Heart age estimated using explainable advanced electrocardiography

BackgroundElectrocardiographic (ECG) Heart Age conveying cardiovascular risk has been estimated by both Bayesian and artificial intelligence approaches. We hypothesized that explainable measures from the 10-second 12-lead ECG could successfully predict Bayesian ECG Heart Age.MethodsAdvanced analysis was performed on ECGs from healthy subjects and patients with cardiovascular risk or proven heart disease. Regression models were used to predict a Bayesian 5-minute ECG Heart Age from the standard resting 10-second 12-lead ECG. The difference between 10-second ECG Heart Age and chronological age was compared.ResultsIn total, 2,771 subjects were included (n=1682 healthy volunteers, n=305 with cardiovascular risk factors, n=784 with cardiovascular disease). Overall, 10-second Heart Age showed strong agreement with the 5-minute Heart Age (R2=0.94, p<0.001, mean±SD bias 0.0±5.1 years). The difference between 10-second ECG Heart Age and chronological age was 0.0±5.7 years in healthy individuals, 7.4±7.3 years in subjects with cardiovascular risk factors (p<0.001), and 14.3±9.2 years for patients with cardiovascular disease (p<0.001).ConclusionsECG Heart Age can be accurately estimated from a 10-second 12-lead ECG in a transparent and explainable fashion based on known ECG measures, without artificial intelligence techniques. The difference between ECG Heart Age and chronological age increases markedly with cardiovascular risk and disease.

Cross-sectional assessment of cardiovascular risk factors in patients with knee osteoarthritis

Background: Osteoarthritis (OA) and cardiovascular disease (CVD) are prevalent in India. However, there is dearth of literature among Indians studying the relationship between the two. This study was carried out to assess various cardiovascular (CV) risk factors in patients with knee OA with an objective to investigate their association, screening and management. Methods: In total, 225 patients were included in this cross-sectional study. Participants were diagnosed with knee OA on the basis of the Kellgren and Lawrence (K-L) classification of their radiograph. Participants were also assessed for CV risk factors (age, body mass index, systolic blood pressure, diabetes mellitus, total cholesterol, high-density lipoprotein, smoking) with the help of the Joint British Society QRisk3 calculator (JBS3) a comprehensive risk score calculator as well as a screening tool, which produces three more variables, namely 10-years risk of developing CVD, physiological heart age and life expectancy. Chi Square, Fishers exact test and one-way ANOVA tests were used to compare the categorical and quantitative variables, respectively. Pearson’s correlation coefficient was used to assess the relationship between CV risk factors and knee OA. Multiple regression analysis was done to adjust the multiple con-founders and determine their significance. Results: Patients with severe knee OA had a statistically significantly higher prevalence of CV risk factors (p<0.05). Grade 4 knee OA patients were found to have a mean JBS3 risk of 38%, heart age of 82 years and life expectancy of 77 years as compared to grade 2 patients who had a mean JBS3 risk of 11%, heart age of 63 years and life expectancy of 82 years. Conclusions: Our study concluded that there is a strong positive correlation between knee OA and CVD, with CV risk score being directly proportional to the severity of OA.

Symptoms and Associations of COVID-19 in Pakistan A Single Centre, Descriptive Study

Introduction: Since SARS-CoV2 is a novel virus, not much was previously known about the disease, however recent studies have shown that it is transmitted via droplet infection and mainly affects the respiratory tract, causing symptoms of fever, fatigue and shortness of breath. Comorbidities increase risk of severe disease. Aims & Objectives: Our study aims to determine the predominant manifestations and correlations of COVID-19 in Pakistan. Place and duration of study: The study was carried out at CMH Lahore over ten-day duration from 1st June 2020 to 10th June 2020. Material & Methods: Samples of 107 confirmed cases of COVID-19 was taken. Participants were administered a questionnaire by attending doctor which enquired regarding their symptoms, presence of complications, and comorbidities. Data was analyzed using SPSS 25.0. A p-value of <0.05 was considered significant. Results: 77.6% of participants were male whereas 22.4% were female, with mean age 40.68 years. Symptoms commonly experienced were fever (71%), cough (32.7%), sore throat (36.4%), and myalgia (57%). Progression to complications was seen in 36 participants, most common being pneumonia (22.4%). Age of participants was significantly associated with symptoms of fever (p= 0.017), shortness of breath (p= 0.048) and fatigue (p= 0.021), and complication of pneumonia (p= 0.001). Comorbidities were associated with many symptoms and complications, most prominently cardiovascular disease was associated with development of complications like acute kidney injury (p= 0.002), cardiac failure (p= 0.005), and stroke (p= 0.005). Conclusion: Symptoms of COVID-19 are respiratory in nature primarily, however, the virus also affects other organs like gastrointestinal tract, neurons, heart. Age and presence of comorbidities increase risk of getting more severe disease, with highest risk of complications occurring in patients with history of cardiovascular disease.

The impact of health literacy-sensitive design and heart age in a cardiovascular disease prevention decision aid: randomised controlled trial and end user testing (Preprint)

BACKGROUND Shared decision making is as an essential principle for cardiovascular disease (CVD) prevention, where asymptomatic people are considering lifelong medication and lifestyle changes. OBJECTIVE This project aimed to develop and evaluate the first literacy-sensitive CVD prevention decision aid (DA) developed for people with low health literacy, and investigate the impact of literacy-sensitive design and heart age. METHODS We developed the standard DA based on international standards. The standard DA was based on our existing GP decision aid; the literacy-sensitive DA included simple language, supporting images, white space and a lifestyle action plan; the control DA used Heart Foundation materials. A randomised trial included 859 people aged 45-74 using a 3 (DA: standard, literacy-sensitive, control) x 2 (heart age: heart age + percentage risk, percentage risk only) factorial design, with outcomes including prevention intentions/behaviours, gist/verbatim knowledge of risk, credibility, emotional response and decisional conflict. We iteratively improved the literacy-sensitive version based on end user testing interviews with 20 people with varying health literacy levels. RESULTS Immediately post-intervention (n=859), there were no differences between the DA groups on any outcome. The heart age group was less likely to have a positive emotional response, perceived the message as less credible, and had higher gist/verbatim knowledge of heart age risk but not percentage risk. After 4 weeks (n=596), the DA groups had better gist knowledge of percentage risk than control. The literacy-sensitive decision aid group had higher fruit consumption, and the standard decision aid group had better verbatim knowledge of percentage risk. Verbatim knowledge was higher for heart age than percentage risk amongst those who received both. CONCLUSIONS The literacy-sensitive DA resulted in increased knowledge and lifestyle change for participants with varying health literacy levels and CVD risk results. Adding heart age did not increase lifestyle change intentions or behaviour but did affect psychological outcomes, consistent with previous findings. CLINICALTRIAL The trial protocol was pre-registered at ANZCTR (Trial number ACTRN12620000806965).

The impact of health literacy-sensitive design and heart age in a cardiovascular disease prevention decision aid: randomised controlled trial and end user testing

Introduction: Shared decision making is as an essential principle for cardiovascular disease (CVD) prevention, where asymptomatic people are considering lifelong medication and lifestyle changes. This project aimed to develop and evaluate the first literacy-sensitive CVD prevention decision aid (DA) developed for people with low health literacy, and investigate the impact of literacy-sensitive design and heart age. Methods: We developed the standard DA based on international standards. The literacy-sensitive version included simple language, supporting images, white space and a lifestyle action plan. A randomised trial included 859 people aged 45-74 using a 3 (DA: standard, literacy-sensitive, control) x 2 (heart age: heart age + percentage risk, percentage risk only) factorial design, with outcomes including prevention intentions/behaviours, gist/verbatim knowledge of risk, credibility, emotional response and decisional conflict. We iteratively improved the literacy-sensitive version based on end user testing interviews with 20 people with varying health literacy levels. Results: Immediately post-intervention (n=859), there were no differences between the DA groups on any outcome. The heart age group was less likely to have a positive emotional response, perceived the message as less credible, and had higher gist/verbatim knowledge of heart age risk but not percentage risk. After 4 weeks (n=596), the DA groups had better gist knowledge of percentage risk than control. The literacy-sensitive decision aid group had higher fruit consumption, and the standard decision aid group had better verbatim knowledge of percentage risk. Verbatim knowledge was higher for heart age than percentage risk amongst those who received both. Discussion: The literacy-sensitive DA resulted in increased knowledge and lifestyle change for participants with varying health literacy levels and CVD risk results. Adding heart age did not increase lifestyle change intentions or behaviour but did affect psychological outcomes, consistent with previous findings. Key words: decision aids, shared decision making, risk communication, heart age, cardiovascular disease prevention, behaviour change, health literacy MeSH Terms: Health Literacy, Cardiovascular Diseases, Decision Making (Shared), Life Style, Decision Support Techniques

Cardiovascular disease risk communication in NHS Health Checks using QRISK®2 and JBS3 risk calculators: the RICO qualitative and quantitative study

Background The NHS Health Check is a national cardiovascular disease prevention programme. There is a lack of evidence on how health checks are conducted, how cardiovascular disease risk is communicated to foster risk-reducing intentions or behaviour, and the impact on communication of using different cardiovascular disease risk calculators. Objectives RIsk COmmunication in Health Check (RICO) study aimed to explore practitioner and patient understanding of cardiovascular disease risk, the associated advice or treatment offered by the practitioner, and the response of the patients in health checks supported by either the QRISK®2 or the JBS3 lifetime risk calculator. Design This was a qualitative study with quantitative process evaluation. Setting Twelve general practices in the West Midlands of England, stratified on deprivation of the local area (bottom 50% vs. top 50%), and with matched pairs randomly allocated to use QRISK2 or JBS3 during health checks. Participants A total of 173 patients eligible for NHS Health Check and 15 practitioners. Interventions The health check was delivered using either the QRISK2 10-year risk calculator (usual practice) or the JBS3 lifetime risk calculator, with heart age, event-free survival age and risk score manipulation (intervention). Results Video-recorded health checks were analysed quantitatively (n = 173; JBS3, n = 100; QRISK2, n = 73) and qualitatively (n = 128; n = 64 per group), and video-stimulated recall interviews were undertaken with 40 patients and 15 practitioners, with 10 in-depth case studies. The duration of the health check varied (6.8–38 minutes), but most health checks were short (60% lasting < 20 minutes), with little cardiovascular disease risk discussion (average < 2 minutes). The use of JBS3 was associated with more cardiovascular disease risk discussion and fewer practitioner-dominated consultations than the use of QRISK2. Heart age and visual representations of risk, as used in JBS3, appeared to be better understood by patients than 10-year risk (QRISK2) and, as a result, the use of JBS3 was more likely to lead to discussion of risk factors and their management. Event-free survival age was not well understood by practitioners or patients. However, a lack of effective cardiovascular disease risk discussion in both groups increased the likelihood of a maladaptive coping response (i.e. no risk-reducing behaviour change). In both groups, practitioners often missed opportunities to check patient understanding and to tailor information on cardiovascular disease risk and its management during health checks, confirming apparent practitioner verbal dominance. Limitations The main limitations were under-recruitment in some general practices and the resulting imbalance between groups. Conclusions Communication of cardiovascular disease risk during health checks was brief, particularly when using QRISK2. Patient understanding of and responses to cardiovascular disease risk information were limited. Practitioners need to better engage patients in discussion of and action-planning for their cardiovascular disease risk to reduce misunderstandings. The use of heart age, visual representation of risk and risk score manipulation was generally seen to be a useful way of doing this. Future work could focus on more fundamental issues of practitioner training and time allocation within health check consultations. Trial registration Current Controlled Trials ISRCTN10443908. Funding This project was funded by the National Institute for Health Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 25, No. 50. See the NIHR Journals Library website for further project information.

Cross-sectional assessment of cardiovascular risk factors in patients with knee osteoarthritis

Background: Osteoarthritis (OA) and cardiovascular disease (CVD) are prevalent in India. However, there is dearth of literature among Indians studying the relationship between the two. This study was carried out to assess various cardiovascular (CV) risk factors in patients with knee OA with an objective to investigate their association, screening and management. Methods: In total, 225 patients were included in this cross-sectional study. Participants were diagnosed with knee OA on the basis of the Kellgren and Lawrence (K-L) classification of their radiograph. Participants were also assessed for CV risk factors (age, body mass index, systolic blood pressure, diabetes mellitus, total cholesterol, high-density lipoprotein, smoking) with the help of the Joint British Society QRisk3 calculator (JBS3), which gave three variables: JBS3 risk score, heart age, and life expectancy. Chi Square, Fishers exact test and one-way ANOVA tests were used to compare the categorical and quantitative variables, respectively. Pearson’s correlation coefficient was used to assess the relationship between CV risk factors and knee OA. Results: Patients with severe knee OA had a statistically significantly higher prevalence of CV risk factors (p<0.05). Grade 4 knee OA patients were found to have a mean JBS3 risk of 38%, heart age of 82 years and life expectancy of 77 years as compared to grade 2 patients who had a mean JBS3 risk of 11%, heart age of 63 years and life expectancy of 82 years. Conclusions: Our study concluded that there is a strong positive correlation between knee OA and CVD, with CV risk score being directly proportional to the severity of OA. JBS3 is a comprehensive risk score calculator as well as a screening tool, which produces three more comprehensive variables, namely 10-years risk of developing CVD, physiological heart age and life expectancy.

Cardiovascular disease risk communication in NHS health checks: video-stimulated recall interviews with practitioners

BackgroundNHS Health Check (NHSHC) is a national programme to identify and manage cardiovascular disease (CVD) risk. Practitioners delivering the programme should be competent in discussing CVD risk, but there is evidence of limited understanding of the recommended 10 year/centage CVD risk scores. Lifetime CVD risk calculators might improve understanding and communication of risk.AimTo explore practitioner understanding, perceptions and experiences of CVD risk communication in NHSHCs when using two different CVD risk calculators.Design & settingQualitative video-stimulated recall (VSR) study with NHSHC practitioners.MethodVSR interviews were conducted with practitioners who delivered NHSHCs using either the QRISK2 10-year risk calculator (n=7) or JBS3 lifetime CVD risk calculator (n=8). Data were analysed using reflexive thematic analysis.ResultsFindings from analysis of VSR interviews with 15 practitioners (9 Healthcare Assistants, 6 General Practice Nurses) are presented by risk calculator. There was limited understanding and confidence of 10-year risk, which was used to guide clinical decisions through determining low/medium/high risk thresholds, rather than as a risk communication tool. Potential benefits of some JBS functions were evident, particularly heart age, risk manipulation and visual presentation of risk.ConclusionsThere is a gap between the expectation and reality of practitioners’ understanding, competencies and training in CVD risk communication for NHS Health Check. Practitioners would welcome heart age and risk manipulation functions of JBS3 to promote patient understanding of CVD risk, but there is a more fundamental need for practitioner training in CVD risk communication.

Dissecting heart age using cardiac magnetic resonance videos, electrocardiograms, biobanks, and deep learning

Heart disease is the first cause of death after age 65 and, with the world population aging, its prevalence is expected to starkly increase. We used deep learning to build a heart age predictor on 45,000 heart magnetic resonance videos [MRI] and electrocardiograms [ECG] from the UK Biobank cohort (age range 45-81 years). We predicted age with a root mean squared error [RMSE] of 2.81+/-0.02 years (R-Squared=85.6+/-0.2%) and found that accelerated heart aging is heritable at more than 35%. MRI-based anatomical features predicted age better than ECG-based electro-physiological features (RMSE=2.89+/-0.02 years vs. 6.09+/-.0.02 years), and heart anatomical and electrical aging are weakly correlated (Pearson correlation=.249+/-.002). Our attention maps highlighted the aorta, the mitral valve, and the interventricular septum as key anatomical features driving heart age prediction. We identified genetic (e.g titin gene) and non-genetic correlates (e.g smoking) of accelerated heart aging.