Deep Learning Models for Predicting Left Heart Abnormalities From Single-Lead Electrocardiogram for the Development of Wearable Devices.

BACKGROUND: Left heart abnormalities are risk factors for heart failure. However, echocardiography is not always available. Electrocardiograms (ECGs), which are now available from wearable devices, have the potential to detect these abnormalities. Nevertheless, whether a model can detect left heart abnormalities from single Lead I ECG data remains unclear.Methods and Results: We developed Lead I ECG models to detect low ejection fraction (EF), wall motion abnormality, left ventricular hypertrophy (LVH), left ventricular dilatation, and left atrial dilatation. We used a dataset comprising 229,439 paired sets of ECG and echocardiography data from 8 facilities, and validated the model using external verification with data from 2 facilities. The area under the receiver operating characteristic curves of our model was 0.913 for low EF, 0.832 for wall motion abnormality, 0.797 for LVH, 0.838 for left ventricular dilatation, and 0.802 for left atrial dilatation. In interpretation tests with 12 cardiologists, the accuracy of the model was 78.3% for low EF and 68.3% for LVH. Compared with cardiologists who read the 12-lead ECGs, the model's performance was superior for LVH and similar for low EF. CONCLUSIONS: From a multicenter study dataset, we developed models to predict left heart abnormalities using Lead I on the ECG. The Lead I ECG models show superior or equivalent performance to cardiologists using 12-lead ECGs.

Diuretic and renal effects of angiotensin receptor-neprilysin inhibitor in patients hospitalized for acute heart failure.

In patients hospitalized for acute decompensation of heart failure (HF), the impact of angiotensin receptor-neprilysin inhibitor (ARNI) on diuresis and renal function has not been fully investigated. Patients with HF and reduced ejection fraction who were hospitalized for acute decompensation and newly initiated ARNI after hemodynamic stabilization were enrolled. Changes in urine volume (UV), body weight, estimated glomerular filtration rate (eGFR), and urine N-acetyl-beta-d-glucosaminidase (uNAG) levels before and after ARNI initiation were investigated. Changes in the diuretic response [DR, calculated as urine volume/(intravenous furosemide volume/40 mg)], N-terminal pro-brain natriuretic peptide (NT-proBNP), hematocrit, and plasma volume (PV) were also evaluated. A total of 60 patients were enrolled. ARNI was initiated at a median of 6 [5, 7] days after hospitalization. After initiation of ARNI, body weight, NT-proBNP, and PV decreased. UV and DR increased only on the day of ARNI initiation (delta UV 400 ± 957 ml and delta DR 1100 ± 3107 ml/40 mg furosemide) and then decreased to baseline levels. In the multivariable linear regression analysis, younger age, higher BMI, and higher NT-proBNP levels were significantly associated with greater UV after ARNI initiation. eGFR and uNAG did not significantly change after the initiation of ARNI [delta eGFR -1.7 ± 12.0 mL/min/1.73 m2 and delta uNAG 2.0 (-5.6, 6.9) IU/L]. In patients hospitalized for HF, the initiation of ARNI was associated with a small and transient increase in UV and DR, and was not associated with worsening of renal function or tubular injury.

Malnutrition, hemodynamics and inflammation in heart failure with reduced, mildly reduced and preserved ejection fraction.

In patients with heart failure (HF) with reduced ejection fraction (HFrEF), malnutrition can be associated with intestinal congestion and systemic inflammation. These relationships have not been fully investigated in HF with mildly reduced EF (HFmrEF) and with preserved EF (HFpEF). We analyzed 420 patients with HF who underwent right heart catheterization. The relationships between hemodynamic parameters, C-reactive protein, and the controlling nutritional (CONUT) score were investigated in HFrEF, HFmrEF and HFpEF. The CONUT score of all patients was 2 [1, 4] (median [interquartile range]), and was not significantly different between the left ventricular EF (LVEF) categories (2 [1, 3] for HFrEF, 2 [1, 3] for HFmrEF, and 3 [1, 4] for HFpEF, p = 0.279). In multivariate linear regression analyses, there was a significant association between CRP and the CONUT score in HFmrEF and HFpEF, while brain natriuretic peptide and right atrial pressure were significantly associated with the CONUT score in HFrEF. Higher CONUT scores predicted a higher incidence of the composite endpoint of death or HF hospitalization within 12 months without an interaction with LVEF (p = 0.980). The CONUT score was an independent predictor of the composite endpoint, death, and HF hospitalization after adjustment for confounders in the multivariate analysis. In conclusion, inflammation was associated with malnutrition in HFmrEF and HFpEF, while congestion was an independent predictor of malnutrition in HFrEF. Malnutrition predicted worse outcomes regardless of LVEF.

Impact of Dapagliflozin on the Renal Function and Damage in Patients with Heart Failure with a Reduced Ejection Fraction.

Objective Whether or not the initial dip in the glomerular filtration rate (GFR) after the initiation of sodium-glucose co-transporter 2 inhibitors (SGLT2is) is associated with renal tubular injury in patients with heart failure with a reduced ejection fraction (HFrEF) is unclear. We therefore investigated the relationship between changes in the estimated GFR (eGFR) and urine N-acetyl-ß-D-glucosaminidase (uNAG) after the initiation of dapagliflozin in patients with HFrEF. Methods We prospectively investigated 89 patients with HFrEF who were newly started on dapagliflozin 10 mg/day. Changes in the eGFR and uNAG-to-creatinine ratio (uNAG/Cre) were evaluated at 2 weeks and 2 months after the initiation of dapagliflozin. Results The eGFR was decreased at 2 weeks but had not declined further by 2 months. The uNAG/Cre was increased at 2 weeks but had not increased further by 2 months. There was no correlation between the changes in the eGFR and uNAG/Cre (r=-0.022, p=0.853 at 2 weeks and r=0.078, p=0.538 at 2 months). The relative change in the systolic blood pressure, hematocrit, plasma volume, and N-terminal pro-brain natriuretic peptide (NT-proBNP) were correlated with the relative change in the eGFR. In a multiple linear regression analysis, the relative change in the eGFR at 2 weeks was significantly associated with NT-proBNP, and the relative change in the uNAG/Cre was significantly associated with the use of loop diuretics and the relative change in urine osmolality at 2 weeks. Conclusion A transient decrease in the eGFR after the initiation of dapagliflozin in patients with HFrEF was not generally associated with renal tubular injury and might have been the result of hemodynamic alteration.

A long-term follow-up of necrobiotic xanthogranuloma with concomitant large-vessel vasculitis and heart failure with reduced ejection fraction: a case report.

Background: Necrobiotic xanthogranuloma (NXG) is a non-Langerhans cell histiocytosis and multisystem disorder. Low level of HDL cholesterol associated with a systemic inflammatory profile, which may result from the interaction of monoclonal immunoglobulin and lipoproteins, is a characteristic feature. There is no evidence of NXG-associated large-vessel vasculitis, nor are there any established treatments, although chemotherapy for comorbid multiple myeloma is most often administered. Case summary: We describe a case of a 53-year-old male with a first history of heart failure with impaired systolic function. He presented with orbital xanthomas and multiple subcutaneous nodules, and laboratory examination showed elevated levels of C-reactive protein, low HDL, and paraproteinemia. A constellation of these clinical features and pathological findings of skin biopsy led to the diagnosis of NXG. 18F-Fluorodeoxyglucose positron emission tomography (PET)/computed tomography (CT) confirmed increased uptake in the aorta and bilateral common carotid arteries. He began prednisolone treatment with reference to treatment for large-vessel vasculitis. After the treatment, C-reactive protein immediately decreased with markedly increased levels of apolipoprotein A1 (Apo-A1) and HDL. Systolic dysfunction was restored at 6-month follow-up. The patient has not experienced heart failure for 5 years after treatment, and the follow-up PET/CT demonstrated resolution of vascular inflammation. Discussion: This is the first report of NXG-associated large-vessel vasculitis. Low-dose prednisolone may benefit for NXG-associated vasculitis and cardiomyopathy. HDL, Apo-A1, and C-reactive protein levels may be useful for monitoring the activity of NXG, and PET/CT was a valuable diagnostic tool for NXG-associated vasculitis.

Guillain-Barré syndrome after percutaneous coronary intervention and transcatheter aortic valve implantation.

Guillain-Barré syndrome (GBS) usually develops after preceding infection, but cardiac surgery can also occasionally cause GBS. Currently, cardiac catheterizations have already become common therapeutic options for heart diseases, but there have been no reports of GBS occurrence after that. Herein, we present a rare case in which GBS occurred following catheterization. An 85-year-old-man with sudden onset chest pain was rushed to our hospital and diagnosed with ST-elevated myocardial infarction. He underwent emergent percutaneous coronary intervention (PCI) to left anterior descending artery, but he still had exertional chest pain. Echocardiography revealed severe aortic stenosis (AS) and our heart team considered AS was the cause of symptom and decided to perform and transcatheter aortic valve implantation (TAVI), 11 days after the PCI. However, 5 days after the TAVI procedure, he presented with symmetrical muscular weakness of extremities. Cranial magnetic resonance imaging showed no significant lesion. Based on several signs including albuminocytologic dissociation in cerebrospinal fluid examination, demyelinating polyneuropathy in nerve conduction study, positive anti-ganglioside antibody, and the lack of preceding infection, he was diagnosed with GBS triggered by cardiac catheterizations. We administered high-dose intravenous immunoglobulin therapy and his motor strength gradually improved, finally discharged with full motor strength after 7 months rehabilitation. Learning objective: •Cardiac surgery has been already reported as a non-infectious risk factor of Guillain-Barré syndrome (GBS) in previous literatures, and cardiac catheterization such as percutaneous coronary intervention and transcatheter aortic valve implantation, which were relatively less invasive procedure, may be a potential risk factor for GBS occurrence as well.•If a patient complains of progressive, symmetrical neurological symptoms after cardiac catheterization, GBS should be considered as the possible cause, and nerve conduction study and cerebrospinal fluid examination may be helpful for the diagnosis.

An explainable artificial intelligence-enabled electrocardiogram analysis model for the classification of reduced left ventricular function.

Aims: The black box nature of artificial intelligence (AI) hinders the development of interpretable AI models that are applicable in clinical practice. We aimed to develop an AI model for classifying patients of reduced left ventricular ejection fraction (LVEF) from 12-lead electrocardiograms (ECG) with the decision-interpretability. Methods and results: We acquired paired ECG and echocardiography datasets from the central and co-operative institutions. For the central institution dataset, a random forest model was trained to identify patients with reduced LVEF among 29 907 ECGs. Shapley additive explanations were applied to 7196 ECGs. To extract the model's decision criteria, the calculated Shapley additive explanations values were clustered for 192 non-paced rhythm patients in which reduced LVEF was predicted. Although the extracted criteria were different for each cluster, these criteria generally comprised a combination of six ECG findings: negative T-wave inversion in I/V5-6 leads, low voltage in I/II/V4-6 leads, Q wave in V3-6 leads, ventricular activation time prolongation in I/V5-6 leads, S-wave prolongation in V2-3 leads, and corrected QT interval prolongation. Similarly, for the co-operative institution dataset, the extracted criteria comprised a combination of the same six ECG findings. Furthermore, the accuracy of seven cardiologists' ECG readings improved significantly after watching a video explaining the interpretation of these criteria (before, 62.9% ± 3.9% vs. after, 73.9% ± 2.4%; P = 0.02). Conclusion: We visually interpreted the model's decision criteria to evaluate its validity, thereby developing a model that provided the decision-interpretability required for clinical application.

Acute Myocarditis in a Patient Following mRNA-1273 SARS-CoV-2 Vaccination.

We herein report a case of acute myocarditis possibly related to the second dose of an mRNA-coronavirus disease 2019 vaccine in a 45-year-old woman with no remarkable medical history. She had a fever for one week following the second dose of the mRNA-1273 severe acute respiratory syndrome coronavirus 2 vaccine. One week later, she presented with chest pain and electrocardiogram changes. Her serum troponin levels were elevated upon admission. Echocardiography showed segmental wall motion abnormalities of the apex, apical portion of the anterior and inferior walls. The findings of cardiac magnetic resonance imaging were consistent with acute myocarditis.

The role of the renal resistance index in patients with heart failure with reduced or preserved ejection fraction.

BACKGROUND: Renal impairment is a common phenomenon that portends a poor prognosis of heart failure (HF). The renal arterial resistance index (RRI) can be useful for defining renal function and predicting outcomes in patients with HF. This study aimed to investigate the determining factors of the RRI in HF patients with preserved ejection fraction (HFpEF) and with reduced EF (HFrEF). METHODS: This retrospective study included 330 patients with HF. We investigated the determining factors for the RRI and the association between the RRI and 1-year composite outcome, comprising all-cause mortality and re-hospitalization for HF. RESULTS: The independent predictors of the RRI were tricuspid regurgitation peak gradient and estimated glomerular filtration rate in HFpEF, and pulse pressure and blood urea nitrogen in HFrEF. During the follow-up, 30 (9.1%) patients presented the composite outcome. Cox proportional hazard analysis revealed the association of the RRI with the composite outcome in both HFrEF (HR 1.08; 95% CI 1.03-1.14) and HFpEF (HR 1.07; 95% CI 1.03-1.12) without an interaction (p for interaction = 0.770). CONCLUSIONS: The RRI was a consistent prognosticator in patients with HFpEF and those with HFrEF, while factors defining RRI were different between these groups.