An analysis of calibration and discrimination among multiple cardiovascular risk scores in a modern multiethnic cohort

Cardiovascular risk in US adults with nonalcoholic steatohepatitis (NASH) vs. matched non-NASH controls, National Health and Nutrition Examination Survey, 2017-2020

BACKGROUND

NASH is considered a contributor to atherosclerotic cardiovascular disease (ASCVD) risk; however, its contribution beyond traditional risk factors for CVD, particularly diabetes, is less clearly understood. This study aimed to quantify the cardiovascular-event risk associated with NASH, independent of diabetes status.

METHODS

A cross-sectional analysis was conducted using the 2017-2020 NHANES pre-pandemic cycle. NASH was defined based on presence of steatosis without other causes of liver disease, and FibroScan+AST score from vibration-controlled transient elastography (VCTE). Significant fibrosis (stages F2-F4) was identified by liver stiffness measurement from VCTE. Predicted primary CV-event risk was estimated using both the Pooled Cohort Equations (PCE) and the Framingham Risk Score (FRS). NASH patients were matched with non-NASH controls on age, sex, race/ethnicity, and diabetes status. Weighted logistic regression was conducted, modeling elevated predicted CV risk (binary) as the dependent variable and indicators for NASH / fibrosis stages as independent variables.

RESULTS

A sample of 125 NASH patients was matched with 2585 controls. NASH with significant fibrosis was associated with elevated predicted 10-year CV risk, although this association was only statistically significant in PCE analyses (odds ratio and 95% CI 2.34 [1.25, 4.36]). Analyses restricting to ages <65 years showed similar results, with associations of greater magnitude.

CONCLUSION

Independent of diabetes, a significant association was observed between NASH with significant liver fibrosis and predicted primary CV-event risk in US adults, particularly for those <65. These findings suggest the importance of accounting for NASH and liver-fibrosis stage in predicting CV-event risk.

The impact of diversity, equity, and inclusion on spinal research - asking different questions

BACKGROUND CONTEXT

In recent years, the field of spine surgery has seen significant advancements in surgical techniques alongside a growing emphasis on diversity, equity, and inclusion (DEI).

PURPOSE

This paper explores the significant impact of DEI on spine surgery, recognizing its potential to drive innovation, improve patient outcomes, and address healthcare disparities.

STUDY DESIGN

Review.

SIGN/SETTING

The review focuses on the impact of DEI on spine surgery, examining how diverse perspectives influence research and practice in the field.

PATIENT SAMPLE

Not applicable.

OUTCOME MEASURES

Not applicable.

METHODS

The review analyzes the role of DEI in driving innovation and improving patient outcomes in spine surgery and discusses ongoing challenges such as unconscious biases and systemic barriers.

RESULTS

Shifting paradigms in research through diverse perspectives is crucial for broadening the scope of inquiry and challenging existing standards. Efforts to promote diversity, including targeted outreach and mentorship initiatives, are essential for cultivating a more inclusive workforce.

CONCLUSIONS

Embracing diverse perspectives and asking unconventional questions are vital for achieving a comprehensive understanding of spinal health and delivering equitable healthcare. Ongoing challenges highlight the need for continued commitment to DEI principles.

What Causes Premature Coronary Artery Disease?

PURPOSE OF REVIEW

This review provides an overview of genetic and non-genetic causes of premature coronary artery disease (pCAD).

RECENT FINDINGS

pCAD refers to coronary artery disease (CAD) occurring before the age of 65 years in women and 55 years in men. Both genetic and non-genetic risk factors may contribute to the onset of pCAD. Recent advances in the genetic epidemiology of pCAD have revealed the importance of both monogenic and polygenic contributions to pCAD. Familial hypercholesterolemia (FH) is the most common monogenic disorder associated with atherosclerotic pCAD. However, clinical overreliance on monogenic genes can result in overlooked genetic causes of pCAD, especially polygenic contributions. Non-genetic factors, notably smoking and drug use, are also important contributors to pCAD. Cigarette smoking has been observed in 25.5% of pCAD patients relative to 12.2% of non-pCAD patients. Finally, myocardial infarction (MI) associated with spontaneous coronary artery dissection (SCAD) may result in similar clinical presentations as atherosclerotic pCAD. Recognizing the genetic and non-genetic causes underlying pCAD is important for appropriate prevention and treatment. Despite recent progress, pCAD remains incompletely understood, highlighting the need for both awareness and research.

Atherosclerosis evaluation and cardiovascular risk estimation using coronary computed tomography angiography

Clinical risk scores based on traditional risk factors of atherosclerosis correlate imprecisely to an individual's complex pathophysiological predisposition to atherosclerosis and provide limited accuracy for predicting major adverse cardiovascular events (MACE). Over the past two decades, computed tomography scanners and techniques for coronary computed tomography angiography (CCTA) analysis have substantially improved, enabling more precise atherosclerotic plaque quantification and characterization. The accuracy of CCTA for quantifying stenosis and atherosclerosis has been validated in numerous multicentre studies and has shown consistent incremental prognostic value for MACE over the clinical risk spectrum in different populations. Serial CCTA studies have advanced our understanding of vascular biology and atherosclerotic disease progression. The direct disease visualization of CCTA has the potential to be used synergistically with indirect markers of risk to significantly improve prevention of MACE, pending large-scale randomized evaluation.

Socioeconomic disadvantage in pregnancy and postpartum risk of cardiovascular disease

BACKGROUND

Pregnancy is an educable and actionable life stage to address social determinants of health (SDOH) and lifelong cardiovascular disease (CVD) prevention. However, the link between a risk score that combines multiple neighborhood-level social determinants in pregnancy and the risk of long-term CVD remains to be evaluated.

OBJECTIVE

To examine whether neighborhood-level socioeconomic disadvantage measured by the Area Deprivation Index (ADI) in early pregnancy is associated with a higher 30-year predicted risk of CVD postpartum, as measured by the Framingham Risk Score.

STUDY DESIGN

An analysis of data from the prospective Nulliparous Pregnancy Outcomes Study-Monitoring Mothers-to-Be Heart Health Study longitudinal cohort. Participant home addresses during early pregnancy were geocoded at the Census-block level. The exposure was neighborhood-level socioeconomic disadvantage using the 2015 ADI by tertile (least deprived [T1], reference; most deprived [T3]) measured in the first trimester. Outcomes were the predicted 30-year risks of atherosclerotic cardiovascular disease (ASCVD, composite of fatal and nonfatal coronary heart disease and stroke) and total CVD (composite of ASCVD plus coronary insufficiency, angina pectoris, transient ischemic attack, intermittent claudication, and heart failure) using the Framingham Risk Score measured 2 to 7 years after delivery. These outcomes were assessed as continuous measures of absolute estimated risk in increments of 1%, and, secondarily, as categorical measures with high-risk defined as an estimated probability of CVD ≥10%. Multivariable linear regression and modified Poisson regression models adjusted for baseline age and individual-level social determinants, including health insurance, educational attainment, and household poverty.

RESULTS

Among 4309 nulliparous individuals at baseline, the median age was 27 years (interquartile range [IQR]: 23-31) and the median ADI was 43 (IQR: 22-74). At 2 to 7 years postpartum (median: 3.1 years, IQR: 2.5, 3.7), the median 30-year risk of ASCVD was 2.3% (IQR: 1.5, 3.5) and of total CVD was 5.5% (IQR: 3.7, 7.9); 2.2% and 14.3% of individuals had predicted 30-year risk ≥10%, respectively. Individuals living in the highest ADI tertile had a higher predicted risk of 30-year ASCVD % (adjusted ß: 0.41; 95% confidence interval [CI]: 0.19, 0.63) compared with those in the lowest tertile; and those living in the top 2 ADI tertiles had higher absolute risks of 30-year total CVD % (T2: adj. ß: 0.37; 95% CI: 0.03, 0.72; T3: adj. ß: 0.74; 95% CI: 0.36, 1.13). Similarly, individuals living in neighborhoods in the highest ADI tertile were more likely to have a high 30-year predicted risk of ASCVD (adjusted risk ratio [aRR]: 2.21; 95% CI: 1.21, 4.02) and total CVD ≥10% (aRR: 1.35; 95% CI: 1.08, 1.69).

CONCLUSION

Neighborhood-level socioeconomic disadvantage in early pregnancy was associated with a higher estimated long-term risk of CVD postpartum. Incorporating aggregated SDOH into existing clinical workflows and future research in pregnancy could reduce disparities in maternal cardiovascular health across the lifespan, and requires further study.

Prognostic potentials of AI in ophthalmology: systemic disease forecasting via retinal imaging

BACKGROUND

Artificial intelligence (AI) that utilizes deep learning (DL) has potential for systemic disease prediction using retinal imaging. The retina's unique features enable non-invasive visualization of the central nervous system and microvascular circulation, aiding early detection and personalized treatment plans for personalized care. This review explores the value of retinal assessment, AI-based retinal biomarkers, and the importance of longitudinal prediction models in personalized care.

MAIN TEXT

This narrative review extensively surveys the literature for relevant studies in PubMed and Google Scholar, investigating the application of AI-based retina biomarkers in predicting systemic diseases using retinal fundus photography. The study settings, sample sizes, utilized AI models and corresponding results were extracted and analysed. This review highlights the substantial potential of AI-based retinal biomarkers in predicting neurodegenerative, cardiovascular, and chronic kidney diseases. Notably, DL algorithms have demonstrated effectiveness in identifying retinal image features associated with cognitive decline, dementia, Parkinson's disease, and cardiovascular risk factors. Furthermore, longitudinal prediction models leveraging retinal images have shown potential in continuous disease risk assessment and early detection. AI-based retinal biomarkers are non-invasive, accurate, and efficient for disease forecasting and personalized care.

CONCLUSION

AI-based retinal imaging hold promise in transforming primary care and systemic disease management. Together, the retina's unique features and the power of AI enable early detection, risk stratification, and help revolutionizing disease management plans. However, to fully realize the potential of AI in this domain, further research and validation in real-world settings are essential.

Development and validation of a new algorithm for improved cardiovascular risk prediction

QRISK algorithms use data from millions of people to help clinicians identify individuals at high risk of cardiovascular disease (CVD). Here, we derive and externally validate a new algorithm, which we have named QR4, that incorporates novel risk factors to estimate 10-year CVD risk separately for men and women. Health data from 9.98 million and 6.79 million adults from the United Kingdom were used for derivation and validation of the algorithm, respectively. Cause-specific Cox models were used to develop models to predict CVD risk, and the performance of QR4 was compared with version 3 of QRISK, Systematic Coronary Risk Evaluation 2 (SCORE2) and atherosclerotic cardiovascular disease (ASCVD) risk scores. We identified seven novel risk factors in models for both men and women (brain cancer, lung cancer, Down syndrome, blood cancer, chronic obstructive pulmonary disease, oral cancer and learning disability) and two additional novel risk factors in women (pre-eclampsia and postnatal depression). On external validation, QR4 had a higher C statistic than QRISK3 in both women (0.835 (95% confidence interval (CI), 0.833-0.837) and 0.831 (95% CI, 0.829-0.832) for QR4 and QRISK3, respectively) and men (0.814 (95% CI, 0.812-0.816) and 0.812 (95% CI, 0.810-0.814) for QR4 and QRISK3, respectively). QR4 was also more accurate than the ASCVD and SCORE2 risk scores in both men and women. The QR4 risk score identifies new risk groups and provides superior CVD risk prediction in the United Kingdom compared with other international scoring systems for CVD risk.

Transforming the cardiometabolic disease landscape: Multimodal AI-powered approaches in prevention and management

The rise of artificial intelligence (AI) has revolutionized various scientific fields, particularly in medicine, where it has enabled the modeling of complex relationships from massive datasets. Initially, AI algorithms focused on improved interpretation of diagnostic studies such as chest X-rays and electrocardiograms in addition to predicting patient outcomes and future disease onset. However, AI has evolved with the introduction of transformer models, allowing analysis of the diverse, multimodal data sources existing in medicine today. Multimodal AI holds great promise in more accurate disease risk assessment and stratification as well as optimizing the key driving factors in cardiometabolic disease: blood pressure, sleep, stress, glucose control, weight, nutrition, and physical activity. In this article we outline the current state of medical AI in cardiometabolic disease, highlighting the potential of multimodal AI to augment personalized prevention and treatment strategies in cardiometabolic disease.

Impact of Race on Classification of Atherosclerotic Risk Using a National Cardiovascular Risk Prediction Tool

Introduction

The use of race in clinical risk prediction tools may exacerbate racial disparities in healthcare access and outcomes. This study quantified the number of individuals reclassified for primary prevention of cardiovascular disease owing to a change in their race alone on the basis of a commonly used risk prediction tool.

Methods

This is a cross-sectional analysis of individuals aged 40-75 years without a history of cardiovascular events, diabetes, or other high-risk features using the 2005-2018 National Health and Nutritional Examination Survey. Authors compared atherosclerotic cardiovascular disease risk scores using the American Heart Association/American College of Cardiology equation recommended for White individuals or individuals of other races with that recommended for Black individuals.

Results

A total of 2,946 White individuals; 1,361 Black individuals; and 2,495 individuals of other races were included in the analysis. Using the American Heart Association/American College of Cardiology equation, the mean 10-year atherosclerotic cardiovascular disease risk was 5.80% (95% CI=5.54, 6.06) for White individuals, 7.04% (956% CI=6.69, 7.39) for Black individuals, and 4.93% (95% CI=4.61, 5.24) for individuals of other races. When using the American Heart Association/American College of Cardiology equation designated for the opposite race (White/other race versus Black), the mean atherosclerotic cardiovascular disease risk score increased by 1.02% (95% CI=0.90, 1.13) for White individuals, decreased by 1.82% (95% CI= -1.67, -1.96) for Black individuals, and increased by 0.98% (95% CI=0.85, 1.10) for individuals of other races. When using clinical atherosclerotic cardiovascular disease categories of <7.5%, 7.5%-10%, and >10%, 16.93% of all individuals were reclassified when using the American Heart Association/American College of Cardiology's equation designated for the opposite race.

Conclusions

Changing race within a commonly used cardiovascular risk prediction tool results in significant changes in risk classification among eligible White and Black individuals in the U.S.

Navigating the Landscape of Cardiovascular Risk Scores: A Comparative Analysis of Eight Risk Prediction Models in a High-Risk Cohort in Lithuania

Background: Numerous cardiovascular risk prediction models (RPM) have been developed, however, agreement studies between these models are scarce. We aimed to assess the inter-model agreement between eight RPMs: assessing cardiovascular risk using SIGN, the Australian CVD risk score (AusCVDRisk), the Framingham Risk Score for Hard Coronary Heart Disease, the Multi-Ethnic Study of Atherosclerosis risk score, the Pooled Cohort Equation (PCE), the QRISK3 cardiovascular risk calculator, the Reynolds Risk Score, and Systematic Coronary Risk Evaluation-2 (SCORE2). Methods: A cross-sectional study was conducted on 11,174 40-65-year-old individuals with diagnosed metabolic syndrome from a single tertiary university hospital in Lithuania. Cardiovascular risk was calculated using the eight RPMs, and the results were categorized into high, intermediate, and low-risk groups. Inter-model agreement was quantified using Cohen's Kappa coefficients. Results: The study revealed significant heterogeneity in risk categorizations with only 1.49% of cases where all models agree on the risk category. SCORE2 predominantly categorized participants as high-risk (67.39%), while the PCE identified the majority as low-risk (62.03%). Cohen's Kappa coefficients ranged from -0.09 to 0.64, indicating varying degrees of inter-model agreement. Conclusions: The choice of RPM can substantially influence clinical decision-making and patient management. The PCE and AusCVDRisk models exhibited the highest degree of agreement while the SCORE2 model consistently exhibited low agreement with other models.

Comparative Analysis of Three Atherosclerotic Cardiovascular Disease Risk Prediction Models in Individuals Aged 75 and Older

Purpose

To evaluate the performance of the Framingham cardiovascular risk score (FRS)/pooled cohort equations (PCE)/China prediction for atherosclerotic cardiovascular disease (ASCVD) risk (China-PAR model) in a prospective cohort of Chinese older adults.

Patients and Methods

We assessed 717 older adults aged 75-85 years without ASCVD at the baseline from the Sichuan province of China. The participants were followed annually from 2011 to 2021. We obtained the participants' information through the medical records of physical examination and evaluated their 10-year ASCVD risk using FRS, PCE, and China-PAR. We further evaluated the predictive abilities of three assessment models.

Results

During the 10-year follow-up, 206 participants developed ASCVD, with an incidence rate of 28.73%. The FRS and China-PAR moderately underestimated the risk of ASCVD (22.1% and 12.4%, respectively), but while PCE overestimated the risk (36.1%). FRS and China-PAR were found to underestimate the risk of ASCVD (26% and 63%, respectively) for men, while PCE overestimated the risk by 8%; For women, FRS and China-PAR were found to underestimate the risk of ASCVD (14% and 35%, respectively), while PCE overestimated the risk by 88%.

Conclusion

The 10-year ASCVD risk was found to be overestimated by PCE. China-PAR had the most accurate predictions in women, while FRS was particularly well-calibrated in males. All three risk models have good discrimination, with FRS and PCE being well-calibrated in men and all three being well-calibrated in women. Therefore, accurate risk models are warranted to facilitate the prevention of ASCVD at the baseline among Chinese older adults.

Comparison of interleukin-6 and high-sensitivity C-reactive protein for cardiovascular risk assessment: Findings from the MESA study

BACKGROUND AND AIMS

Inflammation is a risk factor for major adverse cardiovascular events (MACE). Elevated levels of both high-sensitivity C-reactive protein (hsCRP) and interleukin-6 (IL6) have been associated with MACE. However, few studies have compared IL6 to hsCRP for cardiovascular risk assessment. Using the MESA (Multi-Ethnic Study of Atherosclerosis) study cohort, we aim to compare IL6 to hsCRP.

METHODS

We divided IL6 and hsCRP by their median values and created 4 groups i.e., low-low, high-low, low-high and high-high. The median follow-up was 14 years.

RESULTS

6614 (97 %) participants had complete baseline IL6 and hsCRP data. The correlation between hsCRP and IL6 was modest (Rho = 0.53). IL6 ≥1.2 pg/mL (median) was present in 3309 participants, and hsCRP ≥1.9 mg/L (median) was present in 3339 participants. Compared to participants with low IL6 and low hsCRP, those with high IL6 and high hsCRP were older (64 vs. 60 years), more frequently women (63 % vs. 45 %), and with more cardiovascular co-morbidities. hsCRP outcome associations lost statistical significance when adjusting for IL6: MACE HR (95 %CI) 1.06 (0.93-1.20), p =0.39, whereas IL6 associations remained significant after adjusting for hsCRP: HR (95 %CI) 1.44 (1.25-1.64), p <0.001. The C-index of Framingham score for did not improve with hsCRP but improved with IL6. Compared to participants with low IL6 and low hsCRP, those with high IL6, regardless of hsCRP, experienced an increased risk of MACE, heart failure and mortality.

CONCLUSIONS

In a diverse and asymptomatic population, IL6 showed a stronger association with atherosclerotic, heart failure and fatal outcomes than hsCRP.

Prediction of Cardiovascular Complication in Patients with Newly Diagnosed Type 2 Diabetes Using an XGBoost/GRU-ODE-Bayes-Based Machine-Learning Algorithm

BACKGRUOUND

Cardiovascular disease is life-threatening yet preventable for patients with type 2 diabetes mellitus (T2DM). Because each patient with T2DM has a different risk of developing cardiovascular complications, the accurate stratification of cardiovascular risk is critical. In this study, we proposed cardiovascular risk engines based on machine-learning algorithms for newly diagnosed T2DM patients in Korea.

METHODS

To develop the machine-learning-based cardiovascular disease engines, we retrospectively analyzed 26,166 newly diagnosed T2DM patients who visited Seoul St. Mary's Hospital between July 2009 and April 2019. To accurately measure diabetes-related cardiovascular events, we designed a buffer (1 year), an observation (1 year), and an outcome period (5 years). The entire dataset was split into training and testing sets in an 8:2 ratio, and this procedure was repeated 100 times. The area under the receiver operating characteristic curve (AUROC) was calculated by 10-fold cross-validation on the training dataset.

RESULTS

The machine-learning-based risk engines (AUROC XGBoost=0.781±0.014 and AUROC gated recurrent unit [GRU]-ordinary differential equation [ODE]-Bayes=0.812±0.016) outperformed the conventional regression-based model (AUROC=0.723± 0.036).

CONCLUSION

GRU-ODE-Bayes-based cardiovascular risk engine is highly accurate, easily applicable, and can provide valuable information for the individualized treatment of Korean patients with newly diagnosed T2DM.

Recommendations for statin management in primary prevention: disparities among international risk scores

BACKGROUND AND AIMS

Statin recommendations in primary prevention depend upon risk algorithms. Moreover, with intermediate risk, risk enhancers and de-enhancers are advocated to aid decisions. The aim of this study was to compare algorithms used in North America and Europe for the identification of patients warranting statin or consideration of risk enhancers and de-enhancers.

METHODS

A simulated population (n = 7680) equal in males and females, with/without smoking, aged 45-70 years, total cholesterol 3.5-7.0 mmol/L, high-density lipoprotein cholesterol 0.6-2.2 mmol/L, and systolic blood pressure 100-170 mmHg, was evaluated. High, intermediate, and low risks were determined using the Framingham Risk Score (FRS), Pooled Cohort Equation (PCE), four versions of Systematic Coronary Risk Evaluation 2 (SCORE2), and Multi-Ethnic Study of Atherosclerosis (MESA) algorithm (0-1000 Agatston Units).

RESULTS

Concordance for the three levels of risk varied from 19% to 85%. Both sexes might be considered to have low, intermediate, or high risk depending on the algorithm applied, even with the same burden of risk factors. Only SCORE2 (High Risk and Very High Risk versions) identified equal proportions of males and females with high risk. Excluding MESA, the proportion with moderate risk was 25% (SCORE2, Very High Risk Region), 32% (FRS), 39% (PCE), and 45% (SCORE2, Low Risk Region).

CONCLUSION

Risk algorithms differ substantially in their estimation of risk, recommendations for statin treatment, and use of ancillary testing, even in identical patients. These results highlight the limitations of currently used risk-based approaches for addressing lipid-specific risk in primary prevention.

High-sensitivity cardiac troponin I for cardiovascular risk stratification in apparently healthy individuals

Troponin I and T as cardiac-specific biomarkers are highly useful tools not only in the diagnosis of acute coronary syndromes but also as independent predictors of several other clinical conditions. High-sensitivity cardiac troponin (hs-cTn) assays allow the detection of considerably low concentrations of cardiac troponin in apparently healthy and asymptomatic individuals, being a candidate tool for cardiovascular risk stratification in the general population. A group of Greek experts summarized the bulk of evidence regarding the use of hs-cTnI as a predictor of cardiovascular events and mortality in apparently healthy individuals and its additive value on top of existing risk stratification methods. This document could serve as a guide for the incorporation of hs-cTnI as an additional risk stratification tool in cardiovascular prevention strategies in apparently healthy individuals.

A Practical Nomogram for Predicting the Bleeding Risk in Patients with a History of Myocardial Infarction Treating with Aspirin

BACKGROUND

Aspirin is a widely used antiplatelet medication to prevent blood clots, reducing the risk of cardiovascular event. Healthcare providers need to be mindful of the risk of aspirin-induced bleeding and carefully balancing its benefits against potential risks. The objective of this study was to create a practical nomogram for predicting bleeding risk in patients with a history of myocardial infarction treating with aspirin.

METHODS

A total of 2099 myocardial infarction patients with aspirin were enrolled. The patients were randomly divided into two groups, with a 7:3 ratio, for model development and internal validation. Boruta analysis was utilized to identify clinically significant features associated with bleeding. Logistic regression model based on independent bleeding risk factors was constructed and presented as a nomogram. Model performance was assessed from three aspects: identification, calibration, and clinical utility.

RESULTS

Boruta analysis identified eight clinical features from 25, and further multivariate logistic regression analysis selected four independent risk factors: hemoglobin, platelet count, previous bleeding, and sex. A visual nomogram was created based on these variables. The model achieved an area under the curve of 0.888 (95% CI: 0.845-0.931) in the training dataset and 0.888 (95% CI: 0.808-0.968) in the test dataset. Calibration curve analysis showed close approximation to the ideal curve. Decision curve analysis demonstrated favorable clinical net benefit for the model.

CONCLUSIONS

Our study focused on creating and validating a model to evaluate bleeding risk in patients with a history of myocardial infarction treated with aspirin, which demonstrated outstanding performance in discrimination, calibration, and net clinical benefit.

Machine learning for the prediction of atherosclerotic cardiovascular disease during 3-year follow up in Chinese type 2 diabetes mellitus patients

AIMS/INTRODUCTION

Clinical guidelines for the management of individuals with type 2 diabetes mellitus endorse the systematic assessment of atherosclerotic cardiovascular disease risk for early interventions. In this study, we aimed to develop machine learning models to predict 3-year atherosclerotic cardiovascular disease risk in Chinese type 2 diabetes mellitus patients.

MATERIALS AND METHODS

Clinical records of 4,722 individuals with type 2 diabetes mellitus admitted to 94 hospitals were used. The features included demographic information, disease histories, laboratory tests and physical examinations. Logistic regression, support vector machine, gradient boosting decision tree, random forest and adaptive boosting were applied for model construction. The performance of these models was evaluated using the area under the receiver operating characteristic curve. Additionally, we applied SHapley Additive exPlanation values to explain the prediction model.

RESULTS

All five models achieved good performance in both internal and external test sets (area under the receiver operating characteristic curve >0.8). Random forest showed the highest discrimination ability, with sensitivity and specificity being 0.838 and 0.814, respectively. The SHapley Additive exPlanation analyses showed that previous history of diabetic peripheral vascular disease, older populations and longer diabetes duration were the three most influential predictors.

CONCLUSIONS

The prediction models offer opportunities to personalize treatment and maximize the benefits of these medical interventions.

Performance of the ACC/AHA Pooled Cohort Cardiovascular Risk Equations in Clinical Practice

BACKGROUND

The performance of the American College of Cardiology/American Heart Association pooled cohort equation (PCE) for atherosclerotic cardiovascular disease (ASCVD) in real-world clinical practice has not been evaluated extensively.

OBJECTIVES

The goal of this study was to test the performance of PCE to predict ASCVD risk in the community, and determine if including individuals with values outside the PCE range (ie, age, blood pressure, cholesterol) or statin therapy initiation over follow-up would significantly affect PCE predictive capabilities.

METHODS

The PCE was validated in a community-based cohort of consecutive patients who sought primary care in Olmsted County, Minnesota, between 1997 and 2000, followed-up through 2016. Inclusion criteria were similar to those of PCE derivation. Patient information was ascertained by using the record linkage system of the Rochester Epidemiology Project. ASCVD events (nonfatal and fatal myocardial infarction and ischemic stroke) were validated in duplicate. Calculated and observed ASCVD risk and c-statistics were compared across predefined groups.

RESULTS

This study included 30,042 adults, with a mean age of 48.5 ± 12.2 years; 46% were male. Median follow-up was 16.5 years, truncated at 10 years for this analysis. Mean ASCVD risk was 5.6% ± 8.73%. There were 1,555 ASCVD events (5.2%). The PCE revealed good performance overall (c-statistic 0.78) and in sex and race subgroups; it was highest among non-White female subjects (c-statistic 0.81) and lowest in White male subjects (c-statistic 0.77). Out-of-range values and initiation of statin medication did not affect model performance.

CONCLUSIONS

The PCE performed well in a community cohort representing real-world clinical practice. Values outside PCE ranges and initiation of statin medication did not affect performance. These results have implications for the applicability of current strategies for the prevention of ASCVD.

Prognostic Value of Ankle-Brachial Index in Prediction of Cardiovascular Events in an Asian Population with Multiple Atherosclerotic Risk Factors

We aimed to evaluate the incremental prognostic value after incorporation of the ankle-brachial index (ABI) into the 10-year pool cohort equation (PCE) risk model in patients with multiple risk factors (MRFs). A total of 4332 MRFs patients were divided into 2 groups as ABI ≤.9 or >.9. The primary outcome was hard cardiovascular events (hCVE: including cardiovascular death, myocardial infarction, or ischemic stroke) over a median follow-up of 36 months. The Cox proportional hazards survival model, C-statistic, and net reclassification indices (NRI) were used. The occurrence of the primary outcome in the ABI ≤.9 group (3.7%) was significantly greater than in the ABI > .9 group (1.3%), P < .001. ABI is an independent predictor of hCVE in addition to the variables in the standard risk model (age, gender, and smoking status). ABI modestly improved the C-index when added to the PCE risk model (PCE .70 vs ABI+PCE .74). The addition of ABI to the PCE risk model did not significantly improve the classification of patients (NRI -.029; 95% CI: -.215 to .130). Despite ABI being one of the independent predictors of hCVE, integration of ABI into the PCE model did not improve the efficacy of risk reclassification in patients with MRFs.

Evidence supporting the choice of a new cardiovascular risk equation for Australia

This article reviews the risk equations recommended for use in international cardiovascular disease (CVD) primary prevention guidelines and assesses their suitability for use in Australia against a set of a priori defined selection criteria. The review and assessment were commissioned by the National Heart Foundation of Australia on behalf of the Australian Chronic Disease Prevention Alliance to inform recommendations on CVD risk estimation as part of the 2023 update of the Australian CVD risk assessment and management guidelines. Selected international risk equations were assessed against eight selection criteria: development using contemporary data; inclusion of established cardiovascular risk factors; inclusion of ethnicity and deprivation measures; prediction of a broad selection of fatal and non-fatal CVD outcomes; population representativeness; model performance; external validation in an Australian dataset; and the ability to be recalibrated or modified. Of the ten risk prediction equations reviewed, the New Zealand PREDICT equation met seven of the eight selection criteria, and met additional usability criteria aimed at assessing the ability to apply the risk equation in practice in Australia.

Development of a Cardiovascular Disease Risk Prediction Model: A Preliminary Retrospective Cohort Study of a Patient Sample in Saudi Arabia

Saudi Arabia has an alarmingly high incidence of cardiovascular disease (CVD) and its associated risk factors. To effectively assess CVD risk, it is essential to develop tailored models for diverse regions and ethnicities using local population variables. No CVD risk prediction model has been locally developed. This study aims to develop the first 10-year CVD risk prediction model for Saudi adults aged 18 to 75 years. The electronic health records of Saudi male and female patients aged 18 to 75 years, who were seen in primary care settings between 2002 and 2019, were reviewed retrospectively via the Integrated Clinical Information System (ICIS) database (from January 2002 to February 2019). The Cox regression model was used to identify the risk factors and develop the CVD risk prediction model. Overall, 451 patients were included in this study, with a mean follow-up of 12.05 years. Thirty-five (7.7%) patients developed a CVD event. The following risk factors were included: fasting blood sugar (FBS) and high-density lipoprotein cholesterol (HDL-c), heart failure, antihyperlipidemic therapy, antithrombotic therapy, and antihypertension therapy. The Bayesian information criterion (BIC) score was 314.4. This is the first prediction model developed in Saudi Arabia and the second in any Arab country after the Omani study. We assume that our CVD predication model will have the potential to be used widely after the validation study.

Serum Soluble Lectin-like Oxidized Low-Density Lipoprotein Receptor-1 (sLOX-1) Is Associated with Atherosclerosis Severity in Coronary Artery Disease

Risk-factor-based scoring systems for atherosclerotic coronary artery disease (CAD) remain concerningly inaccurate at the level of the individual and would benefit from the addition of biomarkers that correlate with atherosclerosis burden directly. We hypothesized that serum soluble lectin-like oxidized low-density lipoprotein receptor-1 (sLOX-1) would be independently associated with CAD and investigated this in the BioHEART study using 968 participants with CT coronary angiograms, which were scored for disease burden in the form of coronary artery calcium scores (CACS), Gensini scores, and a semi-quantitative soft-plaque score (SPS). Serum sLOX-1 was assessed by ELISA and was incorporated into regression models for disease severity and incidence. We demonstrate that sLOX-1 is associated with an improvement in the prediction of CAD severity when scored by Gensini or SPS, but not CACS. sLOX-1 also significantly improved the prediction of the incidence of obstructive CAD, defined as stenosis in any vessel >75%. The predictive value of sLOX-1 was significantly greater in the subgroup of patients who did not have any of the standard modifiable cardiovascular risk factors (SMuRFs). sLOX-1 is associated with CAD severity and is the first biomarker shown to have utility for risk prediction in the SMuRFless population.

Incident CHD and ischemic stroke associated with lipoprotein(a) by levels of Factor VIII and inflammation

BACKGROUND

Inflammation and coagulation may contribute to the increased risk for atherosclerotic cardiovascular disease (ASCVD) associated with high lipoprotein(a). The association of lipoprotein(a) with ASCVD is stronger in individuals with high versus low high-sensitivity C-reactive protein (hs-CRP), a marker of inflammation.

OBJECTIVES

Determine the association of lipoprotein(a) with incident ASCVD by levels of coagulation Factor VIII controlling for hs-CRP.

METHODS

We analyzed data from 6,495 men and women 45 to 84 years of age in the Multi-Ethnic Study of Atherosclerosis (MESA) without prevalent ASCVD at baseline (2000-2002). Lipoprotein(a) mass concentration, Factor VIII coagulant activity, and hs-CRP were measured at baseline and categorized as high or low (≥75th or <75th percentile of the distribution). Participants were followed for incident coronary heart disease (CHD) and ischemic stroke through 2015.

RESULTS

Over a median follow-up of 13.9 years, there were 390 CHD and 247 ischemic stroke events. The hazard ratio (95%CI) for CHD associated with high lipoprotein(a) (≥40.1 versus <40.1 mg/dL) including adjustment for hs-CRP among participants with low and high Factor VIII was 1.07 (0.80-1.44) and 2.00 (1.33-3.01), respectively (p-value for interaction 0.016). The hazard ratio (95%CI) for CHD associated with high lipoprotein(a) including adjustment for Factor VIII was 1.16 (0.87-1.54) and 2.00 (1.29-3.09) among participants with low and high hs-CRP, respectively (p-value for interaction 0.042). Lp(a) was not associated with ischemic stroke regardless of Factor VIII or hs-CRP levels.

CONCLUSION

High lipoprotein(a) is a risk factor for CHD in adults with high levels of hemostatic or inflammatory markers.

A multi-ancestry polygenic risk score improves risk prediction for coronary artery disease

Identification of individuals at highest risk of coronary artery disease (CAD)-ideally before onset-remains an important public health need. Prior studies have developed genome-wide polygenic scores to enable risk stratification, reflecting the substantial inherited component to CAD risk. Here we develop a new and significantly improved polygenic score for CAD, termed GPSMult, that incorporates genome-wide association data across five ancestries for CAD (>269,000 cases and >1,178,000 controls) and ten CAD risk factors. GPSMult strongly associated with prevalent CAD (odds ratio per standard deviation 2.14, 95% confidence interval 2.10-2.19, P < 0.001) in UK Biobank participants of European ancestry, identifying 20.0% of the population with 3-fold increased risk and conversely 13.9% with 3-fold decreased risk as compared with those in the middle quintile. GPSMult was also associated with incident CAD events (hazard ratio per standard deviation 1.73, 95% confidence interval 1.70-1.76, P < 0.001), identifying 3% of healthy individuals with risk of future CAD events equivalent to those with existing disease and significantly improving risk discrimination and reclassification. Across multiethnic, external validation datasets inclusive of 33,096, 124,467, 16,433 and 16,874 participants of African, European, Hispanic and South Asian ancestry, respectively, GPSMult demonstrated increased strength of associations across all ancestries and outperformed all available previously published CAD polygenic scores. These data contribute a new GPSMult for CAD to the field and provide a generalizable framework for how large-scale integration of genetic association data for CAD and related traits from diverse populations can meaningfully improve polygenic risk prediction.

Novel Biomarkers and the Multiple-Marker Approach in Early Detection, Prognosis, and Risk Stratification of Cardiac Diseases: A Narrative Review

Cardiac diseases are a primary cause of mortality worldwide, underscoring the importance of early identification and risk stratification to enhance patient outcomes. Biomarkers have become important tools for the risk assessment of cardiovascular disease and monitoring disease progression. This narrative review focuses on the multiple-marker approach, which involves simultaneously evaluating several biomarkers for the early detection and risk stratification of heart diseases. The review covers the clinical applications of novel biomarkers, such as high-sensitivity troponin, galectin-3, source of tumorigenicity 2, B-type natriuretic peptide and N-terminal pro-B-type natriuretic peptide, growth differentiation factor 15, myeloperoxidase, fatty acid-binding protein, C-reactive protein, lipoprotein-associated phospholipase A2, microRNAs, circulating endothelial cells, and ischemia-modified albumin. These biomarkers have demonstrated potential in identifying people who are at high risk for developing heart disease and in providing prognostic data. Given the complexity of cardiac illnesses, the multiple-marker approach to risk assessment is extremely beneficial. Implementing the multiple-marker strategy can improve risk stratification, diagnostic accuracy, and patient care in heart disease patients.

Artificial Intelligence in Cardiology: Applications and Obstacles

Artificial intelligence (AI) technology is poised to alter the flow of daily life, and in particular, medicine, where it may eventually complement the physician's work in diagnosing and treating disease. Despite the recent frenzy and uptick in AI research over the past decade, the integration of AI into medical practice is in its early stages. Cardiology stands to benefit due to its many diagnostic modalities and diverse treatments. AI methods have been applied to various domains within cardiology: imaging, electrocardiography, wearable devices, risk prediction, and disease classification. While many AI-based approaches have been developed that perform equal to or better than the state-of-the-art, few prospective randomized studies have evaluated their use. Furthermore, obstacles at the intersection of medicine and AI remain unsolved, including model understanding, bias, model evaluation, relevance and reproducibility, and legal and ethical dilemmas. We summarize recent and current applications of AI in cardiology, followed by a discussion of the aforementioned complications.

Machine Learning and the Conundrum of Stroke Risk Prediction

Stroke is a leading cause of death worldwide. With escalating healthcare costs, early non-invasive stroke risk stratification is vital. The current paradigm of stroke risk assessment and mitigation is focused on clinical risk factors and comorbidities. Standard algorithms predict risk using regression-based statistical associations, which, while useful and easy to use, have moderate predictive accuracy. This review summarises recent efforts to deploy machine learning (ML) to predict stroke risk and enrich the understanding of the mechanisms underlying stroke. The surveyed body of literature includes studies comparing ML algorithms with conventional statistical models for predicting cardiovascular disease and, in particular, different stroke subtypes. Another avenue of research explored is ML as a means of enriching multiscale computational modelling, which holds great promise for revealing thrombogenesis mechanisms. Overall, ML offers a new approach to stroke risk stratification that accounts for subtle physiologic variants between patients, potentially leading to more reliable and personalised predictions than standard regression-based statistical associations.

Performance of the Pooled Cohort Equations in non-alcoholic fatty liver disease: The Multi-Ethnic Study of Atherosclerosis

BACKGROUND AND AIMS

Non-alcoholic fatty liver disease (NAFLD) is associated with a high risk of cardiovascular disease. Whether risk scores developed in the general population accurately assess cardiovascular risk in the NAFLD population is unknown. This study aimed to evaluate the performance of the Pooled Cohort Equations (PCE) in NAFLD.

METHODS

Individuals in the Multi-Ethnic Study of Atherosclerosis with baseline non-contrast cardiac computed tomography scans with sufficient data to determine the presence of hepatic steatosis were identified and assessed for the development of incident 10-year atherosclerotic cardiovascular disease. The discrimination and calibration of the PCE were evaluated, and the observed and expected events by risk category (<5%, 5-<7.5%, 7.5-<20%, ≥20%) were determined. Risk reclassification with the addition of NAFLD to the PCE was assessed.

RESULTS

Of 4014 participants included, 698 (17.4%) with NAFLD were identified, including 247 (35.3%) with moderate-to-severe steatosis. Discrimination of the PCE was suboptimal in NAFLD (c-statistic 0.69), particularly moderate-to-severe steatosis (0.65), and calibration was overall poor. While risk was overestimated in non-NAFLD, it was underestimated in NAFLD in lower/intermediate risk categories, predominantly in women (5-<7.5% observed/expected ratio = 1.67). The addition of NAFLD to the PCE improved risk classification in women.

CONCLUSIONS

The PCE overall performed suboptimally in cardiovascular risk assessment in NAFLD, particularly in women and individuals with moderate-to-severe steatosis in clinically relevant risk categories. Primary prevention may need to be considered at a lower risk threshold in these groups, and further work is needed to improve risk stratification in this growing high-risk population.

The Role of Imaging in Preventive Cardiology in Women

PURPOSE OF REVIEW

The prevalence of CVD in women is increasing and is due to the increased prevalence of CV risk factors. Traditional CV risk assessment tools for prevention have failed to accurately determine CVD risk in women. CAC has shown to more precisely determine CV risk and is a better predictor of CV outcomes. Coronary CTA provides an opportunity to determine the presence of CAD and initiate prevention in women presenting with angina. Identifying women with INOCA due to CMD with use of cPET or cMRI with MBFR is vital in managing these patients. This review article outlines the role of imaging in preventive cardiology for women and will include the latest evidence supporting the use of these imaging tests for this purpose.

RECENT FINDINGS

CV mortality is higher in women who have more extensive CAC burden. Women have a greater prevalence of INOCA which is associated with higher MACE. INOCA is due to CMD in most cases which is associated with traditional CVD risk factors. Over half of these women are untreated or undertreated. Recent study showed that stratified medical therapy, tailored to the specific INOCA endotype, is feasible and improves angina in women. Coronary CTA is useful in the setting of women presenting with acute chest pain to identify CAD and initiate preventive therapy. CAC confers greater relative risk for CV mortality in women versus (vs.) men. cMRI or cPET is useful to assess MBFR to diagnose CMD and is another useful imaging tool in women for CV prevention.

German Radiological Society and the Professional Association of German Radiologists Position Paper on Coronary computed tomography: Clinical Evidence and Quality of Patient Care in Chronic Coronary Syndrome

This position paper is a joint statement of the German Radiological Society (DRG) and the Professional Association of German Radiologists (BDR), which reflects the current state of knowledge about coronary computed tomography. It is based on preclinical and clinical studies that have investigated the clinical relevance as well as the technical requirements and fundamentals of cardiac computed tomography. CITATION FORMAT: · Langenbach MC, Sandstede J, Sieren M et al. DRG and BDR Position Paper on Coronary CT: Clinical Evidence and Quality of Patient Care in Chronic Coronary Syndrome. Fortschr Röntgenstr 2023; 195: 115 - 133.

[German Radiological Society and the Professional Association of German Radiologists position paper on coronary computed tomography: clinical evidence and quality of patient care in chronic coronary syndrome]

This position paper is a joint statement of the German Radiological Society (DRG) and the Professional Association of German Radiologists (BDR), which reflects the current state of knowledge about coronary computed tomography (CT). It is based on preclinical and clinical studies that have investigated the clinical relevance as well as the technical requirements and fundamentals of cardiac computed tomography.

High Concordance between D:A:Dr and the Framingham Risk Score in Brazilians Living with HIV

People living with HIV (PLHIV) have twice the risk of developing cardiovascular diseases, making it essential to identify high cardiovascular risk (CVR). However, there is no validated CVR calculator for PLHIV in Brazil. We performed a cross-sectional study with 265 individuals living with HIV, aged 40 to 74 years, to assess the agreement between three CVR scores: Framingham Risk Score (FRS), Atherosclerotic Cardiovascular Disease (ASCVD) Risk Score, and a specific for PLHIV, Reduced Data Collection on Adverse Effects of Anti-HIV Drugs (D:A:Dr). We assessed agreement using the weighted Kappa coefficient and the Bland-Altman plot. The median age was 52 years (47-58), 58.9% were men, 34% were hypertensive and 8.3% had a detectable viral load. There was an almost perfect agreement between D:A:Dr x FRS (k = 0.82; 95% CI 0.77-0.87; p < 0.001), and substantial agreement between FRS vs. ASCVD (k = 0.74; 95% CI 0.69-0.79; p < 0.001) and between D:A:Dr vs. ASCVD (k = 0.70; 95% CI 0.64-0.76; p < 0.001). The Bland-Altman plot revealed greater discordance between scores as the CVR increased. Our results suggest that the FRS and the D:A:Dr are adequate to classify the CVR in this population, and the D:A:Dr score can be used as an alternative to the FRS in Brazil, as other international guidelines have already advocated.

Comparing the pooled cohort equations and coronary artery calcium scores in a symptomatic mixed Asian cohort

Background

The value of pooled cohort equations (PCE) as a predictor of major adverse cardiovascular events (MACE) is poorly established among symptomatic patients. Coronary artery calcium (CAC) assessment further improves risk prediction, but non-Western studies are lacking. This study aims to compare PCE and CAC scores within a symptomatic mixed Asian cohort, and to evaluate the incremental value of CAC in predicting MACE, as well as in subgroups based on statin use.

Methods

Consecutive patients with stable chest pain who underwent cardiac computed tomography were recruited. Logistic regression was performed to determine the association between risk factors and MACE. Cohort and statin-use subgroup comparisons were done for PCE against Agatston score in predicting MACE.

Results

Of 501 patients included, mean (SD) age was 53.7 (10.8) years, mean follow-up period was 4.64 (0.66) years, 43.5% were female, 48.3% used statins, and 50.0% had no CAC. MI occurred in 8 subjects while 9 subjects underwent revascularization. In the general cohort, age, presence of CAC, and ln(Volume) (OR = 1.05, 7.95, and 1.44, respectively) as well as age and PCE score for the CAC = 0 subgroup (OR = 1.16 and 2.24, respectively), were significantly associated with MACE. None of the risk factors were significantly associated with MACE in the CAC > 0 subgroup. Overall, the PCE, Agatston, and their combination obtained an area under the receiver operating characteristic curve (AUC) of 0.501, 0.662, and 0.661, respectively. Separately, the AUC of PCE, Agatston, and their combination for statin non-users were 0.679, 0.753, and 0.734, while that for statin-users were 0.585, 0.615, and 0.631, respectively. Only the performance of PCE alone was statistically significant (p = 0.025) when compared between statin-users (0.507) and non-users (0.783).

Conclusion

In a symptomatic mixed Asian cohort, age, presence of CAC, and ln(Volume) were independently associated with MACE for the overall subgroup, age and PCE score for the CAC = 0 subgroup, and no risk factor for the CAC > 0 subgroup. Whilst the PCE performance deteriorated in statin versus non-statin users, the Agatston score performed consistently in both groups.

Magnitude, Distribution and Contextual Risk Enhancing Predictors of High 10-Year Cardiovascular Risk Among Diabetic Patients in Tanzania

Introduction

Atherosclerotic Cardiovascular Disease (ASCVD) is the leading cause of death worldwide. In Diabetics, ASCVD is associated with poor prognosis and a higher case fatality rate compared with the general population. Sub-Saharan Africa is facing an epidemiological transition with ASCVD being prevalent among young adults. To date, over 20 million people have been living with DM in Africa, Tanzania being one of the five countries in the continent reported to have a higher prevalence. This study aimed to identify an individual's 10-year ASCVD absolute risk among a diabetic cohort in Tanzania and define contextual risk enhancing factors.

Methods

A prospective observational study was conducted at the Aga Khan hospital, Mwanza, for a period of 8 months. The hospital is a 42-bed district-level hospital in Tanzania. Individuals 10-year risk was calculated based on the ASCVD 2013 risk calculator by ACC/AHA. Pearson's chi-square or Fischer's exact test was used to compare categorical and continuous variables. Multivariable analysis was applied to determine contextual factors for those who had a high 10-year risk of developing ASCVD.

Results

The overall cohort included 573 patients. Majority of the individuals were found to be hypertensive (n = 371, 64.7%) and obese (n = 331, 58%) having a high 10-year absolute risk (n = 343, 60%) of suffering ASCVD. The study identified duration of Diabetes Mellitus (>10 years) (OR 8.15, 95% CI 5.25-14.42), concomitant hypertension (OR 1.82 95% CI 1.06-3.06), Diabetic Dyslipidemia (OR 1.44, 95% CI 1.08-1.92) and deranged serum creatinine (OR 1.03, 95% CI 1.02-1.03) to be the risk enhancing factors amongst our population.

Conclusion

The study confirms the majority of diabetic individuals in the lake region of Tanzania to have a high 10-year ASCVD risk. The high prevalence of obesity, hypertension and dyslipidemia augments ASCVD risk but provides interventional targets for health-care workers to decrease these alarming projections.

Recalibration of the Framingham risk score for predicting 10-year risk of cardiovascular events: A non-concurrent rural cohort study from Tamil Nadu

OBJECTIVE

To recalibrate the Framingham Risk Score-cardiovascular diseases (FRS-CVD) using 10-year mortality data and baseline risk factor data for a rural cohort and assess the effect of recalibration on proportion categorised as high risk.

METHODS

Participants of a cardiovascular risk factor survey aged 30-64 years in 2011-12, from 9 villages of a rural block in Vellore, Tamil Nadu, were followed up for mortality till 2021, as part of an established demographic surveillance system. We calculated both lipid-based and Body Mass Index-based FRS-CVD risk scores, as well as recalibrated scores using risk factor data from the baseline survey and CVD mortality observed over 10 years.

RESULTS

Based on original lipid-based FRS-CVD scores, 8.48% (109) of 1285 males had a 10-year CVD risk ≥30%, compared to 11.60% (149) with recalibrated scores. Among 1737 females, 1.50% (26) had a 10-year CVD risk of ≥30%, using original FRS-CVD scores, and 3.22% (56) using recalibrated scores. Similarly, for BMI based FRS-CVD scores, overall, 3.63% (110/3028) had a 10-year risk of ≥30%, compared to 6.64% (201) using recalibrated scores. The median 10-year FRS-CVD original score in males was 7.57 (IQR: 3.67-15.83), and 2.53 (IQR: 1.28-5.32) in females, compared to 8.95 (IQR: 4.35-18.52) and 3.79 (IQR: 1.92-7.93) respectively, for the recalibrated FRS-CVD risk scores.

CONCLUSION

The recalibrated Framingham models showed a greater proportion of the population at risk of CVDs compared to the original FRS scores, with males having 2-3 times greater CVD risk scores compared to females.

Predictive value of CAC score combined with clinical features for obstructive coronary heart disease on coronary computed tomography angiography: a machine learning method

OBJECTIVE

We investigated the predictive value of clinical factors combined with coronary artery calcium (CAC) score based on a machine learning method for obstructive coronary heart disease (CAD) on coronary computed tomography angiography (CCTA) in individuals with atypical chest pain.

METHODS

The study included data from 1,906 individuals undergoing CCTA and CAC scanning because of atypical chest pain and without evidence for the previous CAD. A total of 63 variables including traditional cardiovascular risk factors, CAC score, laboratory results, and imaging parameters were used to build the Random forests (RF) model. Among all the participants, 70% were randomly selected to train the models on which fivefold cross-validation was done and the remaining 30% were regarded as a validation set. The prediction performance of the RF model was compared with two traditional logistic regression (LR) models.

RESULTS

The incidence of obstructive CAD was 16.4%. The area under the receiver operator characteristic (ROC) for obstructive CAD of the RF model was 0.841 (95% CI 0.820-0.860), the CACS model was 0.746 (95% CI 0.722-0.769), and the clinical model was 0.810 (95% CI 0.788-0.831). The RF model was significantly superior to the other two models (p < 0.05). Furthermore, the calibration curve and Hosmer-Lemeshow test showed that the RF model had good classification performance (p = 0.556). CAC score, age, glucose, homocysteine, and neutrophil were the top five important variables in the RF model.

CONCLUSION

RF model was superior to the traditional models in the prediction of obstructive CAD. In clinical practice, the RF model may improve risk stratification and optimize individual management.

Performance of the pooled cohort equation in South Asians: insights from a large integrated healthcare delivery system

South Asian ethnicity is associated with increased atherosclerotic cardiovascular disease (ASCVD) risk and has been identified as a "risk enhancer" in the 2018 American College of Cardiology/American Heart Association Guidelines. Risk estimation and statin eligibility in South Asians is not well understood; we studied the accuracy of 10-years ASCVD risk prediction by the pooled cohort equation (PCE), based on statin use, in a South Asian cohort. This is a retrospective cohort study of Kaiser Permanente Northern California South Asian members without existing ASCVD, age range 30-70, and 10-years follow up. ASCVD events were defined as myocardial infarction, ischemic stroke, and cardiovascular death. The cohort was stratified by statin use during the study period: never; at baseline and during follow-up; and only during follow-up. Predicted probability of ASCVD, using the PCE was calculated and compared to observed ASCVD events for low < 5.0%, borderline 5.0 to < 7.5%, intermediate 7.5 to < 20.0%, and high ≥ 20.0% risk groups. A total of 1835 South Asian members were included: 773 never on statin, 374 on statins at baseline and follow-up, and 688 on statins during follow-up only. ASCVD risk was underestimated by the PCE in low-risk groups: entire cohort: 1.8 versus 4.9%, p < 0.0001; on statin at baseline and follow-up: 2.58 versus 8.43%, p < 0.0001; on statin during follow-up only: 2.18 versus 7.77%, p < 0.0001; and never on statin: 1.37 versus 2.09%, p = 0.12. In this South Asian cohort, the PCE underestimated risk in South Asians, regardless of statin use, in the low risk ASCVD risk category.

Estimated versus observed 10-year atherosclerotic cardiovascular event rates in a rural population-based health initiative: The Heart of New Ulm Project

Objective

Assess discrepancy between estimated 10-year atherosclerotic cardiovascular disease (ASCVD) risk and observed 10-year event rates in a rural population participating in cardiovascular health initiative.

Methods

The study included a rural sample of individuals participating in the Heart of New Ulm (HONU), a population-based health initiative aimed at reducing ASCVD risk in a rural community. HONU conducted over 100 baseline screening events with 5221 individuals participating in 2009. For this analysis, we included participants who were aged 40-79 years, free of ASCVD at baseline, and had adequate data to calculate 10-year ASCVD risk. Electronic health record data and state death records were used to determine rates of non-fatal myocardial infarction and stroke, and ASCVD death from 2010-2019. ASCVD event rates were compared to estimated 10-year risks calculated using the Pooled Cohort Equations, stratified by sex and clinically relevant risk categories.

Results

The sample (n = 2819, mean ± SD age 56.1 ± 9.9 years, 59.6% female) had a low prevalence of tobacco use (8.1% current smokers) and diabetes (6.5%) and a high prevalence of hypertension (44.4%) and hyperlipidemia (56.6%). The median estimated 10-year ASCVD risk for the entire sample was 5.7% (IQR 2.3-13.5%) with an observed 10-year ASCVD event rate of 3.4%. The largest gap between observed and estimated risk was in those at intermediate/high (≥7.5%) ASCVD risk (median 10-year risk 15.8% [IQR 10.4-29.0], observed ASCVD event rate 6.4%).

Conclusio

In a sample of rural participants exposed to a multifaceted ASCVD prevention initiative, observed rates of ASCVD were substantially lower compared to estimated ASCVD risk. The potential for significantly lower than predicted ASCVD event rates in certain populations should be included in the clinician-patient risk discussion.

Evaluation of available risk scores to predict multiple cardiovascular complications for patients with type 2 diabetes mellitus using electronic health records

Aims

Various cardiovascular risk prediction models have been developed for patients with type 2 diabetes mellitus. Yet few models have been validated externally. We perform a comprehensive validation of existing risk models on a heterogeneous population of patients with type 2 diabetes using secondary analysis of electronic health record data.

Methods

Electronic health records of 47,988 patients with type 2 diabetes between 2013 and 2017 were used to validate 16 cardiovascular risk models, including 5 that had not been compared previously, to estimate the 1-year risk of various cardiovascular outcomes. Discrimination and calibration were assessed by the c-statistic and the Hosmer-Lemeshow goodness-of-fit statistic, respectively. Each model was also evaluated based on the missing measurement rate. Sub-analysis was performed to determine the impact of race on discrimination performance.

Results

There was limited discrimination (c-statistics ranged from 0.51 to 0.67) across the cardiovascular risk models. Discrimination generally improved when the model was tailored towards the individual outcome. After recalibration of the models, the Hosmer-Lemeshow statistic yielded p-values above 0.05. However, several of the models with the best discrimination relied on measurements that were often imputed (up to 39% missing).

Conclusion

No single prediction model achieved the best performance on a full range of cardiovascular endpoints. Moreover, several of the highest-scoring models relied on variables with high missingness frequencies such as HbA1c and cholesterol that necessitated data imputation and may not be as useful in practice. An open-source version of our developed Python package, cvdm, is available for comparisons using other data sources.

Development of Prediction Models for Acute Myocardial Infarction at Prehospital Stage with Machine Learning Based on a Nationwide Database

Models for predicting acute myocardial infarction (AMI) at the prehospital stage were developed and their efficacy compared, based on variables identified from a nationwide systematic emergency medical service (EMS) registry using conventional statistical methods and machine learning algorithms. Patients in the EMS cardiovascular registry aged >15 years who were transferred from the public EMS to emergency departments in Korea from January 2016 to December 2018 were enrolled. Two datasets were constructed according to the hierarchical structure of the registry. A total of 184,577 patients (Dataset 1) were included in the final analysis. Among them, 72,439 patients (Dataset 2) were suspected to have AMI at prehospital stage. Between the models derived using the conventional logistic regression method, the B-type model incorporated AMI-specific variables from the A-type model and exhibited a superior discriminative ability (p = 0.02). The models that used extreme gradient boosting and a multilayer perceptron yielded a higher predictive performance than the conventional logistic regression-based models for analyses that used both datasets. Each machine learning algorithm yielded different classification lists of the 10 most important features. Therefore, prediction models that use nationwide prehospital data and are developed with appropriate structures can improve the identification of patients who require timely AMI management.

Impact of low/no-charge coronary artery calcium scoring on statin eligibility and outcomes in women: The CLARIFY study

Background

Methods

Results

Conclusion

A tool to help patients visualize ASCVD risk and the potential impact of risk-lowering interventions

Risk assessment is a fundamental step in the current approach to primary prevention of atherosclerotic cardiovascular disease (ASCVD). When considering pharmacotherapy for primary prevention of ASCVD, current prevention guidelines in the United States recommend the use of the pooled cohort equations (PCE) to assess 10-year ASCVD risk and begin the important process of shared decision-making between patients and clinicians. Clinicians should support patients in the decisionmaking process by turning raw data into information that is easily understood and more effectively utilized for decisions around the treatment plan. In this work, we present a tool to help patients visualize ASCVD risk and the projected impact of risk-lowering interventions. We believe this visual tool can facilitate communication of ASCVD risk to patients, and improve patient understanding of risk and the potential impact of risklowering interventions, which we believe may help patients make more informed, empowered decisions that achieve greater risk reduction.

Population-Based Recalibration of the Framingham Risk Score and Pooled Cohort Equations

BACKGROUND

The Framingham Risk Score (FRS) and Pooled Cohort Equations (PCEs) overestimate risk in many contemporary cohorts.

OBJECTIVES

This study sought to determine if recalibration of these scores using contemporary population-level data improves risk stratification for statin therapy.

METHODS

Five-year FRS and PCEs were recalibrated using a cohort of Ontario residents alive January 1, 2011, who were 30 to 79 years of age without cardiovascular disease. Scores were externally validated in a primary care cohort of routinely collected electronic medical record data from January 1, 2010, to December 31, 2014. The relative difference in mean predicted and observed risk, number of statins avoided, and number needed to treat with statins to reduce a cardiovascular event at 5 years were reported.

RESULTS

The FRS was recalibrated in 6,938,971 Ontario residents (51.6% women, mean age 48 years) and validated in 71,450 individuals (56.1% women, mean age 52 years). Recalibration reduced overestimation from 109% to 49% for women and 131% to 32% for men. The recalibrated FRS was estimated to reduce statin prescriptions in up to 26 per 1,000 low-risk women and 80 per 1,000 low-risk men, as well as reduce the number needed to treat from 61 to 47 in women and from 53 to 41 in men. In contrast, after recalibration of the PCEs, risk remained overestimated by 217% in women and 128% in men.

CONCLUSIONS

Recalibration is a feasible solution to improve risk prediction but is dependent on the model being used. Recalibration of the FRS but not the PCEs reduced overestimation and may improve utilization of statins.