Citations

Background: Sex and age have long been known to affect the ECG. Several biologic variables and anatomic factors may contribute to sex and age-related differences on the ECG. We hypothesized that a convolutional neural network (CNN) could be trained through a process called deep learning to predict a person’s age and self-reported sex using only 12-lead ECG signals. We further hypothesized that discrepancies between CNN-predicted age and chronological age may serve as a physiological measure of health. Methods: We trained CNNs using 10-second samples of 12-lead ECG signals from 499 727 patients to predict sex and age. The networks were tested on a separate cohort of 275 056 patients. Subsequently, 100 randomly selected patients with multiple ECGs over the course of decades were identified to assess within-individual accuracy of CNN age estimation. Results: Of 275 056 patients tested, 52% were males and mean age was 58.6±16.2 years. For sex classification, the model obtained 90.4% classification accuracy with an area under the curve of 0.97 in the independent test data. Age was estimated as a continuous variable with an average error of 6.9±5.6 years (R-squared =0.7). Among 100 patients with multiple ECGs over the course of at least 2 decades of life, most patients (51%) had an average error between real age and CNN-predicted age of <7 years. Major factors seen among patients with a CNN-predicted age that exceeded chronologic age by >7 years included: low ejection fraction, hypertension, and coronary disease ( P <0.01). In the 27% of patients where correlation was >0.8 between CNN-predicted and chronologic age, no incident events occurred over follow-up (33±12 years). Conclusions: Applying artificial intelligence to the ECG allows prediction of patient sex and estimation of age. The ability of an artificial intelligence algorithm to determine physiological age, with further validation, may serve as a measure of overall health.

Objective Cardiac rehabilitation (CR) availability, programme characteristics and barriers are not well-known in low/middle-income countries (LMICs). In this study, they were compared with high-income countries (HICs) and by CR funding source. Methods A cross-sectional online survey was administered to CR programmes globally. Need for CR was computed using incident ischaemic heart disease (IHD) estimates from the Global Burden of Disease study. General linear mixed models were performed. Results CR was identified in 55/138 (39.9%) LMICs; 47/55 (85.5% country response rate) countries participated and 335 (53.5% programme response) surveys were initiated. There was one CR spot for every 66 IHD patients in LMICs (vs 3.4 in HICs). CR was most often paid by patients in LMICs (n=212, 65.0%) versus government in HICs (n=444, 60.2%; p<0.001). Over 85% of programmes accepted guideline-indicated patients. Cardiologists (n=266, 89.3%), nurses (n=234, 79.6%; vs 544, 91.7% in HICs, p=0.001) and physiotherapists (n=233, 78.7%) were the most common providers on CR teams (mean=5.8±2.8/programme). Programmes offered 7.3±1.8/10 core components (vs 7.9±1.7 in HICs, p<0.01) over 33.7±30.7 sessions (significantly greater in publicly funded programmes; p<0.001). Publicly funded programmes were more likely to have social workers and psychologists on staff, and to offer tobacco cessation and psychosocial counselling. Conclusion CR is only available in 40% of LMICs, but where offered is fairly consistent with guidelines. Governments should enact policies to reimburse CR so patients do not pay out-of-pocket.

Background The benefits of aerobic exercise are well-studied; there is no consensus on the association between resistance training and major adverse cardiovascular outcomes. This systematic review and meta-analysis aimed to address this issue. Design and methods We searched for randomized trials and cohort studies that evaluated the association between resistance training and mortality and cardiovascular events. Two investigators screened the identified abstracts and full-texts independently and in duplicate. Cochrane tools were used to assess the risk of bias. We calculated hazard ratios and 95% confidence intervals using random effect models. Results From the 1430 studies identified, 11 (one randomized trial and 10 cohort studies) met the inclusion criteria, totaling 370,256 participants with mean follow-up of 8.85 years. The meta-analysis showed that, compared with no exercise, resistance training was associated with 21% (hazard ratio (95% confidence interval (CI)), 0.79 (0.69–0.91)) and 40% (hazard ratio (95% CI), 0.60 (0.49–0.72)) lower all-cause mortality alone and when combined with aerobic exercise, respectively. Furthermore, resistance training had a borderline association with lower cardiovascular mortality (hazard ratio (95% CI), 0.83 (0.67–1.03)). In addition, resistance training showed no significant association with cancer mortality. Risk of bias was low to intermediate in the included studies. One cohort study looked at the effect of resistance training on coronary heart disease events in men and found a 23% risk reduction (risk ratio, 0.77, CI: 0.61–0.98). Conclusion Resistance training is associated with lower mortality and appears to have an additive effect when combined with aerobic exercise. There are insufficient data to determine the potential beneficial effect of resistance training on non-fatal events or the effect of substituting aerobic exercise with resistance training.

Background: In general, women report higher stress levels than men. High baseline anxiety, depression, and stress levels are associated with greater risk of cardiovascular diseases. Current evidence for efficacy of stress management interventions for women is limited. This study aimed at assessing the effect of a stress management and resiliency training (SMART) program for decreasing stress, anxiety, and depressive symptoms. Methods: Fifty moderately or severely stressed Women's Heart/Preventive Cardiology Clinic patients consented to the SMART intervention delivered online ( n = 36) or in-person ( n = 9). Primary outcome measures were the observed changes between baseline and at 12 weeks for the following psychometric tools: General Anxiety Disorder-7 (GAD-7), Patient Health Questionnaires (PHQ-9), Perceived Stress Scale (PSS), and Brief Resiliency Scale (BRS). Results: Forty-five patients completed the study. We observed significant improvements in PSS and GAD-7, but not in PHQ-9 or BRS, after the SMART intervention. When assessing outcomes among those with depressive symptoms at baseline (PHQ-9 > 15), we observed significant changes in PSS, GAD-7, and PHQ-9. No differences between online and in-person program delivery methods were found (all p -values >0.05). Conclusions: Training exposure using the SMART program to decrease stress and anxiety in women seeking preventive cardiology services was feasible and similarly effective, whether delivered online or in a single in-person session. Impacts on depression and resilience likely require a more intensive approach. In the future, larger randomized clinical trials with additional training and longer follow-up are warranted.

AbstractObjectivesWe sought to validate a deep learning algorithm designed to predict an ejection fraction (EF) less than or equal to 35% based on the 12‐lead electrocardiogram (ECG) in a large prospective cohort.BackgroundPatients undergoing routine ECG may have undetected left ventricular (LV) dysfunction that warrants further echocardiographic assessment. However, identification of these patients can be challenging.MethodsWe applied the algorithm to all ECGs interpreted by the Mayo Clinic ECG laboratory in September 2018. The performance of the algorithm was tested among patients with recent echocardiographic assessments of LV function. We also applied the algorithm in patients with no recent echocardiographic assessments of LV function to determine the rate of new “positive screens.”ResultsAmong 16 056 adult patients who underwent routine ECG, 8600 (age 67.1 ± 15.2 years, 45.6% male), had a transthoracic echocardiogram (TTE) and 3874 patients had a TTE and ECG less than 1 month apart. Among these patients, the algorithm was able to detect an EF less than or equal to 35% with 86.8% specificity, 82.5% sensitivity, and 86.5% accuracy, (area under the curve, 0.918). Among 474 “false‐positives screens,” 189 (39.8%) had an EF of 36% to 50%. Among patients with no prior TTE, the algorithm identified 3.5% of the patients with suspected EF less than or equal to 35%. Exploratory analysis suggests false positives could be reduced by assessing NT‐pro‐BNP after the initial “positive screen.”ConclusionsA deep learning algorithm detected depressed LV function with good accuracy in routine practice. Further studies are needed to validate the algorithm in patients with no prior echocardiogram and to assess the impact on echocardiography utilization, cost, and clinical outcomes.

The notion that cardiovascular disease (CVD) primarily occurs in men is slowly disappearing. More women than men die of CVD every year, and when women survive, the burden and consequences are worse than in men. Markers of stress and other psychosocial factors have been associated with poor outcomes. Multiple studies have demonstrated sex-based differences in the vascular and endothelial responses to mental stress. Psychosocial stressors were also found to be independent risk factors for the development and progression of CVD. This review arises from accumulating evidence suggesting that psychological well-being may improve cardiac-related outcomes, independent of cardiac risk factors. Despite the fact that positive physician-patient engagement is likely to play a critical role in promoting positive psychological traits and healthy behaviors, current physician awareness and advocacy are rather suboptimal, despite active awareness campaigns such as the American Heart Association's Go Red for Women ® . There is a need to further study the role and management of stress as a CVD risk factor, especially in women, who are disproportionately affected.

Alternative models of cardiac rehabilitation (CR) delivery, such as home or community-based programs, have been developed to overcome underutilization. However, their availability and characteristics have never been assessed globally. In this cross-sectional study, a piloted survey was administered online to CR programs globally. CR was available in 111/203 (54.7%) countries globally; data were collected in 93 (83.8% country response rate). 1082 surveys (32.1% program response rate) were initiated. Globally, 85 (76.6%) countries with CR offered supervised programs, and 51 (45.9%; or 25.1% of all countries) offered some alternative model. Thirty-eight (34.2%) countries with CR offered home-based programs, with 106 (63.9%) programs offering some form of electronic CR (eCR). Twenty-five (22.5%) countries with CR offered community-based programs. Where available, programs served a mean of 21.4% ± 22.8% of their patients in home-based programs. The median dose for home-based CR was 3 sessions (Q25−Q75 = 1.0–4.0) and for community-based programs was 20 (Q25–Q75 = 9.6–36.0). Seventy-eight (47.0%) respondents did not perceive they had sufficient capacity to meet demand in their home-based program, for reasons including funding and insufficient staff. Where alternative CR models are offered, capacity is insufficient half the time. Home-based CR dose is insufficient to achieve health benefits. Allocation to program model should be evidence-based.

In the European Union, Czech Republic ranks 3rd and 6th for the incidence of obesity and cardiovascular diseases, respectively. Worldwide, short sleep duration and excessive daytime sleepiness (EDS) characterize obese subjects, which in turn exhibit scarce physical activity and unhealthy diet. We aimed to understand the relationship between irregular sleep patterns, obesity and lifestyle factors, such as diet and physical activity, in a vulnerable Czech population. 1482 members of the Kardiovize cohort, a random sample of the Czech urban population, were included in a cross-sectional study. Exposure variables included self-reported sleep duration and EDS, assessed by the Epworth Sleepiness Scale. Primary outcomes were BMI and waist-to-hip ratio or prevalence of obesity and central obesity. Covariates included physical activity and diet. Associations and interactions between variables were evaluated using logistic regression analyses. After adjustment for covariates, short sleep duration (<7 h) was associated with greater odds of overweight (BMI > 25; OR = 1.42; 95%CI = 1.06–1.90; p = 0.020) and obesity (BMI > 30; OR = 1.40; 95%CI = 1.02–1.94; p = 0.047), while EDS was associated with greater odds of central obesity (OR = 1.72; 95%CI = 1.06–2.79; p = 0.030), independent of diet and physical activity. However, due to the cross-sectional nature of our study, further prospective, large-scale studies are needed to evaluate the etiological link and causality between sleep disturbances and obesity.