Objective: To compare the clinical characteristics and outcomes of patients with acute heart failure (AHF) according to clinical profiles based on congestion and perfusion determined in the emergency department (ED). Methods and results: Overall, 11 261 unselected AHF patients from 41 Spanish EDs were classified according to perfusion (normoperfusion = warm; hypoperfusion = cold) and congestion (not = dry; yes = wet). Baseline and decompensation characteristics were recorded as were the main wards to which patients were admitted. The primary outcome was 1-year all-cause mortality; secondary outcomes were need for hospitalisation during the index AHF event, in-hospital all-cause mortality, prolonged hospitalisation, 7-day post-discharge ED revisit for AHF and 30-day post-discharge rehospitalisation for AHF. A total of 8558 patients (76.0%) were warm + wet, 1929 (17.1%) cold + wet, 675 (6.0%) warm + dry, and 99 (0.9%) cold + dry; hypoperfused (cold) patients were more frequently admitted to intensive care units and geriatrics departments, and warm + wet patients were discharged home without admission. The four phenotypes differed in most of the baseline and decompensation characteristics. The 1-year mortality was 30.8%, and compared to warm + dry, the adjusted hazard ratios were significantly increased for cold + wet (1.660; 95% confidence interval 1.400–1.968) and cold + dry (1.672; 95% confidence interval 1.189–2.351). Hypoperfused (cold) phenotypes also showed higher rates of index episode hospitalisation and in-hospital mortality, while congestive (wet) phenotypes had a higher risk of prolonged hospitalisation but decreased risk of rehospitalisation. No differences were observed among phenotypes in ED revisit risk. Conclusions: Bedside clinical evaluation of congestion and perfusion of AHF patients upon ED arrival and classification according to phenotypic profiles proposed by the latest European Society of Cardiology guidelines provide useful complementary information and help to rapidly predict patient outcomes shortly after ED patient arrival.

Aims Dipeptidyl peptidase 3 (DPP3) is a protease involved in the degradation of cardiovascular mediators. Its administration has been shown to be associated with impaired cardiac contraction and kidney haemodynamics while its inhibition restored cardiac contraction in a pre‐clinical model of severe heart failure in mice. Circulating DPP3 (cDPP3) was found to be elevated in shock. The present study aims to assess the association between cDPP3 and worsening haemodynamics, namely refractory shock, in a cohort of cardiogenic shock (CS). Methods and results This is an ancillary study of OptimaCC, a prospective, double‐blind, multicentre, randomized study assessing efficacy and safety of catecholamines in 57 patients with CS after acute myocardial infarction. cDPP3 was measured in plasma at inclusion, 24 h, 48 h, and 72 h, and haemodynamic and biological parameters were recorded at inclusion. cDPP3 values were higher in refractory CS than non‐refractory CS at inclusion (median [interquartile range]; 76.1 [37.9–238.7] ng/mL vs. 32.8 [23.9–47.6] ng/mL, P = 0.014), at 24 h (P < 0.001) and up to 48 h (P = 0.027). Furthermore, cDPP3 at inclusion discriminated CS patients who did develop refractory shock vs. non‐refractory with an area under the curve of 0.73 (95% confidence interval [CI] 0.55–0.92). The high cDPP3 group (cDPP3 ≥59.1 ng/mL) at inclusion had a higher Simplified Acute Physiology Score II (SAPS II), lower cardiac index and lower estimated glomerular filtration rate. More importantly, in CS patients with high cDPP3 at inclusion, those who rapidly decreased cDPP3 at 24 h exhibited a striking reduction in the occurrence of refractory shock and death. Conclusion In CS patients, cDPP3 gives an early prediction of outcome, including development of refractory status and/or survival. Clinical Trial Registration: clinicaltrials.gov Identifier NCT01367743