Johan Engdahl

Purpose: To study left atrial stiffness index (LASI) and left atrial activation time (LA-AT) changes in an older population with silent paroxysmal atrial fibrillation (PAF) compared to age-matched control. Methods: Individuals with newly diagnosed PAF and a randomly selected age-matched control group without AF were enrolled from STROKESTOP2-study. All participants had NT-proBNP levels 125–900 ng/L. TTE examinations were analyzed retrospectively offline using dedicated software. Besides conventional echocardiographic measurements, LA-AT was obtained by measuring the time delay between the onset of the P-wave on ECG and tissue Doppler Á peak wave in lateral LA wall. LASI was calculated as the ratio of left ventricular filling (E/é) to peak atrial longitudinal strain (PALS). Results: Mean age was 76 ± 0.3 years. The PAF group ( n = 67) presented significant increase of LASI (0.41 (0.29–0.60) vs. 0.32 (0.29–0.43), P = 0.001) and prolonged LA-AT (170 (140–180) ms vs. 140 (120–144) ms, P < 0.001) compared to controls ( n = 94). Subgroup analysis in participants with normal LA volume index (LAVI) ≤ 34 ml/m ² , LASI and LA-AT remained significantly increased/prolonged in PAF ( N = 27) compared to controls ( N = 72). NT-proBNP-levels did not differ between PAF (228 (181–348) ng/L) and controls (216 (156–317) ng/L) ( P = 0.18). ROC curve analysis identified thresholds LASI > 0.54 and LA-AT > 139 msec for discrimination of silent PAF. Conclusion: Indices reflecting LA stiffness and atrial conduction delay were significantly elevated in recently diagnosed PAF patients. LASI and LA-AT thresholds might non-invasively detect early atrial remodeling without structural enlargement in elderly patients with high risk of PAF and mildly increased NT-proBNP levels.

Backgrounds and Aims: Atrial fibrillation (AF) screening may reduce adverse outcomes through early initiation of preventive treatment. However, more data are warranted to inform the yield of different screening strategies. This study aimed to evaluate how patient characteristics and screening methodologies influence screening yield. Methods: We conducted an individual participant data meta-analysis of 12 European and North American study cohorts, by using generalized linear mixed models with random intercept at study level. Results: In this meta-analysis of 34,165 participants screened for AF (age 73.4 ± 7.5 years and 53.5% female), the screening yield increased with more intensive approaches, ranging from 0.83% [95% confidence interval (CI): 0.24-2.87] in single-timepoint screening, to 5.06% [95% CI: 3.55-7.17] in intermittent (i.e. repeated single-timepoint) screening, 4.81% [95% CI: 3.79-6.08] in short-term continuous monitoring (≤30 days), and up to 30.39% [95% CI: 28.06-32.84] in long-term continuous monitoring (>30 days). Advanced age, male sex, hypertension, and heart failure (HF) were associated with detection. Significant interactions were observed between screening approach and age, sex, HF, CHADS2, and NT-proBNP. These risk factors had a stronger influence on AF detection with single-timepoint screening, compared with more intensive methods. Conclusions: In this meta-analysis, AF screening yield varied substantially by patient characteristics and screening approach. Advanced age, male sex, and HF were independently associated with greater AF detection, particularly in those having single-timepoint screening. Our findings support tailoring the screening approach to the individual target population in clinical practice. Further studies to assess clinical outcomes following screening are needed and underway.

Aims: Atrial fibrillation is associated with increased risks of stroke, heart failure, and death. While stroke prevention is a key focus, heart failure is more frequent and a major contributor to mortality in atrial fibrillation. This study aimed to assess the incidence and timing of heart failure in screening-detected atrial fibrillation. Methods and results: This post hoc analysis included participants without prior heart failure from the STROKESTOP and STROKESTOP II randomized screening studies, in which individuals aged 75–76 years were invited to ECG-based atrial fibrillation screening. Incident heart failure diagnoses were obtained from national registries. Incidence rates were calculated per 100 person-years, and Cox regression was used to estimate hazard ratios (HRs). In STROKESTOP, 23% of participants with screening-detected atrial fibrillation developed heart failure (3.76 per 100 person-years), compared with 4.43 in patients with known atrial fibrillation and 1.08 in patients without atrial fibrillation. Screening-detected atrial fibrillation was associated with a three-fold higher heart failure risk vs. no atrial fibrillation (adjusted HR 3.19) and a risk comparable to known atrial fibrillation (adjusted HR 2.86). In STROKESTOP II, 20% developed heart failure (4.19 per 100 person-years), compared with 3.52 in known atrial fibrillation and 0.93 without atrial fibrillation. Screening-detected atrial fibrillation was associated with a four-fold higher heart failure risk vs. no atrial fibrillation (adjusted HR 4.73) and a higher risk than known atrial fibrillation (adjusted HR 2.59). Conclusion: Screening-detected atrial fibrillation is associated with a high risk of heart failure, comparable to known atrial fibrillation, and is not a benign condition.

Aims: Conventional screening modalities for atrial fibrillation (AF) are often constrained by considerable resource and logistical demands. The randomized controlled NORwegian atrial fibrillation self-SCREENing (NORSCREEN) trial employs a fully digital approach to recruitment and implementation. The aim of this pre-specified analysis was to evaluate the effectiveness of the fully digital recruitment and screening strategy, determine the prevalence of screen-detected AF, and assess initial adherence to therapeutic recommendations. Methods and results: Individuals aged ≥65 years with ≥1 additional risk factor for stroke were digitally recruited from the general Norwegian population and randomized 1:1 to screening or usual care. Exclusion criteria included prior AF, anticoagulation, implantable cardiac electronic devices, or lack of smartphone access. Participants allocated to screening were mailed a patch-based electrocardiogram sensor for site-less continuous monitoring over 7 days. Those with diagnosed AF were advised to contact their general practitioner to initiate appropriate therapy. A total of 551 555 Norwegian residents were digitally invited; of these, 139 596 (25.3%) completed the digital eligibility assessment, 74 956 (53.7%) were eligible, and 50 549 (67.4%) were randomized to screening or usual care. The median age was 75 years (inter-quartile range: 69–78), 23 445 (46.4%) were female, and the mean CHA2DS2-VA risk score was 2.6 ± 0.9. In the screening group, 19 723 (78.1%) completed the study procedures. Previously undiagnosed AF was identified in 382 (1.9%) screened participants, of whom 366 (95.8%) initiated anticoagulation therapy. Conclusion: The NORSCREEN trial demonstrates that fully digital, large-scale, site-less AF screening is feasible and effective in an at-risk population, identifying previously undiagnosed AF in 1.9% of screened individuals and showing high adherence to therapeutic recommendations. Clinical Trial Registration: NCT05914883

Background: Atrial fibrillation is a common arrhythmia in patients with ischaemic heart disease. New-onset atrial fibrillation after coronary revascularisation is associated with adverse cardiovascular outcomes. This study aimed to determine the long-term cumulative incidence of new-onset atrial fibrillation after percutaneous coronary intervention or coronary artery bypass grafting surgery. Methods: A prospective observational cohort study in a real-world population setting, conducted at three tertiary centres, on new-onset atrial fibrillation incidence after percutaneous coronary intervention (N=123) or coronary artery bypass grafting (N=123). Heart rhythm was monitored the first 30 days in hospital by telemetry and on discharge using a handheld thumb ECG device three times a day, and thereafter for 2-week periods at 3, 12 and 24 months. The primary endpoint was the cumulative incidence of new-onset atrial fibrillation 24 months after the index procedure. Secondary objectives were to describe the incidence of cerebral ischaemic stroke and bleeding, myocardial infarction and major bleeding events during 24 months follow-up. Results: Mean age was 67 years, and male sex was more prevalent. At 30 days, the cumulative incidence of atrial fibrillation was 56% (69/123) in the coronary artery bypass graft group and 2% (3/123) in the percutaneous coronary intervention group. At 24 months, the cumulative incidence of atrial fibrillation was 58% (71/123) in the coronary artery bypass graft group and 6% (7/123) in the percutaneous coronary intervention group. Stroke, myocardial infarction and major bleeding were infrequent during follow-up. Conclusion: Over 24 months of follow-up, incident new-onset atrial fibrillation mainly occurred during the first 30 days after coronary artery bypass grafting but was more evenly distributed during 24 months after percutaneous coronary intervention. Trial registration number: NCT04307225 .

Background and aims: Atrial fibrillation (AF), often asymptomatic and underdiagnosed, is an independent risk factor for ischemic stroke. A knowledge gap remains regarding the optimal target population, and method to use for AF screening. We aimed to test whether screening for AF using a machine learning-based risk-prediction model (RPM) and 14-day continuous patch electrocardiogram (ECG) (Philips ePatch) in high-risk individuals ≥65 years is more effective than standard care. Methods: Individuals ≥65 years were assigned to general or RPM cohort. The general cohort was randomised to control or invitation. In the RPM cohort, high-risk individuals, identified by RPM, were randomised to control or invitation. The primary outcome was six-month AF incidence, analysed as intention-to-invite, comparing RPM + invitation with general + control. Results: Of 2,960 randomised individuals, participation was 43% (632/1480) in invitation arms. AF incidence was higher in RPM + invitation than in general + control arm (3.8%, 28/740 versus 0.7%, 5/740; p<0.001), yielding a risk ratio of 5.6, (95% confidence interval [2.2, 14.4]), and a number needed to invite of 32. AF was more often detected in RPM + invitation than in general + invitation arm (1.1%, 8/740; p<0.001), but not more often than in RPM + control arm (2.2%, 16/740; p=0.07). No difference was found between general + invitation and general + control arms (1.1%, 8/740 versus 0.7%, 5/740; p=0.40). Conclusion: Among high-risk individuals ≥65 years, the combination of a machine learning-based risk-prediction model and long-term ECG recording was superior to standard care in identifying new AF cases.

Background/Objectives: To assess the association between moderate N-terminal natriuretic peptide (NT-proBNP) and cardiac alterations and prognosis in septuagenarians without heart failure (HF). Methods: From the STROKESTOP II screening study, 230 individuals aged 75/76 years with NT-proBNP < 900 ng/L were randomly selected. Subjects with persistent atrial fibrillation (AF), more than mild valvular disease, or HF were excluded. Echocardiography was performed. NT-proBNP ≥ 125 ng/L and paroxysmal AF (pAF) on thumb ECG were used as grouping variables. Participants were followed up during a median of 5 years for cardiovascular mortality, HF, AF, and cerebrovascular events. Cox regression analysis was employed for prognostic assessment. Results: Three groups were identified: SR ≥ 125 (n = 94, no pAF and NT-proBNP ≥ 125 ng/L), pAF (n = 77, pAF and NT-proBNP ≥ 125 ng/L), and controls (n = 30, no pAF and NT-proBNP < 125 ng/L). NT-proBNP was not associated with structural (left atrial volume and left ventricular (LV) mass) or functional (E/e’, LV strain) alterations in any group (p > 0.05). Cardiovascular risk factors (HR: 4.6; CI = 1.7–12.3; p = 0.002), but not NT-proBNP (HR: 1.9; CI = 0.7–5.1; p = 0.2), entailed a prognostic value for the composite endpoint of HF, AF, and cardiovascular death. Conclusions: In septuagenarians without HF, modest NT-proBNP elevation was not associated with echocardiographic changes or prognosis

Aims: The aim of this study was to perform an external validation of an automatic machine learning (ML) algorithm for heart rhythm diagnostics using smartphone photoplethysmography (PPG) recorded by patients with atrial fibrillation (AF) and atrial flutter (AFL) pericardioversion in an unsupervised ambulatory setting. Methods and results: Patients undergoing cardioversion for AF or AFL performed 1-min heart rhythm recordings pericardioversion at least twice daily for 4–6 weeks, using an iPhone 7 smartphone running a PPG application (CORAI Heart Monitor) simultaneously with a single-lead electrocardiogram (ECG) recording (KardiaMobile). The algorithm uses support vector machines to classify heart rhythm from smartphone-PPG. The algorithm was trained on PPG recordings made by patients in a separate cardioversion cohort. Photoplethysmography recordings in the external validation cohort were analysed by the algorithm. Diagnostic performance was calculated by comparing the heart rhythm classification output to the diagnosis from the simultaneous ECG recordings (gold standard). In total, 460 patients performed 34 097 simultaneous PPG and ECG recordings, divided into 180 patients with 16 092 recordings in the training cohort and 280 patients with 18 005 recordings in the external validation cohort. Algorithmic classification of the PPG recordings in the external validation cohort diagnosed AF with sensitivity, specificity, and accuracy of 99.7%, 99.7% and 99.7%, respectively, and AF/AFL with sensitivity, specificity, and accuracy of 99.3%, 99.1% and 99.2%, respectively. Conclusion: A machine learning-based algorithm demonstrated excellent performance in diagnosing atrial fibrillation and atrial flutter from smartphone-PPG recordings in an unsupervised ambulatory setting, minimizing the need for manual review and ECG verification, in elderly cardioversion populations. Clinical Trial Registration: Clinicaltrials.gov, NCT04300270

Background: Atrial fibrillation (AF) is a common but often undiagnosed condition, increasing the risk of stroke and heart failure. Early detection is crucial, yet traditional methods struggle with AF’s transient nature. This study investigates how augmenting ECG data with heart rate variability (HRV) and demographic data (age and sex) can improve AF detection. Methods: We analysed 35 634 12-lead ECG recordings from three public databases (China Physiological Signal Challenge-Extra, PTB-XL and Georgia), each with physician-validated AF labels. A range of convolutional neural network models, including AlexNet, VGG-16, ResNet and transformers, were tested for AF prediction, enriched with HRV and demographic data to explore the effectiveness of the multimodal approach. Each data modality (ECG, HRV and demographic) was assessed for its contribution to model performance using fivefold cross-validation. Performance improvements were evaluated across key metrics, and saliency maps were generated to provide further insights into model behaviour and identify critical features in AF detection. Results: Integrating HRV and demographic data with ECG substantially improved performance. AlexNet and VGG-16 outperformed more complex models, achieving AUROC of 0.9617 (95% CI 0.95 to 0.97) and 0.9668 (95% CI 0.96 to 0.97), respectively. Adding HRV data showed the most significant improvement in sensitivity, with AlexNet increasing from 0.9117 to 0.9225 and VGG-16 from 0.9216 to 0.9225. Combining both HRV and demographic data led to further improvements, with AlexNet achieving a sensitivity of 0.9225 (up from 0.9192 with HRV) and VGG-16 reaching 0.9113 (up from 0.9097 with HRV). The combination of HRV and demographic data resulted in the highest gains in sensitivity and area under the receiver operating characteristic curve. Saliency maps confirmed the models identified key AF features, such as the absence of the P-wave, validating the multimodal approach. Conclusions: AlexNet and VGG-16 excelled in AF detection, with HRV data improving sensitivity, and demographic data providing additional benefits. These results highlight the potential of multimodal approaches, pending further clinical validation.