European heart journal. Digital health最新文献

Aims: Wearable devices are transforming the electrocardiogram (ECG) into a ubiquitous medical test. This study assesses the association between premature ventricular and atrial contractions (PVCs and PACs) detected on wearable-format ECGs (15 s single lead) and cardiovascular outcomes in individuals without cardiovascular disease (CVD).

Methods and results: Premature atrial contractions and PVCs were identified in 15 s single-lead ECGs from N = 54 016 UK Biobank participants (median age, interquartile range, age 58, 50-63 years, 54% female). Cox regression models adjusted for traditional risk factors were used to determine associations with atrial fibrillation (AF), heart failure (HF), myocardial infarction (MI), stroke, life-threatening ventricular arrhythmias (LTVAs), and mortality over a period of 11.5 (11.4-11.7) years. The strongest associations were found between PVCs (prevalence 2.2%) and HF (hazard ratio, HR, 95% confidence interval = 2.09, 1.58-2.78) and between PACs (prevalence 1.9%) and AF (HR = 2.52, 2.11-3.01), with shorter prematurity further increasing risk. Premature ventricular contractions and PACs were also associated with LTVA (P < 0.05). Associations with MI, stroke, and mortality were significant only in unadjusted models. In a separate UK Biobank sub-study sample [UKB-2, N = 29,324, age 64, 58-60 years, 54% female, follow-up 3.5 (2.6-4.8) years] used for independent validation, after adjusting for risk factors, PACs were associated with AF (HR = 1.80, 1.12-2.89) and PVCs with HF (HR = 2.32, 1.28-4.22).

Conclusion: In middle-aged individuals without CVD, premature contractions identified in 15 s single-lead ECGs are strongly associated with an increased risk of AF and HF. These data warrant further investigation to assess the role of wearable ECGs for early cardiovascular risk stratification.

Aims: We aimed to assess longer-term results (accessibility, hospital admissions, and mortality) in elderly patients referred to a cardiology department (CD) from primary care using e-consultation in outpatient care.

Methods and results: We included 9963 patients >80 years from 1 January 2010 to 31 December 2019. Until 2012, all patients attended an in-person consultation (2010-2012). In 2013, we instituted an e-consult programme (2013-2019) for all primary care referrals to cardiologists that preceded a patient's in-person consultation when considered. We used an interrupted time series (ITS) regression approach to investigate the impact of e-consultation on (i) cardiovascular hospital admissions and mortality. We also analysed (ii) the total number and referral rate (population-adjusted referred rate) in both periods, and (iii) the accessibility was measured as the number of consultations and variation according to the distance from the municipality and reference hospital. During e-consultation, the demand for care increased (12.8 ± 4.3% vs. 25.5 ± 11.1% per 1000 inhabitants, P < 0.001) and referrals from different areas were equalized. After the implementation of e-consultation, we observed that the increase in hospital admissions and mortality were stabilized [incidence rate ratio (iRR): 1.351 (95% CI, 0.787, 2.317), P = 0.874] and [iRR: 1.925 (95% CI: 0.889, 4.168), P = 0.096], respectively. The geographic variabilities in hospital admissions and mortality seen during the in-person consultation were stabilized after e-consultation implementation.

Conclusions: Implementation of a clinician-to-clinician e-consultation programme in outpatient care was associated with improved accessibility to cardiology healthcare in elderly patients. After e-consultations were implemented, hospital admissions and mortality were stabilized.

Aims: Current non-invasive screening methods for cardiac allograft rejection have shown limited discrimination and are yet to be broadly integrated into heart transplant care. Given electrocardiogram (ECG) changes have been reported with severe cardiac allograft rejection, this study aimed to develop a deep-learning model, a form of artificial intelligence, to detect allograft rejection using the 12-lead ECG (AI-ECG).

Methods and results: Heart transplant recipients were identified across three Mayo Clinic sites between 1998 and 2021. Twelve-lead digital ECG data and endomyocardial biopsy results were extracted from medical records. Allograft rejection was defined as moderate or severe acute cellular rejection (ACR) based on International Society for Heart and Lung Transplantation guidelines. The extracted data (7590 unique ECG-biopsy pairs, belonging to 1427 patients) was partitioned into training (80%), validation (10%), and test sets (10%) such that each patient was included in only one partition. Model performance metrics were based on the test set (n = 140 patients; 758 ECG-biopsy pairs). The AI-ECG detected ACR with an area under the receiver operating curve (AUC) of 0.84 [95% confidence interval (CI): 0.78-0.90] and 95% (19/20; 95% CI: 75-100%) sensitivity. A prospective proof-of-concept screening study (n = 56; 97 ECG-biopsy pairs) showed the AI-ECG detected ACR with AUC = 0.78 (95% CI: 0.61-0.96) and 100% (2/2; 95% CI: 16-100%) sensitivity.

Conclusion: An AI-ECG model is effective for detection of moderate-to-severe ACR in heart transplant recipients. Our findings could improve transplant care by providing a rapid, non-invasive, and potentially remote screening option for cardiac allograft function.

Aims: Remote monitoring (RM) is the standard of care for follow up of patients with cardiac implantable electronic devices. The aim of this study was to compare smartphone-based RM (SM-RM) using patient applications (myMerlinPulse™ app) with traditional bedside monitor RM (BM-RM).

Methods and results: The retrospective study included de-identified US patients who received either SM-RM or BM-RM capable of implantable cardioverter defibrillators or cardiac resynchronization therapy defibrillators (Abbott, USA). Patients in SM-RM and BM-RM groups were propensity-score matched on age and gender, device type, implant year, and month. Compliance with RM was quantified as the proportion of patients enrolling in the RM system (Merlin.net™) and transmitting data at least once. Connectivity was measured by the median number of days between consecutive transmissions per patient. Of the initial 9714 patients with SM-RM and 26 679 patients with BM-RM, 9397 patients from each group were matched. Remote monitoring compliance was higher in SM-RM; significantly more patients with SM-RM were enrolled in RM compared with BM-RM (94.4 vs. 85.0%, P < 0.001), similar number of patients in the SM-RM group paired their device (95.1 vs. 95.0%, P = 0.77), but more SM-RM patients transmitted at least once (98.1 vs. 94.3%, P < 0.001). Connectivity was significantly higher in the SM-RM, with patients transmitting data every 1.2 (1.1, 1.7) vs. every 1.7 (1.5, 2.0) days with BM-RM (P < 0.001) and remained better over time. Significantly more SM-RM patients utilized patient-initiated transmissions compared with BM-RM (55.6 vs. 28.1%, P < 0.001).

Conclusion: In this large real-world study, patients with SM-RM demonstrated improved compliance and connectivity compared with BM-RM.

Aims: Heart failure with preserved ejection fraction (HFpEF) is associated with stiffened myocardium and elevated filling pressure that may be captured by heart sound (HS). We investigated the relationship between phonocardiography (PCG) and echocardiography in symptomatic patients suspected of HFpEF.

Methods and results: Consecutive symptomatic patients with sinus rhythm and left ventricular ejection fraction >45% were enrolled. Echocardiography was performed to evaluate the patients' diastolic function, accompanied by PCG measurements. Phonocardiography features including HS amplitude, frequency, and timing intervals were calculated, and their abilities to differentiate the ratio between early mitral inflow velocity and early diastolic mitral annular velocity (E/e') were investigated. Of 45 patients, variable ratio matching was applied to obtain two groups of patients with similar characteristics but different E/e'. Patients with a higher E/e' showed higher first and second HS frequencies and more fourth HS and longer systolic time intervals. The interval from QRS onset to first HS was the best feature for the prediction of E/e' > 9 [area under the curve (AUC): 0.72 (0.51-0.88)] in the matched patients. In comparison, N-terminal pro-brain natriuretic peptide (NT-proBNP) showed an AUC of 0.67 (0.46-0.85), a value not better than any PCG feature (P > 0.05).

Conclusion: Phonocardiography features stratify E/e' in symptomatic patients suspected of HFpEF with a diagnostic performance similar to NT-proBNP. Heart sound may serve as a simple non-invasive tool for evaluating HFpEF patients.

Aims: Deep learning models (DLMs) have shown superiority in electrocardiogram (ECG) analysis and have been applied to diagnose dyskalaemias. However, no study has explored the performance of DLM-enabled ECG in continuous follow-up scenarios. Therefore, we proposed a dynamic revision of DLM-enabled ECG to use personal pre-annotated ECGs to enhance the accuracy in patients with multiple visits.

Methods and results: We retrospectively collected 168 450 ECGs with corresponding serum potassium (K⁺) levels from 103 091 patients as development samples. In the internal/external validation sets, the numbers of ECGs with corresponding K⁺ were 37 246/47 604 from 13 555/20 058 patients. Our dynamic revision method showed better performance than the traditional direct prediction for diagnosing hypokalaemia [area under the receiver operating characteristic curve (AUC) = 0.730/0.720-0.788/0.778] and hyperkalaemia (AUC = 0.884/0.888-0.915/0.908) in patients with multiple visits.

Conclusion: Our method has shown a distinguishable improvement in DLMs for diagnosing dyskalaemias in patients with multiple visits, and we also proved its application in ejection fraction prediction, which could further improve daily clinical practice.

Aims: Secondary prevention reduces coronary heart disease (CHD) progression. Traditional prevention programs including cardiac rehabilitation are under-accessed, which smartphone apps may overcome. To evaluate the effect of a game-based mobile app intervention (MyHeartMate) to improve cardiovascular risk factors and lifestyle behaviours.

Methods and results: Single-blind randomized trial of CHD patients in Sydney, 2017-2021. Intervention group were provided the MyHeartMate app for 6 months. Co-designed features included an avatar of the patient's heart and tokens earned by risk factor work (tracking, challenges, and quizzes). The control group received usual care. Primary outcome was self-reported physical activity [metabolic equivalents (METs), Global Physical Activity Questionnaire] and secondary outcomes included lipid levels, blood pressure (BP), body mass index, and smoking. Pre-specified sample size was achieved (n = 390), age 61.2 ± 11.5 years; 82.5% men and 9.2% current smokers. At 6 months, adjusted for baseline levels, the intervention group achieved more physical activity than control (median difference 329 MET mins/wk), which was not statistically significant (95% CI -37.4, 696; P = 0.064). No differences occurred between groups on secondary outcomes except for lower triglyceride levels in the intervention [mean difference -0.3 (95% CI -0.5, -0.1 mmoL/L, P = 0.004)]. Acceptability was high: 94.8% of intervention participants engaged by tracking exercise or BP and completing missions; 26.8% continued to engage for ≥30 days. Participants (n = 14) reported the app supported tracking behaviours and risk factors, reinforcing and improving self-care confidence, and decreasing anxiety.

Conclusion: A game-based app proved highly acceptable for patients with CHD but did not improve risk factors or lifestyle behaviours other than triglyceride levels.

Aims: One of the major challenges in the quantification of myocardial blood flow (MBF) from stress perfusion cardiac magnetic resonance (CMR) is the estimation of the arterial input function (AIF). This is due to the non-linear relationship between the concentration of gadolinium and the MR signal, which leads to signal saturation. In this work, we show that a deep learning model can be trained to predict the unsaturated AIF from standard images, using the reference dual-sequence acquisition AIFs (DS-AIFs) for training.

Methods and results: A 1D U-Net was trained, to take the saturated AIF from the standard images as input and predict the unsaturated AIF, using the data from 201 patients from centre 1 and a test set comprised of both an independent cohort of consecutive patients from centre 1 and an external cohort of patients from centre 2 (n = 44). Fully-automated MBF was compared between the DS-AIF and AI-AIF methods using the Mann-Whitney U test and Bland-Altman analysis. There was no statistical difference between the MBF quantified with the DS-AIF [2.77 mL/min/g (1.08)] and predicted with the AI-AIF (2.79 mL/min/g (1.08), P = 0.33. Bland-Altman analysis shows minimal bias between the DS-AIF and AI-AIF methods for quantitative MBF (bias of -0.11 mL/min/g). Additionally, the MBF diagnosis classification of the AI-AIF matched the DS-AIF in 669/704 (95%) of myocardial segments.

Conclusion: Quantification of stress perfusion CMR is feasible with a single-sequence acquisition and a single contrast injection using an AI-based correction of the AIF.