Modern Instrumental Methods of Diagnostics and Risk Assessment of Developing Antitumor Therapy Cardiovasculotoxicity

The most important component of cardio-oncology is the assessment of the risk of development and diagnosis of cardiovascular toxicity of the antitumor therapy, the detection of which is largely based on visualization of the cardiovascular system. The article addresses up-to-date methods of non-invasive visualization of the heart and blood vessels, according to the 2022 European Society of Cardiology Clinical Guidelines on cardio-oncology. Also, the article discusses promising cardiovascular imaging techniques that are not yet included in the guidelines: assessment of coronary calcium using multislice computed tomography and positron emission computed tomography with 18F-labeled 2-deoxy-2-fluoro-d-glucose.

From Atherosclerotic Plaque to Myocardial Infarction-The Leading Cause of Coronary Artery Occlusion

Cardiovascular disease (CVD) constitutes the most common cause of death worldwide. In Europe alone, approximately 4 million people die annually due to CVD. The leading component of CVD leading to mortality is myocardial infarction (MI). MI is classified into several types. Type 1 is associated with atherosclerosis, type 2 results from inadequate oxygen supply to cardiomyocytes, type 3 is defined as sudden cardiac death, while types 4 and 5 are associated with procedures such as percutaneous coronary intervention and coronary artery bypass grafting, respectively. Of particular note is type 1, which is also the most frequently occurring form of MI. Factors predisposing to its occurrence include, among others, high levels of low-density lipoprotein cholesterol (LDL-C) in the blood, cigarette smoking, chronic kidney disease (CKD), diabetes mellitus (DM), hypertension, and familial hypercholesterolaemia (FH). The primary objective of this review is to elucidate the issues with regard to type 1 MI. Our paper delves into, amidst other aspects, its pathogenesis, risk assessment, diagnosis, pharmacotherapy, and interventional treatment options in both acute and long-term conditions.

Implementation of risk-based lipid-lowering therapies in older (age ≥ 65 years) and very-old adults (age ≥ 75 years) with ischemic heart disease in the greater Salzburg region

Introduction: European guidelines recommend the implementation of lipid-lowering therapies (LLTs) in adults (≥ 65 years) with established atherosclerotic cardiovascular disease (ASCVD) and for risk-based primary prevention in older adults (≤ 75 years), yet their use in very-old adults (> 75 years) is controversial, discretionary, and oriented on the presence of risk factors. The aim of this retrospective study is to assess guideline-directed LLT implementation and low-density lipoprotein cholesterol (LDL-C) target achievement in high-/very-high-risk older/very-old adults (65-74 and ≥ 75 years) at presentation for ST-segment elevation myocardial infarction (STEMI) and also to assess evidence-based care delivery to older adults in our region. Methods: All STEMI patients with available LDL-C and total cholesterol presenting for treatment at a large tertiary center in Salzburg, Austria, 2018-2020, were screened (n = 910). High-risk/very-high-risk patients (n = 369) were classified according to European guidelines criteria and divided into cohorts by age: < 65 years (n = 152), 65-74 years (n = 104), and ≥ 75 years (n = 113). Results: Despite being at high-/very-high-risk, prior LLT use was < 40% in the total cohort, with no significant difference by age. Statin monotherapy predominated; 20%-23% of older/very-old adults in the entire cohort were using low-/moderate-intensity stains, 11%-13% were using high-intensity statins, 4% were on ezetimibe therapy, and none were taking proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitors. In the secondary prevention cohort, 53% of older/very-old patients used prior LLTs. Significantly higher percentages of older/oldest ASCVD patients (43% and 49%) met LDL-C targets < 70 mg/dL compared to patients < 65 years (29%; p = 0.033), although just 22% and 30% of these older groups attained stricter LDL-C targets of < 55 mg/dL. Low LLT uptake (16%) among older adults aged 64-74 years for primary prevention resulted in 17% and 10% attainment of risk-based LDL-C targets < 70 mg/dL and < 55 mg/dL, respectively. Oldest adults (≥ 75 years) in both primary and secondary prevention groups more often met risk-based targets than older and younger adults, despite predominantly receiving low-/moderate-intensity statin monotherapy. Conclusion: Secondary prevention was sub-optimal in our region. Less than half of older/very-old adults with established ASCVD met LDL-C targets at the time of STEMI, suggesting severe care-delivery deficits in LLT implementation. Shortcomings in initiation of risk-based LLTs were also observed among high-/very-high-risk primary prevention patients < 75 years, with the achievement of risk-based LDL-C targets in 10%-48% of these patients.

Performance of the non-laboratory based 2019 WHO cardiovascular disease risk prediction chart in Eastern Sub-Saharan Africa

BACKGROUND AND AIMS

The World Health Organization (WHO) updated its cardiovascular disease (CVD) risk prediction charts in 2019 to cover 21 global regions. We aimed to assess the performance of an updated non-lab-based risk chart for people with normoglycaemia, impaired fasting glucose (IFG), and diabetes in Eastern Sub-Saharan Africa.

METHODS AND RESULTS

We used data from six WHO STEPS surveys conducted in Eastern Sub-Saharan Africa between 2012 and 2017. We included 9857 participants aged 40-69 years with no CVD history. The agreement between lab- and non-lab-based charts was assessed using Bland-Altman plots and Cohen's kappa. The median age of the participants was 50 years (25-75th percentile: 44-57). The pooled median 10-year CVD risk was 3 % (25-75th percentile: 2-5) using either chart. According to the estimation, 7.5 % and 8.4 % of the participants showed an estimated CVD risk ≥10 % using the non-lab-based chart or the lab-based chart, respectively. The concordance between the two charts was 91.3 %. The non-lab-based chart underestimated the CVD risk in 57.6 % of people with diabetes. In the Bland-Altman plots, the limits of agreement between the two charts were widest among people with diabetes (-0.57-7.54) compared to IFG (-1.75-1.22) and normoglycaemia (-1.74-1.06). Kappa values of 0.79 (substantial agreement), 0.78 (substantial agreement), and 0.43 (moderate agreement) were obtained among people with normoglycaemia, IFG, and diabetes, respectively.

CONCLUSIONS

Given limited healthcare resources, the updated non-lab-based chart is suitable for CVD risk estimation in the general population without diabetes. Lab-based risk estimation is suitable for individuals with diabetes to avoid risk underestimation.

First Report of Inclisiran Utilization for Hypercholesterolemia Treatment in Real-world Clinical Settings in a Middle East Population

PURPOSE

Inclisiran is the first small interfering RNA-based treatment approved for reducing pro-atherogenic lipoproteins in patients with heterozygous familial hypercholesterolemia or clinical atherosclerotic cardiovascular disease, who require additional lowering of low-density lipoprotein cholesterol (LDL-C). We report the first evaluation of the effects of inclisiran in a Middle East population.

METHODS

We conducted a retrospective review of patients initiating inclisiran treatment at an outpatient diabetology, endocrinology, and cardiology center between May 2021 and December 2022. All patients followed up for 90 days or more, or with at least 1 lipid determination post initiation were included. Participants were categorized into primary prevention (n = 57) and secondary prevention (n = 89) groups according to previous atherosclerotic cardiovascular disease.

FINDINGS

Inclisiran was initiated in 146 patients; mean (SD) age was 54.8 (12.12) years, 82 patients (56.2%) were male, 28 patients (19.2%) had received a diagnosis of familial hypercholesterolemia, 89 patients (61%) had received a diagnosis of diabetes mellitus, and 35 patients (23.9%) had a statin intolerance. Median (interquartile range) follow-up was 137 (90 to 193) days. At 90 days, median (interquartile range) reductions in serum LDL-C and triglycerides were -37.9% (-9.5% to -51.2%) and -12.0% (-9.8% to -40.5%), respectively, in primary prevention and -54.1% (-17.1% to -71.4%) and -15.3% (-14% to -38.8%), respectively, in secondary prevention (all, P < 0.001). LDL-C goals were attained in 110 patients (75.3%). Nonattainment of LDL-C goal was attributed to system effect in 26 patients (72.2%), biological effect in 5 patients (13.9%), and discontinuation of treatment in 5 patients (13.9%). Therapy was well tolerated.

IMPLICATIONS

This study is the first from the Middle East and North Africa region that reported the real-world efficacy and safety profile of inclisiran in a mixed-risk population of patients with heterozygous familial hypercholesterolemia and other non-familial hypercholesterolemia indications. Clinically meaningful and sustained reductions in pro-atherogenic lipids with good tolerability were observed after inclisiran initiation. Fewer AEs were reported in this predominantly Arabic population, consistent with previous safety reports for inclisiran. It is important to note that no patient stopped inclisiran treatment due to AEs. Strengths of our study included an optimal cohort, patient heterogeneity, and high retention. In addition, we were able to report mean robust effects of inclisiran and good medication tolerability, quite like randomized studies and open-label extension periods. Despite this, our study had some limitations, including selection bias due to the retrospective design and the absence of a comparative group.

Quantifying cardiovascular disease risk and heart age predictions for men in the prison environment

OBJECTIVES

Cardiovascular disease (CVD) and associated risk factors within the prison population often present at a younger age in this cohort. Given CVD is largely preventable, it warrants investigation to fully quantify this risk. This study explored the relative predicted 10-year CVD risk and examined the calculated heart age in a representative sample of male individuals aged 25-84 years within the prison environment.

STUDY DESIGN

This was a cross-sectional study.

METHODS

Data were collected on 299 men who underwent a cardiometabolic risk assessment in HMP Parc, Bridgend. The QRISK2 algorithm was used to calculate 10-year CVD risk, relative risk (to general population) and the predicted heart age of an individual. Between-group differences (prison population vs general community) in cardiovascular risk predictions (10-year CVD risk and heart age) were assessed.

RESULTS

We observed that at all age groups, the relative risk of predicted 10-year CVD scores in the prison population was double that of the community risk (2.1 ± 0.6), and this was most apparent in the oldest age group (≥50 years: 17.0% compared to 8.8%; P < 0.001). Overall, the heart age of the sample was 7.5 (6.7-8.2) years higher than their own chronological age, and this difference increased to above 9 years in those aged ≥40 years.

CONCLUSIONS

This study provides quantifiable evidence to the elevated CVD risk in prison. Heart age predictions were almost a decade higher in those aged ≥40 years. Lowering the screening age for CVD by around 5 years in the prison population should be considered.

Validity of the models predicting 10-year risk of cardiovascular diseases in Asia: A systematic review and prediction model meta-analysis

We aimed to review the validity of existing prediction models for cardiovascular diseases (CVDs) in Asia. In this systematic review and meta-analysis, we included studies that validated prediction models for CVD risk in the general population in Asia. Various databases, including PubMed, Web of Science conference proceedings citation index, Scopus, Global Index Medicus of the World Health Organization (WHO), and Open Access Thesis and Dissertations (OATD), were searched up to November 2022. Additional studies were identified through reference lists and related reviews. The risk of bias was assessed using the PROBAST prediction model risk of bias assessment tool. Meta-analyses were performed using the random effects model, focusing on the C-statistic as a discrimination index and the observed-to-expected ratio (OE) as a calibration index. Out of 1315 initial records, 16 studies were included, with 21 external validations of six models in Asia. The validated models consisted of Framingham models, pooled cohort equations (PCEs), SCORE, Globorisk, and WHO models, combined with the results of the first four models. The pooled C-statistic for men ranged from 0.72 (95% CI 0.70 to 0.75; PCEs) to 0.76 (95% CI 0.74 to 0.78; Framingham general CVD). In women, it varied from 0.74 (95% CI 0.22 to 0.97; SCORE) to 0.79 (95% CI 0.74 to 0.83; Framingham general CVD). The pooled OE ratio for men ranged from 0.21 (95% CI 0.018 to 2.49; Framingham CHD) to 1.11 (95%CI 0.65 to 1.89; PCEs). In women, it varied from 0.28 (95%CI 0.33 to 2.33; Framingham CHD) to 1.81 (95% CI 0.90 to 3.64; PCEs). The Framingham, PCEs, and SCORE models exhibited acceptable discrimination but poor calibration in predicting the 10-year risk of CVDs in Asia. Recalibration and updates are necessary before implementing these models in the region.

Shared Decision-Making and Cardiovascular Health: A Scientific Statement From the American Heart Association

Shared decision-making is increasingly embraced in health care and recommended in cardiovascular guidelines. Patient involvement in health care decisions, patient-clinician communication, and models of patient-centered care are critical to improve health outcomes and to promote equity, but formal models and evaluation in cardiovascular care are nascent. Shared decision-making promotes equity by involving clinicians and patients, sharing the best available evidence, and recognizing the needs, values, and experiences of individuals and their families when faced with the task of making decisions. Broad endorsement of shared decision-making as a critical component of high-quality, value-based care has raised our awareness, although uptake in clinical practice remains suboptimal for a range of patient, clinician, and system issues. Strategies effective in promoting shared decision-making include educating clinicians on communication techniques, engaging multidisciplinary medical teams, incorporating trained decision coaches, and using tools (ie, patient decision aids) at appropriate literacy and numeracy levels to support patients in their cardiovascular decisions. This scientific statement shines a light on the limited but growing body of evidence of the impact of shared decision-making on cardiovascular outcomes and the potential of shared decision-making as a driver of health equity so that everyone has just opportunities. Multilevel solutions must align to address challenges in policies and reimbursement, system-level leadership and infrastructure, clinician training, access to decision aids, and patient engagement to fully support patients and clinicians to engage in the shared decision-making process and to drive equity and improvement in cardiovascular outcomes.

Analysis of the Nursing Effect of Anesthesia Care Integration Combined with Preventive Nursing on Older Patients with Lumbar Disc Herniation During the Perioperative Period

Objective

This study explored the nursing effect of anesthesia care integration combined with preventive nursing on older patients with perioperative lumbar disc herniation (LDH).

Methods

Clinical data of 100 older patients with LDH who were admitted to our hospital between May 2017 and May 2022 were used, and there were no patients who had not had surgery between January and May 2020 because of the COVID-19 pandemic. Based on the different nursing methods, the patients were divided into control and observation groups, with 50 cases each. The control group received anesthesia care integration, whereas the observation group received anesthesia care integration combined with preventive nursing. Lumbar spine function, pain score, anesthesia recovery assessment, and nursing effects were compared between the two groups.

Results

The scores of the anesthesia recovery assessment of the two groups were compared, and the vital signs of the observation group during recovery from anesthesia were significantly better than those of the control group (P<0.05). After nursing care, the Japanese Orthopaedic Association (JOA) score of the observation group was significantly higher than that of the control group; however, the numerical scale (NRS) score of the observation group was significantly lower than that of the control group (P<0.05). After nursing care, the physical comfort, emotional state, psychological support, self-care ability, and pain scores were higher in the observation group than in the control group; however, the NRS score of the observation group was significantly lower than that of the control group (P<0.05).

Conclusion

Anesthesia care integration combined with preventive nursing has a positive effect on older patients with perioperative LDH, and it significantly improves lumbar spine function, reduces pain, shortens recovery time, and benefits physical and mental health.

Coronary Artery Calcium Score and Polygenic Risk Score for the Prediction of Coronary Heart Disease Events

Importance

Coronary artery calcium score and polygenic risk score have each separately been proposed as novel markers to identify risk of coronary heart disease (CHD), but no prior studies have directly compared these markers in the same cohorts.

Objective

To evaluate change in CHD risk prediction when a coronary artery calcium score, a polygenic risk score, or both are added to a traditional risk factor-based model.

Design, Setting, and Participants

Two observational population-based studies involving individuals aged 45 years through 79 years of European ancestry and free of clinical CHD at baseline: the Multi-Ethnic Study of Atherosclerosis (MESA) study involved 1991 participants at 6 US centers and the Rotterdam Study (RS) involved 1217 in Rotterdam, the Netherlands.

Exposure

Traditional risk factors were used to calculate CHD risk (eg, pooled cohort equations [PCEs]), computed tomography for the coronary artery calcium score, and genotyped samples for a validated polygenic risk score.

Main Outcomes and Measures

Model discrimination, calibration, and net reclassification improvement (at the recommended risk threshold of 7.5%) for prediction of incident CHD events were assessed.

Results

The median age was 61 years in MESA and 67 years in RS. Both log (coronary artery calcium+1) and polygenic risk score were significantly associated with 10-year risk of incident CHD (hazards ratio per SD, 2.60; 95% CI, 2.08-3.26 and 1.43; 95% CI, 1.20-1.71, respectively), in MESA. The C statistic for the coronary artery calcium score was 0.76 (95% CI, 0.71-0.79) and for the polygenic risk score, 0.69 (95% CI, 0.63-0.71). The change in the C statistic when each was added to the PCEs was 0.09 (95% CI, 0.06-0.13) for the coronary artery calcium score, 0.02 (95% CI, 0.00-0.04) for the polygenic risk score, and 0.10 (95% CI, 0.07-0.14) for both. Overall categorical net reclassification improvement was significant when the coronary artery calcium score (0.19; 95% CI, 0.06-0.28) but was not significant when the polygenic risk score (0.04; 95% CI, -0.05 to 0.10) was added to the PCEs. Calibration of the PCEs and models with coronary artery calcium and/or polygenic risk scores was adequate (all χ2<20). Subgroup analysis stratified by the median age demonstrated similar findings. Similar findings were observed for 10-year risk in RS and in longer-term follow-up in MESA (median, 16.0 years).

Conclusions and Relevance

In 2 cohorts of middle-aged to older adults from the US and the Netherlands, the coronary artery calcium score had better discrimination than the polygenic risk score for risk prediction of CHD. In addition, the coronary artery calcium score but not the polygenic risk score significantly improved risk discrimination and risk reclassification for CHD when added to traditional risk factors.

Cost-Effectiveness of the CNIC-Polypill Strategy Compared With Separate Monocomponents in Secondary Prevention of Cardiovascular and Cerebrovascular Disease in Portugal: The MERCURY Study

Background: Cardiovascular (CV) diseases remain a leading and costly cause of death globally. Patients with previous CV events are at high risk of recurrence. Secondary prevention therapies improve CV risk factor control and reduce disease costs. Objectives: To assess the cost-effectiveness of a CV polypill strategy (CNIC-Polypill) compared with the loose combination of monocomponents to improve the control of CV risk factors in patients with previous coronary heart disease or stroke. Methods: A Markov model cost-utility analysis was developed using 4 health states, SMART risk equation, and 3-month cycles for year 1 and annual cycles thereafter, over a lifetime horizon from the perspective of the National Health System in Portugal (base case). The NEPTUNO study, Portuguese registries, mortality tables, official reports, and the literature were consulted to define effectiveness, epidemiological costs, and utility data. Outcomes were costs (estimated in 2020 euros) per life-year (LY) and quality-adjusted LY (QALY) gained. A 4% discount rate was applied. Alternative scenarios and one-way and probabilistic sensitivity analyses tested the consistency and robustness of results. Results: The CNIC-Polypill strategy in secondary prevention provides more LY and QALY, at a higher cost, than monocomponents. The incremental cost-utility ratio is €1557/QALY gained. Assuming a willingness-to-pay threshold of €30 000/QALY gained, there is a 79.7% and a 44.4% probability of the CNIC-Polypill being cost-effective and cost-saving, respectively, compared with the loose combination of monocomponents. Results remain consistent in the alternative scenarios and robust in the sensitivity analyses. Discussion: The model reflects increments in the number of years patients would live and in quality of life with the CNIC-Polypill. The clinical effectiveness of the CNIC-Polypill strategy initially demonstrated in the NEPTUNO study has been recently corroborated in the SECURE trial. The incremental cost of the CNIC-Polypill strategy emerges slightly above the comparator, but willingness-to-pay estimates and sensitivity analyses indicate that the CNIC-Polypill strategy is consistently cost-effective compared with monocomponents and remains within acceptable affordability margins. Conclusion: The CNIC-Polypill is a cost-effective secondary prevention strategy. In patients with histories of coronary heart disease or stroke, the CNIC-Polypill more effectively controls CV risk factors compared with monocomponents.

The First Report of a Real-World Experience With a PCSK9 Inhibitor in a Large Familial Hyperlipidemia and Very-High-Risk Middle Eastern Population

PURPOSE

Evolocumab, a monoclonal inhibitor of proprotein convertase subtilisin/kexin 9, has been shown to reduce proatherogenic lipoproteins in patients with or without familial hypercholesterolemia (FH), diabetes mellitus, or atherosclerotic cardiovascular disease (ASCVD). We explored the safety profile and clinical effectiveness of evolocumab in an outpatient population of Emirati individuals with FH diagnosed per Dutch Lipid Clinic Network criteria, previous ASCVD, or statin intolerance.

METHODS

This study was a retrospective review of patients initiating evolocumab treatment for any indication at Imperial College London Diabetes Centre between 2017 and 2020. All individuals followed up for at least 90 days or with at least one lipid panel postinitiation were included. Participants were subclassified into primary prevention (no previous ASCVD event, n = 81) and secondary prevention (any prior clinical ASCVD event, n = 102) groups.

FINDINGS

Evolocumab was initiated in 183 individuals (mean [SD] age, 51.5 [12.4] years; 51% male); 108 (59%) had a clinical or genetic FH diagnosis, and 70.5% had diabetes mellitus. Statin intolerance was a treatment indication in 60 (32.8%) individuals. At 90 days, substantial reductions in serum LDL-C, triglycerides (TG), and total cholesterol:HDL-C (TC:HDL-C) were observed in both the primary and secondary prevention groups, and both FH and non-FH individuals. In the primary prevention group, median (interquartile range) reduction in LDL-C was 43.7% (10.8%; 63.0%); TG, 15.0% (7.2%; 35.3%); and TC:HDL-C, 31.5% (11.1%; 46.0%). In the secondary prevention group, median (interquartile range) reduction in LDL-C was 48.3% (22%; 70%); TG, 19.6% (1.2%; 32.5%); and TC:HDL-C, 32.6% (14.6%; 46.3%) (all, P < 0.0001). American College of Cardiology/American Heart Association LDL-C targets were consistently achieved in 114 (62.3%) patients during a follow-up of 359 (79-639) days. Nonattainment of the LDL-C target was attributed to nonadherence in 36 (52.2%) patients and discontinuation of treatment in 14 (20.3%) patients. Evolocumab was discontinued in 4 patients because of adverse events.

IMPLICATIONS

This study is the first from the Middle East and North Africa region that reports the real-world efficacy of evolocumab in a mixed risk population of individuals with FH and other non-FH indications. Clinically meaningful and sustained reductions in LDL-C, TG, and cholesterol ratios were observed after evolocumab initiation. Few adverse events were reported in this predominantly Arabic population, consistent with previous safety reports for evolocumab. Notable strengths of this study include a relatively large cohort, patient heterogeneity and high retention, and a minimum follow-up of 1 year. Despite these strengths, the study has some limitations, including the selection bias due to the retrospective design and absence of comparative group. (Clin Ther. 2022;44:XXX-XXX) © 2022 Elsevier HS Journals, Inc.

Relationships between work, lifestyles, and obesity: cross-sectional study based on the 2017 Spanish National Health Survey

AIMS

Overweight and obesity are affected by factors such as dietary transgressions and lack of exercise, as well as less obvious factors such as shift work and sedentary working. (i) To explore the associations between overweight and obesity (O/O) with working conditions such as type of job position, and work shifts; (ii) to know the associations between O/O with lifestyles (physical activity, eating habits, and tobacco and alcohol consumption).

METHODS AND RESULTS

A cross-sectional, population-based design was used based on the Spanish National Health Survey. The final sample consisted of 9097 workers aged between 18 and 65. The influence of the explanatory variables on the outcome variable (body mass index) was examined using logistic regression. The prevalence of O/O increased by 4% for every additional year of age (P < 0.001) and was 3.1 times more frequent among men (P < 0.001). It also was 20% higher in night work or rotating shifts (P < 0.01), and 14.9% lower in sedentary job positions (P < 0.05). Low consumption of the following foods was significantly associated with a reduction in O/O: meat (16.2%), fish (75.4%), cured meats (35.1%), dairy products (33.3%), and snacks (47.1%).

CONCLUSION

Age, men, engaging in moderate or low levels of physical activity, being an ex-smoker, working during the night or rotating shifts, holding a non-sedentary job position, having a low to moderate consumption of pasta, potatoes, and rice, a moderate consumption of bread and cereals, as well as dairy products, vegetables, sugary foods, and fast food, have been significantly associated with a higher prevalence of O/O. No relationships were identified between the consumption of alcohol, fruit, eggs, pulses, and sugary soft drinks.

Predicting 30-day mortality after ST elevation myocardial infarction: Machine learning- based random forest and its external validation using two independent nationwide datasets

BACKGROUND

Various prognostic models for mortality prediction following ST-segment elevation myocardial infarction (STEMI) have been developed over the past two decades. Our group has previously demonstrated that machine learning (ML)-based models can outperform known risk scores for 30-day mortality post-STEMI. The study aimed to redevelop an ML-based random forest prediction model for 30-day mortality post-STEMI and externally validate it on a large cohort.

METHODS

This was a retrospective, supervised learning, data mining study developed on the Acute Coronary Syndrome Israeli Survey (ACSIS) registry and the Myocardial Ischemia National Audit Project (MINAP) for external validation. Patients included received reperfusion therapy for STEMI between 2006 and 2016. Discrimination and calibration performances were assessed for two developed models and compared with the Global Registry of Acute Cardiac Events (GRACE) score.

RESULTS

The ACSIS cohort (2,782 included /15,212 total) and MINAP cohort (22,693 included/735,000 total) were significantly different in most variables, yet similar in 30-day mortality rate (4.3-4.4%). Random forest models were developed on the ACSIS cohort with a full model including all 32 variables and a simple model including the 10 most important ones. Features' importance was calculated using the varImp function measuring how much each feature contributes to the data's homogeneity. Applying the optimized models on the MINAP validation cohort showed high discrimination of area under the curve (AUC) = 0.804 (0.786-0.822) for the full model, and AUC = 0.787 (0.748-0.780) using the simple model, compared with the GRACE risk score discrimination of AUC = 0.764 (0.748-0.780). All models were not well calibrated for the MINAP data. Following Platt scaling on 20% of the MINAP data, the random forest models calibration improved while the GRACE calibration did not change.

CONCLUSIONS

The random forest predictive model for 30-day mortality post STEMI, developed on the ACSIS national registry, has been validated in the MINAP large external cohort and can be applied early at admission for risk stratification. The model performed better than the commonly used GRACE score. Furthermore, to the best of our knowledge, this is the first externally validated ML-based model for STEMI.

Age and sex affect deep learning prediction of cardiometabolic risk factors from retinal images

Deep neural networks can extract clinical information, such as diabetic retinopathy status and individual characteristics (e.g. age and sex), from retinal images. Here, we report the first study to train deep learning models with retinal images from 3,000 Qatari citizens participating in the Qatar Biobank study. We investigated whether fundus images can predict cardiometabolic risk factors, such as age, sex, blood pressure, smoking status, glycaemic status, total lipid panel, sex steroid hormones and bioimpedance measurements. Additionally, the role of age and sex as mediating factors when predicting cardiometabolic risk factors from fundus images was studied. Predictions at person-level were made by combining information of an optic disc centred and a macula centred image of both eyes with deep learning models using the MobileNet-V2 architecture. An accurate prediction was obtained for age (mean absolute error (MAE): 2.78 years) and sex (area under the curve: 0.97), while an acceptable performance was achieved for systolic blood pressure (MAE: 8.96 mmHg), diastolic blood pressure (MAE: 6.84 mmHg), Haemoglobin A1c (MAE: 0.61%), relative fat mass (MAE: 5.68 units) and testosterone (MAE: 3.76 nmol/L). We discovered that age and sex were mediating factors when predicting cardiometabolic risk factors from fundus images. We have found that deep learning models indirectly predict sex when trained for testosterone. For blood pressure, Haemoglobin A1c and relative fat mass an influence of age and sex was observed. However, achieved performance cannot be fully explained by the influence of age and sex. In conclusion we confirm that age and sex can be predicted reliably from a fundus image and that unique information is stored in the retina that relates to blood pressure, Haemoglobin A1c and relative fat mass. Future research should focus on stratification when predicting person characteristics from a fundus image.

Optimising implementation of European guidelines on cardiovascular disease prevention in clinical practice: what is needed?

Cardiovascular disease is a model example of a preventable condition for which practice guidelines are particularly important. In 2016, the joint task force created by the European Society of Cardiology (ESC) together with 10 other societies released the new version of the European guidelines on cardiovascular disease prevention. To facilitate the implementation of the ESC guidelines, a dedicated prevention implementation committee has been established within the European Association of Preventive Cardiology. The paper will first explore potential barriers to the guidelines’ implementation. It then develops a discussion that seeks to inform the future development of the committee’s work, including a new definition of the guidelines’ stakeholders (health policy-makers, healthcare professionals and health educators, patient organisations, entrepreneurs and the general public), future activities within four specific areas: strengthening awareness of the guidelines among stakeholders; supporting organisational changes to facilitate the guidelines’ implementation; motivating stakeholders to utilise the guidelines; and present ideas on new implementation strategies. Providing multifaceted cooperation between healthcare professionals, healthcare management executives and health policy-makers, the novel approach proposed in this paper should contribute to a wider use of the 2016 ESC guidelines and produce desired effects of less cardiovascular disease morbidity and mortality. Furthermore, the solutions presented within the paper may constitute a benchmark for the implementation of practice guidelines in other medical disciplines.