Comprehensive characterization of non-cardiac comorbidities in acute heart failure: an analysis of ESC-HFA EURObservational Research Programme Heart Failure Long-Term Registry.

AIMS: To evaluate the prevalence and associations of non-cardiac comorbidities (NCCs) with in-hospital and post-discharge outcomes in acute heart failure (AHF) across the ejection fraction (EF) spectrum. METHODS AND RESULTS: The 9326 AHF patients from European Society of Cardiology (ESC)-Heart Failure Association (HFA)-EURObservational Research Programme Heart Failure Long-Term Registry had complete information for the following 12 NCCs: anaemia, chronic obstructive pulmonary disease (COPD), diabetes, depression, hepatic dysfunction, renal dysfunction, malignancy, Parkinson's disease, peripheral vascular disease (PVD), rheumatoid arthritis, sleep apnoea, and stroke/transient ischaemic attack (TIA). Patients were classified by number of NCCs (0, 1, 2, 3, and ≥4). Of the AHF patients, 20.5% had no NCC, 28.5% had 1 NCC, 23.1% had 2 NCC, 15.4% had 3 NCC, and 12.5% had ≥4 NCC. In-hospital and post-discharge mortality increased with number of NCCs from 3.0% and 18.5% for 1 NCC to 12.5% and 36% for ≥4 NCCs.Anaemia, COPD, PVD, sleep apnoea, rheumatoid arthritis, stroke/TIA, Parkinson, and depression were more prevalent in HF with preserved EF (HFpEF). The hazard ratio (95% confidence interval) for post-discharge death for each NCC was for anaemia 1.6 (1.4-1.8), diabetes 1.2 (1.1-1.4), kidney dysfunction 1.7 (1.5-1.9), COPD 1.4 (1.2-1.5), PVD 1.2 (1.1-1.4), stroke/TIA 1.3 (1.1-1.5), depression 1.2 (1.0-1.5), hepatic dysfunction 2.1 (1.8-2.5), malignancy 1.5 (1.2-1.8), sleep apnoea 1.2 (0.9-1.7), rheumatoid arthritis 1.5 (1.1-2.1), and Parkinson 1.4 (0.9-2.1). Anaemia, kidney dysfunction, COPD, and diabetes were associated with post-discharge mortality in all EF categories, PVD, stroke/TIA, and depression only in HF with reduced EF, and sleep apnoea and malignancy only in HFpEF. CONCLUSION: Multiple NCCs conferred poor in-hospital and post-discharge outcomes. Ejection fraction categories had different prevalence and risk profile associated with individual NCCs.

The current analysis from ESC-Heart Failure Long-Term Registry represents the largest and most comprehensive study in an acute heart failure (AHF) population with HF with reduced ejection fraction (HFrEF), HF with mildly reduced EF (HFmrEF), and HF with preserved EF (HFpEF), on prevalence and association with in-hospital and post-discharge outcomes of a large number of non-cardiac comorbidities.A greater number of non-cardiac comorbidities (CNNs) were associated at admission with older age, preserved EF, more severe NYHA class, and longer duration of HF. In-hospital and post-discharge mortality gradually increased with number of CNNs.The association between each individual comorbidity and post-discharge outcomes varied substantially in AHF patients with HFrEF, HFmrEF, and HFpEF, suggesting that an 'EF-specific' multidisciplinary approach with distinct comorbidity management programs should be applied in post-discharge phase.

Deep neural networks for ECG-free cardiac phase and end-diastolic frame detection on coronary angiographies.

Invasive coronary angiography (ICA) is the gold standard in Coronary Artery Disease (CAD) imaging. Detection of the end-diastolic frame (EDF) and, in general, cardiac phase detection on each temporal frame of a coronary angiography acquisition is of significant importance for the anatomical and non-invasive functional assessment of CAD. This task is generally performed via manual frame selection or semi-automated selection based on simultaneously acquired ECG signals - thus introducing the requirement of simultaneous ECG recordings. In this paper, we evaluate the performance of a purely image based workflow relying on deep neural networks for fully automated cardiac phase and EDF detection on coronary angiographies. A first deep neural network (DNN), trained to detect coronary arteries, is employed to preselect a subset of frames in which coronary arteries are well visible. A second DNN predicts cardiac phase labels for each frame. Only in the training and evaluation phases for the second DNN, ECG signals are used to provide ground truth labels for each angiographic frame. The networks were trained on 56,655 coronary angiographies from 6820 patients and evaluated on 20,780 coronary angiographies from 6261 patients. No exclusion criteria related to patient state (stable or acute CAD), previous interventions (PCI or CABG), or pathology were formulated. Cardiac phase detection had an accuracy of 98.8 %, a sensitivity of 99.3 % and a specificity of 97.6 % on the evaluation set. EDF prediction had a precision of 98.4 % and a recall of 97.9 %. Several sub-group analyses were performed, indicating that the cardiac phase detection performance is largely independent from acquisition angles, the heart rate of the patient, and the angiographic view (LCA / RCA). The average execution time of cardiac phase detection for one angiographic series was on average less than five seconds on a standard workstation. We conclude that the proposed image based workflow potentially obviates the need for manual frame selection and ECG acquisition, representing a relevant step towards automated CAD assessment.