Reducing the burden of inconclusive smart device single-lead ECG tracings via a novel artificial intelligence algorithm.

Background: Multiple smart devices capable of automatically detecting atrial fibrillation (AF) based on single-lead electrocardiograms (SL-ECG) are presently available. The rate of inconclusive tracings by manufacturers' algorithms is currently too high to be clinically useful. Method: This is a prospective, observational study enrolling patients presenting to a cardiology service at a tertiary referral center. We assessed the clinical value of applying a smart device artificial intelligence (AI)-based algorithm for detecting AF from 4 commercially available smart devices (AliveCor KardiaMobile, Apple Watch 6, Fitbit Sense, and Samsung Galaxy Watch3). Patients underwent a nearly simultaneous 12-lead ECG and 4 smart device SL-ECGs. The novel AI algorithm (PulseAI, Belfast, United Kingdom) was compared with each manufacturer's algorithm. Results: We enrolled 206 patients (31% female, median age 64 years). AF was present in 60 patients (29%). Sensitivity and specificity for the detection of AF by the novel AI algorithm vs manufacturer algorithm were 88% vs 81% (P = .34) and 97% vs 77% (P < .001) for the AliveCor KardiaMobile, 86% vs 81% (P = .45) and 95% vs 83% (P < .001) for the Apple Watch 6, 91% vs 67% (P < .01) and 94% vs 82% (P < .001) for the Fitbit Sense, and 86% vs 82% (P = .63) and 94% vs 80% (P < .001) for the Samsung Galaxy Watch3, respectively. In addition, the proportion of SL-ECGs with an inconclusive diagnosis (1.2%) was significantly lower for all smart devices using the AI-based algorithm compared to manufacturer's algorithms (14%-17%), P < .001. Conclusion: A novel AI algorithm reduced the rate of inconclusive SL-ECG diagnosis massively while maintaining sensitivity and improving the specificity compared to the manufacturers' algorithms.

Left atrial posterior wall isolation using pulsed-field ablation: procedural characteristics, safety, and mid-term outcomes.

BACKGROUND: Non-pulmonary vein (PV) ablation targets such as posterior wall isolation (PWI) have been tested in patients with persistent atrial fibrillation (AF). Pulsed-field ablation (PFA) offers a novel ablation technology possibly able to overcome the obstacles of incomplete PWI and concerns of damage to adjacent structures compared to thermal energy sources. Our aim was to assess procedural characteristics, safety, and mid-term outcomes of patients undergoing PWI using PFA in a clinical setting. METHODS: Patients undergoing PFA-PVI with PWI were included. First-pass isolation was controlled using a multipolar mapping catheter. RESULTS: One hundred consecutive patients were included (median age 69 [IQR 63-75] years, 33 females (33%), left atrial size 43 [IQR 39-47] mm, paroxysmal AF 24%). Median procedure time was 66 (IQR 59-77) min, and fluoroscopy time was 11 (8-14) min. PWI using PFA was achieved in 100% of patients with a median of 19 applications (IQR 14-26). There were no major complications. Overall, in 15 patients (15%), recurrent AF/AT was noted during a median follow-up of 144 (94-279) days. CONCLUSIONS: PWI using PFA appears safe and results in high acute isolation rates and high arrhythmia survival during mid-term follow-up. Further randomized trials are essential and warranted.

Durability of pulmonary vein isolation for atrial fibrillation: a meta-analysis and systematic review.

AIMS: Pulmonary vein isolation (PVI) plays a central role in the interventional treatment of atrial fibrillation (AF). Uncertainties remain about the durability of ablation lesions from different energy sources. We aimed to systematically review the durability of ablation lesions associated with various PVI-techniques using different energy sources for the treatment of AF. METHODS AND RESULTS: Structured systematic database search for articles published between January 2010 and January 2023 reporting PVI-lesion durability as evaluated in the overall cohort through repeat invasive remapping during follow-up. Studies evaluating only a proportion of the initial cohort in redo procedures were excluded. A total of 19 studies investigating 1050 patients (mean age 60 years, 31% women, time to remap 2-7 months) were included. In a pooled analysis, 99.7% of the PVs and 99.4% of patients were successfully ablated at baseline and 75.5% of the PVs remained isolated and 51% of the patients had all PVs persistently isolated at follow-up across all energy sources. In a pooled analysis of the percentages of PVs durably isolated during follow-up, the estimates of RFA were the lowest of all energy sources at 71% (95% CI 69-73, 11 studies), but comparable with cryoballoon (79%, 95%CI 74-83, 3 studies). Higher durability percentages were reported in PVs ablated with laser-balloon (84%, 95%CI 78-89, one study) and PFA (87%, 95%CI 84-90, 2 studies). CONCLUSION: We observed no significant difference in the durability of the ablation lesions of the four evaluated energies after adjusting for procedural and baseline populational characteristics.

Accuracy in detecting atrial fibrillation in single-lead ECGs: an online survey comparing the influence of clinical expertise and smart devices.

BACKGROUND: Manual interpretation of single-lead ECGs (SL-ECGs) is often required to confirm a diagnosis of atrial fibrillation. However accuracy in detecting atrial fibrillation via SL-ECGs may vary according to clinical expertise and choice of smart device. AIMS: To compare the accuracy of cardiologists, internal medicine residents and medical students in detecting atrial fibrillation via SL-ECGs from five different smart devices (Apple Watch, Fitbit Sense, KardiaMobile, Samsung Galaxy Watch, Withings ScanWatch). Participants were also asked to assess the quality and readability of SL-ECGs. METHODS: In this prospective study (BaselWearableStudy, NCT04809922), electronic invitations to participate in an online survey were sent to physicians at major Swiss hospitals and to medical students at Swiss universities. Participants were asked to classify up to 50 SL-ECGs (from ten patients and five devices) into three categories: sinus rhythm, atrial fibrillation or inconclusive. This classification was compared to the diagnosis via a near-simultaneous 12-lead ECG recording interpreted by two independent cardiologists. In addition, participants were asked their preference of each manufacturer's SL-ECG. RESULTS: Overall, 450 participants interpreted 10,865 SL-ECGs. Sensitivity and specificity for the detection of atrial fibrillation via SL-ECG were 72% and 92% for cardiologists, 68% and 86% for internal medicine residents, 54% and 65% for medical students in year 4-6 and 44% and 58% for medical students in year 1-3; p <0.001. Participants who stated prior experience in interpreting SL-ECGs demonstrated a sensitivity and specificity of 63% and 81% compared to a sensitivity and specificity of 54% and 67% for participants with no prior experience in interpreting SL-ECGs (p <0.001). Of all participants, 107 interpreted all 50 SL-ECGs. Diagnostic accuracy for the first five interpreted SL-ECGs was 60% (IQR 40-80%) and diagnostic accuracy for the last five interpreted SL-ECGs was 80% (IQR 60-90%); p <0.001. No significant difference in the accuracy of atrial fibrillation detection was seen between the five smart devices; p = 0.33. SL-ECGs from the Apple Watch were considered as having the best quality and readability by 203 (45%) and 226 (50%) participants, respectively. CONCLUSION: SL-ECGs can be challenging to interpret. Accuracy in correctly identifying atrial fibrillation depends on clinical expertise, while the choice of smart device seems to have no impact.

Clinical validation of an artificial intelligence algorithm offering cross-platform detection of atrial fibrillation using smart device electrocardiograms.

BACKGROUND: Several smart devices are able to detect atrial fibrillation automatically by recording a single-lead electrocardiogram, and have created a work overload at the hospital level as a result of the need for over-reads by physicians. AIM: To compare the atrial fibrillation detection performances of the manufacturers' algorithms of five smart devices and a novel deep neural network-based algorithm. METHODS: We compared the rate of inconclusive tracings and the diagnostic accuracy for the detection of atrial fibrillation between the manufacturers' algorithms and the deep neural network-based algorithm on five smart devices, using a physician-interpreted 12-lead electrocardiogram as the reference standard. RESULTS: Of the 117 patients (27% female, median age 65 years, atrial fibrillation present at time of recording in 30%) included in the final analysis (resulting in 585 analyzed single-lead electrocardiogram tracings), the deep neural network-based algorithm exhibited a higher conclusive rate relative to the manufacturer algorithm for all five models: 98% vs. 84% for Apple; 99% vs. 81% for Fitbit; 96% vs. 77% for AliveCor; 99% vs. 85% for Samsung; and 97% vs. 74% for Withings (P<0.01, for each model). When applying our deep neural network-based algorithm, sensitivity and specificity to correctly identify atrial fibrillation were not significantly different for all assessed smart devices. CONCLUSION: In this clinical validation, the deep neural network-based algorithm significantly reduced the number of tracings labeled inconclusive, while demonstrating similarly high diagnostic accuracy for the detection of atrial fibrillation, thereby providing a possible solution to the data surge created by these smart devices.

Echocardiographic pattern of left ventricular function recovery in tachycardia-induced cardiomyopathy patients.

AIMS: Tachycardia-induced cardiomyopathy (TCM) represents a partially reversible type of cardiomyopathy (CM) that is often underdiagnosed and cardiac chamber remodelling in TCM remains incompletely understood. We aim to explore differences in the dimensions of the left ventricle and functional recovery in patients with TCM compared with patients with other forms of CM. METHODS AND RESULTS: We identified patients with reduced ejection fraction (≤50%) and/or atrial fibrillation or flutter with a left ventricular ejection fraction that improved from baseline (≥15% in left ventricular ejection fraction at follow-up or normalization of cardiac function with at least 10% improvement). Patients were then divided into two groups: (A) TCM patients and (B) patients with other forms of CM (controls). Two hundred thirty-eight patients were included (31% female, 70 years median age), 127 patients had TCM, and 111 had other forms of CM. Patients with TCM did not significantly improve indexed left ventricular volume (LVEDVI) after treatment (60 [45, 84] mL/m2 versus 56 [45, 70] mL/m2 , P = ns) compared with controls (67 [54, 81] mL/m2 versus 52 [42, 69] mL/m2 , P < 0.001). Patients with TCM patients had significantly worse fractional shortening at baseline than controls (15.5 [12, 23] vs. 20 [13, 30], P = 0.01) and higher indexed left atrial volume (LAVI) at baseline than controls (48 [37, 58] vs. 41 [33, 51], P = 0.01) that remained dilated at follow-up (follow-up LAVI 41 [33, 52] mL/m2 ). Good predictors of TCM were: normal LVEDVI (LVEDVI < 58 mL/m2 (M) and < 52 mL/m2 (F)) (odds ratio [OR] 5.2; 95% confidence interval [CI] 2.2-13.3, P < 0.001), fractional shortening < 30% (OR 3.5; 95% CI 1.4-9.2, P = 0.009), LAVI >40 mL/m2 (OR 3.4; 95% CI 1.6-7.3, P = 0.001) and normal wall thickness left ventricle (OR 3.2; 95% CI 1.4-7.8, P = 0.008). 54% of patients with TCM demonstrated diastolic dysfunction at follow-up, without differences from controls (54% vs. 43%, P = ns). 21% of patients with TCM showed persistent heart failure symptoms at follow-up compared with 4.5% of controls, P = 0.004. CONCLUSIONS: TCM patients have a specific pattern of functional recovery with persistent remodelling of the left atria and left ventricle. Several echocardiographic parameters might help identify TCM before treatment.

Clinical Validation of 5 Direct-to-Consumer Wearable Smart Devices to Detect Atrial Fibrillation: BASEL Wearable Study.

BACKGROUND: Multiple smart devices capable to detect atrial fibrillation (AF) are presently available. Sensitivity and specificity for the detection of AF may differ between available smart devices, and this has not yet been adequately investigated. OBJECTIVES: The aim was to assess the accuracy of 5 smart devices in identifying AF compared with a physician-interpreted 12-lead electrocardiogram as the reference standard in a real-world cohort of patients. METHODS: We consecutively enrolled patients presenting to a cardiology service at a tertiary referral center in a prospective, diagnostic study. RESULTS: We prospectively analyzed 201 patients (31% women, median age 66.7 years). AF was present in 62 (31%) patients. Sensitivity and specificity for the detection of AF were comparable between devices: 85% and 75% for the Apple Watch 6, 85% and 75% for the Samsung Galaxy Watch 3, 58% and 75% for the Withings Scanwatch, 66% and 79% for the Fitbit Sense, and 79% and 69% for the AliveCor KardiaMobile, respectively. The rate of inconclusive tracings (the algorithm was unable to determine the heart rhythm) was 18%, 17%, 24%, 21%, and 26% for the Apple Watch 6, Samsung Galaxy Watch 3, Withings Scan Watch, Fitbit Sense, and AliveCor KardiaMobile (P < 0.01 for pairwise comparison), respectively. By manual review of inconclusive tracings, the rhythm could be determined in 955 (99%) of 969 single-lead electrocardiograms. Regarding patient acceptance, the Apple Watch was ranked first (39% of participants). CONCLUSIONS: In this clinical validation of 5 direct-to-consumer smart devices, we found differences in the amount of inconclusive tracings diminishing sensitivity and specificity of the smart devices. In a clinical setting, manual review of tracings is required in about one-fourth of cases.

Risk factors for the development of premature ventricular complex-induced cardiomyopathy: a systematic review and meta-analysis.

BACKGROUND: Premature ventricular complexes (PVCs) are a potentially reversible cause of heart failure. However, the characteristics of patients most likely to develop impaired left ventricular function are unclear. Hence, the objective of this study is to systematically assess risk factors for the development of PVC-induced cardiomyopathy. METHODS: We performed a structured database search of the scientific literature for studies investigating risk factors for the development of PVC-induced cardiomyopathy (PVC-CM). We investigated the reporting of PVC-CM risk factors (RF) and assessed the comparative association of the different RF using random-effect meta-analysis. RESULTS: A total of 26 studies (9 prospective and 17 retrospective studies) involving 16,764,641 patients were analyzed (mean age 55 years, 58% women, mean PVC burden 17%). Eleven RF were suitable for quantitative analysis (≥ 3 occurrences in multivariable model assessing a binary change in left ventricular (LV) function). Among these, age (OR 1.02 per increase in the year of age, 95% CI [1.01, 1.02]), the presence of symptoms (OR 0.18, 95% CI [0.05, 0.64]), non-sustained ventricular tachycardias (VT) (OR 3.01, 95% CI [1.39, 6.50]), LV origin (OR 2.20, 95% CI [1.14, 4.23]), epicardial origin (OR 4.72, 95% CI [1.81, 12.34]), the presence of interpolation (OR 4.93, 95% CI [1.66, 14.69]), PVC duration (OR 1.05 per ms increase in QRS-PVC duration [1.004; 1.096]), and PVC burden (OR 1.06, 95% CI [1.04, 1.08]) were all significantly associated with PVC-CM. CONCLUSIONS: In this meta-analysis, the most consistent risk factors for PVC-CM were age, non-sustained VT, LV, epicardial origin, interpolation, and PVC burden, whereas the presence of symptoms significantly reduced the risk. These findings help tailor stringent follow-up of patients presenting with frequent PVCs and normal LV function.

Magnetic field interactions of smartwatches and portable electronic devices with CIEDs - Did we open a Pandora's box?

Introduction: Magnetic interaction of portable electronic devices (PEDs), such as state-of-the art mobile phones, with cardiovascular implantable electronic devices (CIEDs) has been reported. The aim of the study was to quantify the magnetic fields of latest generation smartwatches and other PEDs and to evaluate and predict their risk of CIED interactions. Methods: High resolution magnetic field characterization of five smartwatches (Apple Watch 6/7, Fitbit Sense, Samsung Galaxy 3, Withings Scanwatch) was performed using a novel magnetic field camera. Ex vivo measurements of the minimal safety distance (MSD) at which no mode switch can be observed were performed between 11 PEDs and six representative CIEDs. Results: Maximal 1 mT distances ranged between 10 mm (Withings) and 19 mm (Fitbit and AppleWatch), and 1 mT volumes between 6 cm3 (Withings) and 19 cm3 (Fitbit). All these measures were observed only for the back side of the smartwatches. While most smartwatches with measured 1 mT distance < 15 mm posed low ex vivo interaction within a distance of < 10 mm, PEDs such as electronic pens and in-ear-headphones with measured 1 mT distance > 15 mm showed device interaction up to > 15 mm. Linear regression analysis showed a linear relationship of the MSD with 1 mT distance (B coefficient: 0.46; 95 %-CI: 0.25-0.67, p < 0.001). Conclusion: Smartwatches are safer compared to other PEDs such as electronic pens or in-ear headphones with regards to CIED interaction. With a standardized magnetic field camera, the risk assessment of CIED interaction of novel PEDs is feasible.

Clinical Validation of Automated Corrected QT-Interval Measurements From a Single Lead Electrocardiogram Using a Novel Smartwatch.

Introduction: The Withings Scanwatch (Withings SA, Issy les Moulineaux, France) offers automated analysis of the QTc. We aimed to compare automated QTc-measurements using a single lead ECG of a novel smartwatch (Withings Scanwatch, SW-ECG) with manual-measured QTc from a nearly simultaneously recorded 12-lead ECG. Methods: We enrolled consecutive patients referred to a tertiary hospital for cardiac workup in a prospective, observational study. The QT-interval of the 12-lead ECG was manually interpreted by two blinded, independent cardiologists through the tangent-method. Bazett's formula was used to calculate QTc. Results were compared using the Bland-Altman method. Results: A total of 317 patients (48% female, mean age 63 ± 17 years) were enrolled. HR-, QRS-, and QT-intervals were automatically calculated by the SW in 295 (93%), 249 (79%), and 177 patients (56%), respectively. Diagnostic accuracy of SW-ECG for detection of QTc-intervals ≥ 460 ms (women) and ≥ 440 ms (men) as quantified by the area under the curve was 0.91 and 0.89. The Bland-Altman analysis resulted in a bias of 6.6 ms [95% limit of agreement (LoA) -59 to 72 ms] comparing automated QTc-measurements (SW-ECG) with manual QTc-measurement (12-lead ECG). In 12 patients (6.9%) the difference between the two measurements was greater than the LoA. Conclusion: In this clinical validation of a direct-to-consumer smartwatch we found fair to good agreement between automated-SW-ECG QTc-measurements and manual 12-lead-QTc measurements. The SW-ECG was able to automatically calculate QTc-intervals in one half of all assessed patients. Our work shows, that the automated algorithm of the SW-ECG needs improvement to be useful in a clinical setting.

Incremental diagnostic and prognostic value of the QRS-T angle, a 12-lead ECG marker quantifying heterogeneity of depolarization and repolarization, in patients with suspected non-ST-elevation myocardial infarction.

BACKGROUND: The value of the 12-lead ECG in the diagnosis of non-ST-elevation myocardial infarction (NSTEMI) is limited due to insufficient sensitivity and specificity of standard ECG criteria. The QRS-T angle reflects depolarization-repolarization heterogeneity and might assist in detecting patients with a NSTEMI (diagnosis) as well as predicting patients with an increased mortality risk (prognosis). METHODS: We prospectively enrolled 2705 consecutive patients with symptoms suggestive of NSTEMI. The QRS-T angle was automatically derived from the standard 10 s 12-lead ECG recorded at presentation to the ED. Patients were followed up for all-cause mortality for 2 years. RESULTS: NSTEMI was the final diagnosis in 15% (n = 412) of patients. QRS-T angles were significantly greater in patients with NSTEMI compared to those without (p < 0.001). The use of the QRS-T angle in addition to standard ECG criteria indicative of ischemia improved the diagnostic accuracy for NSTEMI as quantified by the area under the ROC curve from 0.68 to 0.72 (p < 0.001). An algorithm for the combined use of standard ECG criteria and the QRS-T angle improved the sensitivity of the ECG for NSTEMI from 45% to 78% and the specificity from 86% to 91% (p < 0.001 for both comparisons). The 2-year survival rates were 98%, 97% and 87% according to QRS-T angle tertiles (p < 0.001). CONCLUSION: In patients with suspected NSTEMI, the QRS-T angle derived from the standard 12-lead ECG provides incremental diagnostic accuracy on top of standard ECG criteria indicative of ischemia, and independently predicts all-cause mortality during 2 years of follow-up.