulnerable period. Since hospitalisation could have a role in limiting adverse post-discharge events, and EAA in reducing mortality, these relationships should be prospectively explored in further studies.