Diana My Frodi

Aims: Patterns in physical behaviour may be associated with an increased risk of ventricular arrhythmia. We examined associations between temporal dynamics in physical behaviour measured through wearable technology and the risk of ventricular arrhythmia Methods and results: The multicentre, prospective SAFEHEART study recruited 303 patients with an implantable cardioverter-defibrillator (ICD) from May 2021 to September 2022. Continuous physical behaviour data were collected over 365 consecutive days and nights using wearable accelerometers. We characterized each 28-day behavioural time series using statistical summary indices and deep representations learned by neural networks. Logistic regression analyses were used to estimate the associations between physical behaviour and the subsequent risk of ventricular arrhythmia as detected by ICD and treated appropriately. Predictive performance was assessed using k-fold cross-validation and quantified by the area under the receiver operating characteristic curve (AUROC). From the SAFEHEART cohort, 277 patients and 56 ventricular arrhythmia were analysed, contributing to a total of 64 995 days of behavioural data. Reduced numbers of daily physical activity bouts and low day-to-day variation in sleep characteristics were associated with an increased risk of an arrhythmic event. Deep representations improved the predictive performance, with an AUROC of 0.74 ± 0.05, compared with statistical summary indices with an AUROC 0.67 ± 0.14 (P = 0.05 for comparison). Conclusion: Our findings suggest that behavioural patterns captured through wearable technology are independently associated with ventricular arrhythmia, particularly when these are characterised by deep representations.

Aims: Physical activity has shown association with ventricular arrhythmia, however, the role of specific behavioral patterns over a 24 h cycle remains unknown. Therefore, we aimed to explore associations between physical behavior and appropriate implantable cardioverter defibrillator (ICD) therapy. Methods and results: We included patients with an ICD at two European sites, who wore wrist-based accelerometers capturing 24 h movement and sleep behaviours for 28 days. Behavioural measures included activity volume, duration and intensity, sleep duration, and efficiency. Participants were followed for 12 months for the outcome of appropriate ICD therapy. Cox proportional hazard models with restricted cubic splines were used for the analysis. Lastly, the predictive capacity was tested. A total of 253 ICD patients were included (mean age 63.5 (±10.2), 48 (19.0%) female). During follow-up, 40 participants (15.8%) received appropriate ICD therapy; 32 anti-tachycardia pacing (ATP) only (12.6%), 5 shock only (2.0%), and 3 combined ATP and shock (1.2%). In the adjusted model, high inactive duration (HR 1.40 (95% 1.10–1.78)), peak walking cadence (HR 1.07 (95% 1.03–1.12)), and total sleep duration (HR 1.50 (1.02–2.22)) were associated with the outcome. The dose–response relationship was U-shaped for inactive duration with a cut-off at 16 h, and linear for peak cadence and sleep. The prediction model reached an area under the receiver operating characteristic curve of 0.70 ± 0.03, with highest accuracy in the first months. Conclusion: Wearable-derived 24 h movement and sleep behaviours collected over 28 days were associated with later appropriate ICD therapy risk. Testing of the predictive value of digital biomarkers for enhanced risk stratification of ventricular arrhythmia warrants larger prospective studies. Clinical Trial Registration: National Trial Registration (NL9218, http://onderzoekmetmensen.nl/).

We aimed to identify and characterise behavioural profiles in patients at high risk of SCD, by using deep representation learning of day-to-day behavioural recordings. We present a pipeline that employed unsupervised clustering on low-dimensional representations of behavioural time-series data learned by a convolutional residual variational neural network (ResNet-VAE). Data from the prospective, observational SafeHeart study conducted at two large tertiary university centers in the Netherlands and Denmark were used. Patients received an implantable cardioverter-defibrillator (ICD) between May 2021 and September 2022 and wore wearable devices using accelerometer technology during 180 consecutive days. A total of 272 patients (mean age of 63.1 ± 10.2 years, 81% male) were eligible with a total sampling of 37,478 days of behavioural data (138 ± 47 days per patient). Deep representation learning identified five distinct behavioural profiles: Cluster A (n = 46) had very low physical activity levels and a disturbed sleep pattern. Cluster B (n = 70) had high activity levels, mainly at light-to-moderate intensity. Cluster C (n = 63) exhibited a high-intensity activity profile. Cluster D (n = 51) showed above-average sleep efficiency. Cluster E (n = 42) had frequent waking episodes and poor sleep. Annual risks of malignant ventricular arrhythmias ranged from 30.4% in Cluster A to 9.8% and 9.5% for Clusters D-E, respectively. Compared to low-risk profiles (D-E), Cluster A demonstrated a three-to-four fold increased risk of malignant ventricular arrhythmias adjusted for clinical covariates (adjusted HR 3.63, 95% CI 1.54–8.53, p < 0.001). These behavioural profiles may guide more personalised approaches to ventricular arrhythmia and SCD prevention.

Context: Subclinical thyroid dysfunction is a marker for atrial fibrillation (AF) and stroke risk. Objective: This study explored the effects of AF screening according to thyroid-stimulating hormone (TSH) levels. Methods: An AF screening trial (the LOOP study) was analyzed post hoc according to baseline TSH. The primary outcome was stroke or systemic embolism (SE). Secondary outcomes included major bleeding, all-cause death, and the combination of stroke, SE, and cardiovascular death. Results: TSH measurements were available in 6003 of 6004 trial participants, 1500 randomized to implantable loop recorder (ILR) screening for AF and anticoagulation upon detection vs 4503 to usual care; mean age was 74.7 ± 4.1 years and 2836 (47%) were women. AF detection was approximately triple for ILR vs usual care across TSH tertiles (adjusted P interaction = 0.44). In the first tertile, screening was associated with decreased risk of the primary outcome (hazard ratio [HR] 0.52, 95% CI 0.30-0.90; P = .02) and stroke, SE, or cardiovascular death (HR 0.54, 95% CI 0.34-0.84; P = .006) compared with usual care, while no effect was observed among participants with higher TSH (adjusted P interaction .03 and .01, respectively). There was no effect on other outcomes. Analyses of continuous TSH or excluding those with abnormal TSH or thyroid medication showed similar results. Conclusion: AF screening and subsequent treatment was associated with decreased stroke risk among participants with low TSH, though the yield of screening was similar across TSH levels. TSH may be useful as a marker to indicate benefit from AF screening vs overdiagnosis and overtreatment. These findings should be considered exploratory and warrant further study.

Background: Advances in medical treatment and outcomes in implantable cardioverter-defibrillator (ICD) recipients incentivize a need for improved candidate selection and identification of risk factors for ICD therapy. We examined contemporary rates of and risk factors for ICD therapy. Methods: Patients with ICD for primary (PP) or secondary prevention (SP), implanted between January 2010 and December 2020, were followed for appropriate and inappropriate incident and recurrent shock. Results: Overall, 2998 patients (mean age 61.8 ± 12.7 years, 20% female, 73% ICD carriers, and 47.1% SP) were analyzed with a median follow-up of 4.3 (interquartile range (IQR) 2.1–7.4) years. A total of 426/2998 (14.2%) patients had shock; 364/2998 (12.1%) had appropriate and 82/2998 (2.7%) inappropriate shock, with annualized event rates of 2.34 (2.11–2.59) and 0.49 (0.39–0.61) per 100 person-years, respectively. Of those with shock, 133/364 (36.5%) experienced recurrent appropriate shock and 8/364 (2.2%) received recurrent inappropriate shock, with event rates of 10.57 (8.85–12.53) and 0.46 (0.20–0.92), respectively. In multivariable analyses, female sex was associated with a reduced risk of incident appropriate shock (hazard ratio 0.69 [95% confidence interval 0.52; 0.91]). Of other variables, only revascularization status was associated with recurrent appropriate shock in PP, and CRT-D with recurrent appropriate shock in the overall cohort. Conclusion: One in eight ICD recipients received appropriate shock 2–7 years after guideline-directed implantation. More than one-third of patients with a first shock experienced recurrent shock. Few clinical variables showed potential in predicting shocks, illustrating a need for more advanced tools to select candidates for implantation. Graphical

Aims: Wearable health technologies are increasingly popular. Yet, wearable monitoring only works when devices are worn as intended, and adherence reporting lacks standardization. In this study, we aimed to explore the long-term adherence to a wrist-worn activity tracker in the prospective SafeHeart study and identify patient characteristics associated with adherence. Methods and results: This study enrolled 303 participants, instructed to wear a wrist-worn accelerometer day and night for 6 months. Long-term adherence was defined as valid days (≥22 h of wear time) divided by expected days, and daily adherence as mean hours of wear time per 24 h period. Optimal, moderate, and low long-term and daily adherence groups were defined as long-term adherence above or below 95 and 75% and daily adherence above or below 90 and 75%. Regression models were used to identify patient characteristics associated with long-term adherence. In total, 296 participants [median age 64 years; interquartile range (IQR) 57–72; 19% female] were found eligible, yielding 44 003 days for analysis. The median long-term adherence was 88.2% (IQR 74.6–96.5%). A total of 83 (28%), 127 (42.9%), and 86 (29.1%) participants had optimal, moderate, and low long-term adherence, and 163 (55.1%), 87 (29.4%), and 46 (15.5%) had optimal, moderate, and low daily adherence, respectively. Age and smoking habits differed significantly between adherence levels, and increasing changeover intervals improved the degree of long-term adherence. Conclusion: Long-term adherence to a wearable activity tracker was 88.2% over a 6-month period. Older age and longer changeover interval were positively associated with long-term adherence. This serves as a benchmark for future studies that rely on wearable devices. Trial registration number: The National Trial Registration number: NL9218 (https://onderzoekmetmensen.nl/).

Introduction: Patients at high risk of sudden cardiac death are treated with an implantable cardioverter-defibrillator (ICD). Patterns in physical behaviour may be prognostic for clinical deterioration. Deep embedded clustering (DEC) employs deep neural networks to learn compact and meaningful representations from high-dimensional data, on which an unsupervised clustering then operates. Purpose: The objectives of this study were to use DEC to identify and describe behavioural profiles in a prospective ICD cohort and assess their risk of clinical outcomes. Methods: Data from the multicenter, prospective SafeHeart study conducted at our 2 centers, was used. Patients received an implantable cardioverter defibrillator (ICD) between May 2021 and September 2022, who then wore wearable devices during 6 months that capture physical behaviour. Latent representations were extracted from the time-series behavioural data using a variational autoencoder, and used as input to an unsupervised machine learning algorithm (Figure 1). Kaplan-Meier survival curves were used to visualise the annual risk of appropriate ICD-therapy, and the composite of all ICD-therapy and death. Results: A total of 272 patients (mean age of 63.1 ±10.2 years, 81% male) were eligible, with a total of 37478 days of behavioural data (138 ±47 days per patient). DEC identified five distinct behavioural profiles: Cluster A (n=70) had high activity levels, mainly at light-to-moderate intensity. Cluster B (n=51) showed above-average sleep efficiency. Cluster C (n=63) exhibited a high-intensity activity profile. Cluster D (n=42) had frequent waking episodes and poor sleep. Cluster E (n=46) had very low physical activity levels and an impaired sleep architecture. Significantly differences were observed between clusters A-E in ICD-therapy and mortality (log-rank p-value 0.04) (Figure 2). Clusters B and D exhibited a threefold decrease in the annual risk of appropriate ICD therapy compared to cluster E. Conclusion: DEC reveals distinct behavioural profiles in a prospective ICD cohort, with their inherent risk on outcomes. Wearable accelerometer devices enable behavioural monitoring that may provide opportunities for improved personalised risk prediction in a notoriously heterogeneous high-risk patient population.Deep embedded clustering of behaviourEvent rates stratified for clusters

Aims: Patient-reported outcome measures (PROMs) serve multiple purposes, including shared decision-making and patient communication, treatment monitoring, and health technology assessment. Patient monitoring using PROMs is constrained by recall and non-response bias, respondent burden, and missing data. We evaluated the potential of behavioural digital biomarkers obtained from a wearable accelerometer to achieve personalized predictions of PROMs. Methods and results: Data from the multicentre, prospective SafeHeart study conducted at Amsterdam University Medical Center in the Netherlands and Copenhagen University Hospital, Rigshospitalet in Copenhagen, Denmark, were used. The study enrolled patients with an implantable cardioverter defibrillator between May 2021 and September 2022 who then wore wearable devices with raw acceleration output to capture digital biomarkers reflecting physical behaviour. To collect PROMs, patients received the Kansas City Cardiomyopathy Questionnaire (KCCQ) and EuroQoL 5-Dimensions 5-Level (EQ5D-5L) questionnaire at two instances: baseline and after six months. Multivariable Tobit regression models were used to explore associations between digital biomarkers and PROMs, specifically whether digital biomarkers could enable PROM prediction. The study population consisted of 303 patients (mean age 62.9 ± 10.9 years, 81.2% male). Digital biomarkers showed significant correlations to patient-reported physical and social limitations, severity and frequency of symptoms, and quality of life. Prospective validation of the Tobit models indicated moderate correlations between the observed and predicted scores for KCCQ [concordance correlation coefficient (CCC) = 0.49, mean difference: 1.07 points] and EQ5D-5L (CCC = 0.38, mean difference: 0.02 points). Conclusion: Wearable digital biomarkers correlate with PROMs, and may be leveraged for real-time prediction. These findings hold promise for monitoring of PROMs through wearable accelerometers.