Citations

Introduction: Shared decision-making (SDM) about anticoagulant treatment in patients with atrial fibrillation (AF) is widely recommended but its effectiveness is unclear. We assessed the extent to which using an SDM tool promotes high-quality SDM. Hypothesis: SDM will improve communication quality, knowledge, and decisional conflict and will translate to improved medication adherence and outcomes. Methods: In this multicentre encounter-randomized trial, we included patients with nonvalvular AF considering starting/reviewing anticoagulation, and their clinicians. We compared usual care with or without ANTICOAGULATION CHOICE, an SDM conversation tool for use during the clinical encounter that presents individualized risk estimates and compares anticoagulation options across patient important issues. Results: We enrolled and video-recorded 922 patient-clinician encounters. So far, our data have shown that participants in both arms reported near-optimal communication quality, knowledge, and decisional conflict (JAMA IM-2020). Patients in intervention-arm estimated their stroke risks more accurately and were more involved in decision-making (JAMA IM-2020), clinicians were more satisfied (JAMA IM-2020), and cost conversation were more common (JAMA NO-2021). Use of the intervention had no effect on treatment decisions, encounter duration (JAMA IM-2020), treatment adherence, clinical outcomes (JAHA-epub) or patient-perceived sense of the care plan (PEC-epub). New analyses have shown that compared to eligible white participants, black participants more frequently did not enrol to the study. Enrolment of Black, indigenous, and people of colour benefited most form including and prioritizing practices most likely to care for them. We found no evidence of contamination (tool use, functional, or learned) between study arms. Conclusions: Use of an SDM tool improved some measures of SDM quality, without affecting treatment decisions and their sensibility, encounter duration, patient adherence, or clinical outcomes. Our results should calibrate expectations as to what could be accomplished by implementing SDM tools about anticoagulation in the care of patients with AF

Background: An electrocardiogram (ECG)-based artificial intelligence (AI) algorithm has shown good performance in detecting hypertrophic cardiomyopathy (HCM AI-score). However, its clinical application may be challenging due to the low disease prevalence and potentially high false-positive rate when screening unenriched cohorts. Aims: To identify clinical characteristics associated with false-positive HCM AI-score results to improve screening tool application. Methods: From a cohort of 20,677 patients with a digital 12-lead ECG in routine clinical practice in 2021, we performed a detailed chart review of the 200 patients with highest HCM AI-score to determine their HCM status and collect clinical data. Stepwise forward logistic regression was used to create a clinical variable-based “candidacy for HCM screening” score (CHS score), differentiating true- and likely true-positive from likely false-positive HCM AI-score results. We then validated the CHS score in 200 patients with the highest HCM AI-scores from an independent cohort of 15,147 patients with an ECG in 2022. Results: Among the 200 patients in 2021 (median age 71 years, 48% female), 96 (48%) were likely false-positive HCM AI-score results, 71 (36%) definite or likely true-positive results, and 33 (16%) indeterminate. The CHS score included a history of surgical septal myectomy, coronary artery bypass graft surgery, cardiac valve replacement, and pacemaker, and had an AUC of 0.88 (95% CI 0.83-0.92) for the differentiation of true- versus false-positive AI results. In the 2022 cohort, the CHS score had an AUC of 0.82 (95% CI 0.76 - 0.88). After limiting the cohort to HCM candidates identified by the CHS score, the false-positive results rate was reduced to 27/114 (24%). Conclusions: By applying the CHS score to patients with positive HCM AI-screening results, the false-positive rate decreased from 49% to 24%. Utilizing a clinical model used in tandem with an AI model improved diagnostic yield.

Evaluating the Electrocardiographic Characteristics of Transgender Patients Undergoing Gender-affirming Hormone Therapy with Artificial Intelligence Introduction: An artificial intelligence (AI) electrocardiogram (ECG) algorithm has been demonstrated to identify a patient’s birth designated sex with 90.4% accuracy with an AUC = 0.97. However, the algorithm has not been evaluated in transgender and gender diverse individuals. We aimed to determine how the algorithm output varies among transgender individuals seen at the Mayo Clinic Transgender and Intersex Specialty Care Clinic (TISCC) who are undergoing gender-affirming hormonal treatment (GAHT). Methods: We applied the algorithm (which assessed the probability of “male” between 0 and 1, with < 0.5 considered as “female” and > 0.5 as “male”) to ECGs obtained from patients in the TISCC before and after the initiation of GAHT. GAHT frequently involves the use of testosterone in transgender men, and estrogen and anti-androgen agents in transgender women. Patient characteristics, including the date of the initiation of GAHT, were collected through chart review. Results: Among transgender women, the AI model probability of “male” decreased from 0.84± 0.25 (12/86 classified as males) to 0.59±0.36 (68/173 classified as male) with GAHT (p < 7.8e-10). Among transgender men, the AI model probability of “male” increased from 0.16 ± 0.28 (7/47 classified as male) to 0.41 ±0.38 (22/53 classified as male) with GAHT (p < 2.4e-4). Conclusion: After GAHT initiation, the Mayo Clinic AI ECG algorithm assigns a higher probability of being male among transgender men and a lower probability of being male among transgender women. These findings warrant further investigation into the electrocardiographic differences across genders and the effects of GAHT on the human heart.

Introduction: Over 100 FDA-approved medications, electrolyte perturbations, and many disease states can prolong the QTc beyond its 99 th percentile value resulting in acquired QT prolongation. In contrast, approximately 1 in 2000 people have congenital long QT syndrome (LQTS) hallmarked by pathological QT prolongation secondary to genetic defects in the heart. Hypothesis: An artificial intelligence (AI) deep neural network (DNN) can distinguish patients with LQTS from those with acquired QT prolongation. Methods: The study cohort included all patients with LQTS evaluated in the Windland Smith Rice Genetic Heart Rhythm Clinic and controls from Mayo Clinic’s ECG data vault comprising over 2.5 million patients. For the AI-DNN model, every patient/control with ≥ 1 ECG above age- and sex- specific 99 th percentile values for QTc [> 460 ms for all patients (male/female) < 13 years of age, or > 470 ms for men and > 480 ms for women above this age] was included. An AI-DNN involving a multi-layer c onvolutional n eural n etwork (CNN) was developed to classify patients. LQTS patients were age- and sex- matched to controls at 1:5 ratio. Results: Among the 1,599 patients with genetically confirmed LQTS, 808 had ≥ 1 ECG with QTc above the threshold (2,987 ECGs) compared to 361,069/2.5M controls (14% of Mayo Clinic patients getting an ECG, ‘presumed negative’; 989,313 ECGs). Following age- and sex- matching and splitting, 3,309 (training), 411 (validation) and 887 (testing) control ECGs were used. This model distinguished LQTS from those with acquired QT prolongation with an AUC of 0.896 (accuracy 85%, sensitivity 77%, specificity 88%, PPV 0.58, NPV 0.94). After exclusion of patients with a wide QRS (>150 ms) or pacemaker, the model remained successful in distinguishing the two groups (AUC 0.85, accuracy 78%, sensitivity 78%, specificity 87%, PPV 59%, NPV 94%). Conclusions: For patients with a QTc exceeding its 99 th percentile values, this novel AI-DNN functions as a LQTS mutation detector being able to identify patients with abnormal QT prolongation secondary to a LQTS-causative mutation rather than acquired QT prolongation with a >50% positive predictive value. This algorithm may facilitate screening for this potentially lethal, yet highly treatable, genetic heart disease.

BACKGROUND: Differentiation of wide QRS complex tachycardias (WCTs) into ventricular tachycardia (VT) and supraventricular wide QRS tachycardia (SWCT) through 12-lead electrocardiogram (ECG) interpretation is one of the most critical yet challenging tasks in clinical practice. Recent research highlights novel automated methods that successfully diagnose WCTs. We sought to directly compare the diagnostic efficacy of these novel automated WCT differentiation methods (i.e., WCT Formula, VT Prediction Model, and WCT Formula II) to traditional ECG interpretation approaches (i.e., Brugada and Vereckei aVR algorithms). METHODS: A collection of paired WCT and baselined ECGs were retrospectively analyzed. Next, an electrophysiologist blindly and prospectively applied the Brugada and Vereckei aVR algorithms. Separately, computerized measurements were used to apply the three automated WCT differentiation methods. The diagnostic performance of each method was then evaluated. RESULTS: 213 WCTs (111 VT and 102 SWCT) from 105 patients were analyzed. The WCT Formula demonstrated superior overall accuracy (84.7% vs. 64.8%), specificity (86.2% vs. 35.3%), positive predictive value (PPV) (85.7% vs. 60.7%), and positive (+) likelihood ratio (LR) (6.06 vs. 1.42) compared to Vereckei aVR algorithm. Similarly, the WCT Formula attained higher specificity (86.2% vs. 61.8%), PPV (85.7% vs. 71.9%), and (+) LR (6.06 vs. 2.36) than the Brugada algorithm. The WCT Formula II demonstrated superior overall accuracy (89.2% vs. 64.8%), specificity (84.3 % vs. 35.3%), PPV (86.7% vs. 60.7%), and positive (+) LR (5.97 vs.1.42) compared to Vereckei aVR algorithm as well as higher specificity (84.3 % vs. 61.8 %), PPV (86.7% vs. 71.9%), and (+) LR (5.97 vs. 2.36) than the Brugada algorithm. The VT prediction model yielded superior specificity (79.4% vs. 35.3%), PPV (79.8 vs. 60.7%) and (+) LR (3.63 vs.1.42) compared to the Vereckei aVR algorithm but did not demonstrate superior diagnostic performance to the Brugada algorithm. CONCLUSIONS: Novel automated WCT differentiation methods achieved similar or superior diagnostic performance metrics compared to traditional ECG interpretation approaches.

Introduction: Deep-learning artificial intelligence (AI) algorithms can identify hypertrophic cardiomyopathy (HCM) from the electrocardiogram (ECG). Preliminary data indicate that AI-ECG may also track treatment response in HCM patients (pts) receiving mavacamten (Mava). In this placebo-controlled analysis, we investigated the performance of AI-ECG in HCM detection and its changes during Mava treatment compared to placebo in the phase 3, EXPLORER-HCM trial. Methods: Pts with non-paced ECGs enrolled in EXPLORER-HCM, evaluating Mava over 30 weeks of treatment, were included. All pts had a 12-lead ECG pre-treatment, at each study visit, and 8 weeks after treatment end. Two independently developed AI-ECG convolutional neural network algorithms from the Mayo Clinic and the University of California, San Francisco (UCSF) were applied to all ECGs suitable for analysis. Serial AI-ECG-predicted scores, which correlate with HCM likelihood (range 0-1), were examined between Mava and placebo groups. Results: A total of 101 Mava and 110 placebo pts were included (mean age 58.6, female 40%). At baseline, the mean Mayo and UCSF AI-ECG scores for HCM were 0.82 and 0.65, with 94% and 88% of pts exceeding the respective diagnostic thresholds of 0.11 (Mayo) and 0.4 (UCSF). During 30 weeks of follow-up in the Mava group, the algorithms showed a mean AI-ECG score reduction of 0.32 (Mayo) and 0.17 (UCSF) in all pts between start and end of treatment. In comparison, AI-ECG scores decreased by 0.05 (Mayo) and increased by 0.02 (UCSF) in the placebo group (Mava vs placebo, t-test p-values <0.001 for both algorithms; Figure ) Conclusion: In the EXPLORER-HCM trial population, two independent deep-learning AI-ECG algorithms correctly identified most HCM pts and demonstrated significant reductions in predicted HCM scores after treatment with Mava when compared to placebo. This relative normalization of the ECG suggests disease-modifying properties of Mava, deserving further investigation.

Introduction: Cardiac resynchronization therapy (CRT) has known benefit for heart failure (HF) with reduced left ventricular ejection fraction (LVEF) and dyssynchrony. However, very little is known about its benefit in infiltrative cardiomyopathies such as cardiac amyloidosis (CA) due to relative rarity of these conditions. Methods: This retrospective case series (2003-2021) comprised patients with HF due to transthyretin wild-type CA with (case) or without (control) CRT implant. Only controls with an indication for CRT or LVEF <40% were included. To further match disease severity, cases were 1:1 matched (propensity score tolerance 0.04) to controls for age, NYHA class, and CA stage (UK Staging). Primary outcome was MACE (major adverse cardiac events) which included stroke, myocardial infarction, HF hospitalization and cardiac mortality during a 3-year follow-up. Loss to follow-up was censored as per last visit date. LVEF and NYHA were noted closest to 1-year follow-up. Survival analysis for MACE was performed with Kaplan-Meier (KM) and multivariable cox regression models. Results: Study Cohort (n=160) had a mean age of 75.5 (SD 7.8) years and 97.5% were males. A majority had NYHA class 2 or 3 (92.3%) and CA stage 2 (42.2.%). Notably, median LVEF at baseline was lower in controls (33.5 vs 36.5; P = .001) and 77.5% of controls had an indication for CRT device. A total of 79 MACE events occurred, most frequent being HF hospitalization (n=45). KM curves were comparable for MACE events in the two groups (log-rank test; P = .059). Even after adjusting, as in table, CRT device was not associated with MACE events (adjusted HR 0.66; 95% CI 0.40-1.09; P = .106). Lastly, change in LVEF ( P = .84) and NYHA ( P = .86) at follow-up was similar in cases vs controls. Conclusions: The findings of this largest yet observational study to examine this question suggest that CRT in patients with transthyretin wild-type CA does not favorably impact outcomes for MACE events, improve LVEF or NYHA functional class.

Background: The communities in which people live affect their health. We aimed to determine whether neighborhood characteristics are associated with outcomes in patients with atrial fibrillation (AF) in a Midwest community. Methods: Patients with incident AF from 2013-2017 from a 27-county region who survived >90 days after their diagnosis were identified (N=14,725). Patient addresses were linked to publicly-available data on population density (rural-urban commuting area; RUCA) and neighborhood socioeconomic status (area deprivation index; ADI). Patients were followed for ischemic stroke or transient ischemic attack (TIA), heart failure (HF), dementia/delirium, and all-cause death through 2021. Cox regression determined risk of outcomes by RUCA category (urban, micropolitan, small rural, isolated) and ADI quartile after adjustment for age, sex, race/ethnicity, and CHA2DS2-VASc score. Results: After excluding persons missing race/ethnicity (n=46), RUCA (n=1,465), and ADI (n=1,640), 13,233 patients with AF (mean age 73.8 years, 57% men) remained. After a median follow-up of 4.7 years, 5,989 patients died; 582 experienced a stroke/TIA, 4,122 (of 9,156 patients at risk) developed HF, and 1,758 (of 12,154 patients at risk) developed dementia/delirium. Persons living in isolated areas (HR 1.14, 95% CI 1.05-1.24) and the highest ADI quartile (HR 1.59, 95% CI 1.37-1.85) experienced the greatest risk of death (Figure). A trend toward an increased risk of HF was observed for the 2 highest ADI quartiles (p-trend <0.01), although the estimates did not reach statistical significance. The risk of other outcomes did not differ by RUCA or ADI category. Conclusion: Patients with AF residing in isolated areas and neighborhoods with greater area deprivation experience increased risks of death. Additional research is needed to better understand how neighborhood characteristics influence treatment, co-morbid conditions, and other outcomes in patients with AF.

Background: Recognition of the causes of early mortality after atrial fibrillation (AF) catheter ablation is essential for the improvement of patient safety. Objective: This study sought to determine the causes of early mortality (≤90-days) after AF ablation and to further understand the circumstances surrounding death after the procedure. Methods: We performed a retrospective analysis of AF ablation from January 1, 2013, to December 1, 2021 at the Mayo Clinic (Rochester, Phoenix, and Jacksonville). Causes of death were identified through a comprehensive chart review of the electronic health record from within the Mayo Clinic system and outside records when available. Results: A total of 6,723 patients were included in the study. The 30-day and 90-day all-cause mortality rates were 0.09% and 0.22%, respectively. Among all 90-day deaths, majority of the deaths (73.3%) did not have a direct relationship with the procedure. Sudden death was the most common cause of early death (20%), followed by peri-procedural stroke (13%), respiratory failure (13%), atrioesophageal fistula (13%), infection (7%), heart failure (7%), and traumatic brain injury (7%). The 90-day mortality rate directly due to AF ablation procedural complications was 0.06%. Most of the early deaths (93.3%) occurred after discharge from index hospitalization, among which 64.3% were readmitted before their death, 14% died at home, and 7% were discharged on comfort care and died in hospice. The main reason for readmission leading to death was respiratory failure (33%). Conclusion: Atrial fibrillation ablation procedure has a 90-day mortality of 0.22% and the most common cause of early mortality was sudden death. The majority (73.3%) of early mortality was not directly associated with a procedural complication, and the mortality rate due to complications associated with the AF ablation procedure was 0.06%.

Introduction: Prior AF screening trials demonstrated low yield, highlighting the need for more targeted approaches. An AI algorithm was developed to identify ECG signatures of AF risk during normal sinus rhythm, which has been validated in diverse external populations. Hypothesis: An AI-guided, targeted screening approach could improve the diagnosis of AF. Methods We conducted a pragmatic decentralized trial to prospectively recruit patients with stroke risk factors but no prior AF. The AI algorithm was applied to the ECGs performed in routine practice and divided patients into high- or low-risk groups. The primary endpoint was AF lasting ≥ 30 seconds on a subsequent 30-day continuous cardiac rhythm monitor. In a secondary analysis, trial participants were 1:1 propensity score-matched to real-world controls derived from the eligible but unenrolled population. Results A total of 1,003 patients from 40 U.S. states completed the study, with a mean age of 74 [SD 8.8] years. Over a mean of 22.3 days of continuous monitoring, AF was detected in 6 (1.6%) of low-risk patients and 48 (7.6%) of high-risk patients (OR 4.98 [2.11-11.75], p<0.001). Compared to usual care, AI-guided AF screening was associated with increased detection of AF (high-risk group: 4.2% vs. 11.1%, p<0.001; low-risk group: 0.9% vs. 2.4%, p=0.12) over a median follow-up of 10 months. Conclusions A prospective pragmatic study found that the AI a can risk-stratify a relatively uniform population (i.e., older adults at risk for stroke) to detect AF during short-term cardiac monitoring. Furthermore, when compared with usual care, AI-guided cardiac monitoring was associated with increased AF detection. As such, an AI-guided AF screening approach, leveraging existing EHR data and infrastructure, could be effective, patient-centered, and massively scalable, thereby reducing unnecessary health utilization and diagnosis-related anxiety (Clinicaltrials.gov: NCT04208971).