Pooled Cohort Risk Equations Research Articles

Generic and disease-adapted cardiovascular risk scores as predictors of atherosclerosis progression in SLE

ObjectiveStudies show that generic cardiovascular risk (CVR) prediction tools may underestimate CVR in SLE. We examined, for the first time to our knowledge, whether generic and disease-adapted CVR scores may...

Abstract P337: Determinants and Influence of Breast Arterial Calcifications on Risk of Cardiovascular Disease

Background: Breast arterial calcifications (BAC) are incidentally observed on screening mammography. BAC has been shown to be associated with the presence of coronary artery calcification (CAC) and increased risk of coronary artery disease (CAD). Given that population-based mammography is currently recommended to women, the evaluation of BAC may be important in identifying high-risk women without additional cost or radiation exposure. Aims: We sought to identify reproductive and cardiovascular risk factors associated with the presence of detectable BAC and CAC in women participated to mammography screening. Also, we aimed to determine the association between BAC and presence of CAC. Further, we investigated presence of BAC, CAC and estimated 10-year risk of atherosclerotic cardiovascular disease (ASCVD). Methods: In this cross-sectional study, reproductive history and CVD risk factors were obtained in 215 women ≥ 18 years of age that underwent mammography and cardiac computed tomographic angiography (CCTA) within 2 years of each other between January 2007 and September 2017 at Weill Cornell Medicine/ New York Presbyterian hospital (WCM/NYP). BAC was recorded in two ways: a binary scale (presence/absence) and a semi-quantitative scale (mild, moderate, severe). CAC was calculated using the Agatston method and was recorded as a binary variable (presence/absence). Information regarding reproductive history and CVD risk factors, medical history, and relevant demographics were obtained by chart review. Odds ratios (ORs) and 95% confidence intervals (CIs), with adjustment for age at study entry as a potential confounder, were estimated. The 10-year risk of atherosclerotic cardiovascular disease (ASCVD) was calculated using the Pooled Cohort Risk Equations. Results: The odds of presence of BAC increased with increasing age . Women age ≥60 had a near 6-fold higher odds of BAC (OR = 5.77; 95% CI = 2.45 to 16.00) compared with women <60 years old. Other factors associated with presence of BAC after controlling for age were diastolic blood pressure ≥80 ( P = 0.0008), systolic blood pressure ≥140 ( P = 0.0009), number of children (P = 0.01). Younger age at first birth (≤28 years) was associated with 3-fold higher odds of BAC compared with women with age at first birth >28 years. Except for age at study entry, the only factor associated with presence of CAC was hyperlipidemia ( P = 0.002). We found no association between presence of BAC with CAC. We observed women with presence of both BAC and CAC had the highest estimated 10-year risk of ASCVD: 18.54%, followed by presence of BAC but absence of CAC 11.65%, absence of BAC and presence of CAC 6.01%, and women with no BAC and no CAC presence had a mean 10-year risk of ASCVD of 5.25%. Conclusions: These findings support the value of BAC in identifying women at potentially increased risk of future cardiovascular disease without additional cost and radiation exposure.

Abstract P341: A Flexible, Interpretable Approach to Recalibrate Risk Prediction Models: Adapting the Pooled Cohorts Equation to a Local New England Contemporary Population

Introduction: Atherosclerotic cardiovascular disease (ASCVD) prognostic models have well-documented poor generalizability in contemporary populations. Sustained, accurate performance of risk prediction models is challenging given demographic shifts, changes in practice patterns, and geographic variation. We propose an approach to tailor known risk prediction models for local settings while preserving known disease risk mechanisms. Here, as a case study, we recalibrate the Pooled Cohort Equations (PCE) in a contemporary (1997-2018), local electronic health records (EHR) cohort in New England. Methods: We conducted a retrospective cohort study using EHR from large Boston-area hospitals. 61,808 patients 45+ years old, without ASCVD in 1997-2006 and complete data to calculate PCE model were split into training (50%), validation (10%), and test sets (40%). We used XGBoost to recalibrate PCE model (XGB-PCE) using PCE risk score, age, race/ethnicity, and sex (Figure). We constrained the XGB-PCE model to preserve known physiologic trends i.e. increase in ASCVD risk with increase in total cholesterol and systolic blood pressure. Models’ discrimination and calibration were calculated in a set aside test set using the Harrell’s c-index and modified Nam D’Agostino χ2 test. Results: In the test set, XGB-PCE exhibited improved calibration across ASCVD risk categories (p=0.37) and race/ethnicity populations (p=0.73). In contrast, even after recalibration in the large, PCE was poorly calibrated across risk categories (p<0.001) and race/ethnicity populations (p<0.001). For model discrimination, both models produced similar c-index (Δc-index of ±.01). Conclusions: Our recalibration approach maintained comparable discriminative performance and demonstrated improved calibration relative to the PCE while preserving known disease risk mechanisms. Methods that are flexible and preserve disease mechanisms have the potential to locally tailor existing models to be better calibrated for contemporary populations.

Evaluating and mitigating bias in machine learning models for cardiovascular disease prediction

ObjectiveThe study aims to investigate whether machine learning-based predictive models for cardiovascular disease (CVD) risk assessment show equivalent performance across demographic groups (such as race and gender) and if bias mitigation methods can reduce any bias present in the models. This is important as systematic bias may be introduced when collecting and preprocessing health data, which could affect the performance of the models on certain demographic sub-cohorts. The study is to investigate this using electronic health records data and various machine learning models. MethodsThe study used large de-identified Electronic Health Records data from Vanderbilt University Medical Center. Machine learning (ML) algorithms including logistic regression, random forest, gradient-boosting trees, and long short-term memory were applied to build multiple predictive models. Model bias and fairness were evaluated using equal opportunity difference (EOD, 0 indicates fairness) and disparate impact (DI, 1 indicates fairness). In our study, we also evaluated the fairness of a non-ML baseline model, the American Heart Association (AHA) Pooled Cohort Risk Equations (PCEs). Moreover, we compared the performance of three different de-biasing methods: removing protected attributes (e.g., race and gender), resampling the imbalanced training dataset by sample size, and resampling by the proportion of people with CVD outcomes. ResultsThe study cohort included 109,490 individuals (mean [SD] age 47.4 [14.7] years; 64.5% female; 86.3% White; 13.7% Black). The experimental results suggested that most ML models had smaller EOD and DI than PCEs. For ML models, the mean EOD ranged from −0.001 to 0.018 and the mean DI ranged from 1.037 to 1.094 across race groups. There was a larger EOD and DI across gender groups, with EOD ranging from 0.131 to 0.136 and DI ranging from 1.535 to 1.587. For debiasing methods, removing protected attributes didn’t significantly reduced the bias for most ML models. Resampling by sample size also didn’t consistently decrease bias. Resampling by case proportion reduced the EOD and DI for gender groups but slightly reduced accuracy in many cases. ConclusionsAmong the VUMC cohort, both PCEs and ML models were biased against women, suggesting the need to investigate and correct gender disparities in CVD risk prediction. Resampling by proportion reduced the bias for gender groups but not for race groups.

Atherosclerotic cardiovascular disease risk and small dense low-density lipoprotein cholesterol in men, women, African Americans and non-African Americans: The pooling project

Background and aimsElevated small dense low-density lipoprotein-cholesterol (sdLDL-C) has been reported to be associated with increased atherosclerotic cardiovascular disease (ASCVD) risk. Our aims were to determine whether direct and calculated sdLDL-C were significant independent ASCVD risk factors in sex and race subgroups. MethodsIn a total of 15,933 participants free of ASCVD at baseline (median age 62 years, 56.7% female, 19.7% African American) fasting plasma lipids and sdLDL-C were measured by standardized automated methods. All subjects were followed for 10 years for incident ASCVD, which developed in 9.7% of subjects. SdLDL-C values were also calculated. Univariate and multivariate analyses were carried out to assess for independent associations with incident ASCVD after adjustment for all standard risk factors. ResultsAll standard risk factors were significantly associated with incident ASCVD on univariate analysis, as were direct and calculated sdLDL-C. These latter parameters were also significant when added to the pooled cohort risk equation. However, associations were significantly stronger for direct sdLDL-C and were not significant for calculated values once direct values were in the model. In contrast to calculated values, top quartile direct sdLDL-C was significantly independently associated with incident ASCVD versus bottom quartile values in all subjects and subgroups, except African Americans (hazard ratios ≥1.50, p < 0.01). Subjects with direct values ≥ 50 mg/dL versus <25 mg/dL had significantly higher independent ASCVD risk in all groups (hazard ratios >1.49, all p < 0.01). ConclusionsHaving a direct small dense low-density lipoprotein cholesterol value ≥ 50 mg/dL is a significant independent ASCVD risk-enhancer.

Conduction Disturbances and Arrhythmia Risk After Septal Reduction Therapy with Alternative Agents: A Pilot Study with EVOH-DMSO and Systematic Review.

Surgical septal myectomy and alcohol septal ablation are recommended treatment modalities for alleviating Left ventricular outflow tract (LVOT) gradient in obstructive HCM. Alcohol septal ablation offers advantages over surgery in many ways. However, it is associated with some life-threatening complications. For this purpose, our center used alternative agents for septal artery embolization. This study compared and evaluated conduction system defects and arrhythmia risk after EVOH-DMSO septal ablation with other alternative agents and alcohol septal ablation. Twenty-five patients who received septal reduction therapy with EVOH-DMSO were analyzed retrospectively, and all non-alcoholic agent's septal ablation studies were systematically reviewed and compared. Twenty-five patients (52% female; mean age: 55.8 ± 17.1) with symptomatic obstructive HCM were enrolled. The Peak LVOT gradient was significantly reduced after the procedure (68 vs. 20 mmHg; P <0.001). During the 12-month follow-up, no mortality occurred. The complete atrioventricular block was noted in 2 (8%) patients. The incidence of right bundle branch block (RBBB) increased after the procedure (pre-procedural 2 patients (8%), post-procedural 9 patients (36%) P = 0.002). On ECG and Holter monitorization, no sustained ventricular tachyarrhythmia occurred during follow-up, and no change was found in the frequency of atrial fibrillation. We systematically compared EVOH-DMSO to other non-alcohol agents, and we found that EVOH-DMSO can cause conduction system problems more commonly than other non-alcohol agents. EVOH-DMSO could cause conduction system problems more common than other non-alcohol agents but less than alcohol septal ablation.

Pregnancy-related factors may signal additional protection or risk of future cardiovascular diseases

BackgroundCardiovascular disease (CVD) guidelines recommend using the Pooled Cohort Equation (PCE) to assess 10-year CVD risk based on traditional risk factors. Pregnancy-related factors have been associated with future CVD. We examined the contribution of two pregnancy-related factors, (1) history of a low birthweight (LBW) infant and (2) breastfeeding to CVD risk accounting for traditional risk factors as assessed by the PCE.MethodsA nationally representative sample of women, ages 40–79, with a history of pregnancy, but no prior CVD, was identified using NHANES 1999–2006. Outcomes included (1) CVD death and (2) CVD death plus CVD surrogates. We used Cox proportional hazards models to adjust for PCE risk score.ResultsAmong 3,758 women, 479 had a LBW infant and 1,926 reported breastfeeding. Mean follow-up time was 12.1 years. Survival models showed a consistent reduction in CVD outcomes among women with a history of breastfeeding. In cause-specific survival models, breastfeeding was associated with a 24% reduction in risk of CVD deaths (HR 0.76; 95% CI 0.45─1.27, p = 0.30) and a 33% reduction in risk of CVD deaths + surrogate CVD, though not statistically significant. (HR 0.77; 95% CI 0.52─1.14, p = 0.19). Survival models yielded inconclusive results for LBW with wide confidence intervals (CVD death: HR 0.98; 95% CI 0.47─2.05; p = 0.96 and CVD death + surrogate CVD: HR 1.29; 95% CI 0.74─2.25; p = 0.38).ConclusionPregnancy-related factors may provide important, relevant information about CVD risk beyond traditional risk factors. While further research with more robust datasets is needed, it may be helpful for clinicians to counsel women about the potential impact of pregnancy-related factors, particularly the positive impact of breastfeeding, on cardiovascular health.

Comparison of Framingham risk score and pooled cohort equations for the prediction of coronary atherosclerosis in patients who meet the target LDL-C level of Korean dyslipidemia guideline.

This study aims to compare the predicting performance of coronary atherosclerosis between Framingham Risk Score (FRS) and Pooled Cohort Equations (PCE) in moderate to high-risk patients who meet the target low-density lipoprotein cholesterol (LDL-C) level of Korean dyslipidemia guidelines. Among 1207 patients aged 40 to 65 who underwent coronary computed tomography angiography at outpatient for chest discomfort, we included 414 moderate-risk patients (non-diabetes) and 86 high-risk patients (diabetes). They were divided into 3 groups according to FRS and PCE, then compared with coronary artery calcification score (CACS) and plaque burden degree strata. We presented receiver operating characteristic curves for the presence of coronary artery calcification (CAC) and any plaque. In moderate-risk patients, the distribution of CACS and plaque burden degree according to FRS and PCE risk strata showed significant differences between groups and a consistent trend (P < .001). Both FRS and PCE showed good discrimination for the presence of CAC [area under the curve (AUC); 0.711 vs 0.75, P = .02] and any plaque (AUC; 0.72 vs 0.756, P = .025). However, in high-risk patients, there was no significant differences or consistent trend between groups and the AUC values of FRS and PCE were (0.537 vs 0.571, P = .809) for CAC and (0.478 vs 0.65 P = .273) for any plaque showing poor discrimination. In predicting coronary atherosclerosis in moderate to high-risk patients who meet the target LDL-C level of Korean dyslipidemia guidelines, both FRS and PCE can be used in moderate-risk patients but not in high-risk patients.

Abstract 9843: Dynamic Importance of Genomic and Nongenomic Risk Factors for Coronary Artery Disease Over the Life Course

Introduction: Current cardiovascular risk models are limited by fixed window estimates of risk rather than lifetime trajectories. We create a novel framework to understand the time-varying profiles of genomic and non-genomic risk factors over the life span. Methods: We examine 325,948 individuals from the UK Biobank (UKB) (57% female, enrollment age 40-70 years, mean follow-up 11.9 years) with baseline clinical risk factors, composite Pooled Cohort Equations (PCE), and coronary artery disease (CAD) polygenic risk score (PRS). We compare Cox hazard ratios (HR) and the proportion of variation explained (PVE) at each age of enrollment. We extend analyses to a younger cohort of 3,821 individuals from the Framingham Health Study (FHS) Offspring Cohort, (51% female, enrollment ages 18-57, mean follow-up 35.1 years). Results: Results: HR and PVE of CAD PRS are greater early in life with HR 2.47, [95% CI 2.45-2.50] at age 40 to HR 1.67 [1.64,1.70] at age 70 in UKB; PVE of 7.8% to 2.3% in UKB. Parallel results were observed in FHS (Fig 1A-C) earlier in the life span. While the variation in lifetime risk increases with time, the HR for single risk factors estimated at each age are dynamic. The PCE, via inclusion of age, absorb much of this variation. Reclassification of intermediate 10-year PCE risk (7.5-19.9%) using PRS is most informative in younger individuals. In the UKB, the top-quintile of PRS, when compared to PCE alone, accurately predicts more individuals under age 55 (1108 versus 844) who ultimately develop CAD, but loses superiority over age 55 (Fig. 1D). Conclusions: Depending on when the question is asked, the HR of CAD of both genomic and non-genomic risk factors changes dramatically, with PRS demonstrating superiority early on. Our novel framework incorporates time-varying effects in portraying risk trajectory rather than fixed-window estimates. We show that polygenic scores also predict lifetime risk of disease early on rather than predicting only early-onset disease.

Association of a Multiancestry Genome-Wide Blood Pressure Polygenic Risk Score With Adverse Cardiovascular Events.

Traditional cardiovascular risk factors and the underlying genetic risk of elevated blood pressure (BP) determine an individual's composite risk of developing adverse cardiovascular events. We sought to evaluate the relative contributions of the traditional cardiovascular risk factors to the development of adverse cardiovascular events in the context of varying BP genetic risk profiles. Genome-wide polygenic risk score (PRS) was computed using multiancestry genome-wide association estimates among US adults who underwent whole-genome sequencing in the Trans-Omics for Precision program. Individuals were stratified into high, intermediate, and low genetic risk groups (>80th, 20-80th, and <20th centiles of systolic BP [SBP] PRS). Based on the ACC/AHA Pooled Cohort Equations, participants were stratified into low and high (10 year-atherosclerotic cardiovascular disease [CVD] risk: <10% or ≥10%) cardiovascular risk factor profile groups. The primary study outcome was incident cardiovascular event (composite of incident heart failure, incident stroke, and incident coronary heart disease). Among 21 897 US adults (median age: 56 years; 56.0% women; 35.8% non-White race/ethnicity), 1 SD increase in the SBP PRS, computed using 1.08 million variants, was associated with SBP (β: 4.39 [95% CI, 4.13-4.65]) and hypertension (odds ratio, 1.50 [95% CI, 1.46-1.55]), respectively. This association was robustly seen across racial/ethnic groups. Each SD increase in SBP PRS was associated with a higher risk of the incident CVD (multivariable-adjusted hazards ratio, 1.07 [95% CI, 1.04-1.10]) after controlling for ACC/AHA Pooled Cohort Equations risk scores. Among individuals with a high SBP PRS, low atherosclerotic CVD risk was associated with a 58% lower hazard for incident CVD (multivariable-adjusted hazards ratio, 0.42 [95% CI, 0.36-0.50]) compared to those with high atherosclerotic CVD risk. A similar pattern was noted in intermediate and low genetic risk groups. In a multiancestry cohort of >21 000 US adults, genome-wide SBP PRS was associated with BP traits and adverse cardiovascular events. Adequate control of modifiable cardiovascular risk factors may reduce the predisposition to adverse cardiovascular events among those with a high SBP PRS.

Socioeconomic status and modification of atherosclerotic cardiovascular disease risk prediction: epidemiological analysis using data from the atherosclerosis risk in communities study

ObjectiveExamine whether the relationship between the pooled cohort equations (PCE) predicted 10-year risk for atherosclerotic cardiovascular disease (ASCVD) and absolute risk for ASCVD is modified by socioeconomic status (SES).DesignPopulation-based longitudinal...

A New Marker for Determining Cardiovascular Risk: Salusin Alpha.

Background and objective Prevention of atherosclerotic cardiovascular diseases (ASCVD) is possible with early recognition of individuals at risk. Salusin alpha is an endogenous bioactive peptide with anti-atherogenic properties. We aimed to reveal the relationship between salusin alpha levels and cardiovascular (CV) risk using the Systematic COronary Risk Estimation 2 (SCORE2) and Pooled Cohort Equation (PCE) algorithms. Methods A total of 137 asymptomatic outpatients were included in the study. The participants were divided into four quartile groups according to the distribution of salusin alpha levels: Q1 (n = 34), Q2 (n = 35), Q3 (n = 34), and Q4 (n = 34). The relationships of salusin alpha with cardiovascular risk scores and groups were investigated. Results The means of SCORE2 and PCE risk scores (11.24 ± 1.24 and 13.30 ± 1.71, respectively; p < 0.001 for both), baseline scores in the presence of SCORE2 and PCE optimal risk factors (4.82 ± 0.71 (p = 0.034); 4.20 ± 0.77 (p = 0.010)), and means of Δ SCORE2 and Δ PCE risk score (6.41 ± 0.67 and 9.10 ± 1.10, respectively; p < 0.001 for both) were significantly higher in the Q1 group. The SCORE2 "very high"cardiovascular (CV)risk group was significantly represented in the Q1 group (n = 17 (50%)), while the "low-moderate" risk group was more heavily represented in the Q4 group (n = 15 (44.1%)) (p < 0.001). The PCE"high"CV risk group was significantly represented in the Q1 group (n = 9 (26.5%)), while the "low" risk group was more intensely represented in the Q4 group (n = 20 (58.8%)) (p < 0.001). Salusin alpha had a significantly negative moderate correlation with SCORE2 and PCE risk scores. As a result of receiver operating characteristic (ROC) analysis, it was found that salusin alpha had significant diagnostic power in the prediction of CV risk groups determined by SCORE2 and PCE risk scores. Salusin alpha was observed to be a risk-reducing factor in SCORE2 and PCE CV risk groups (odds ratio (OR)(95% confidence interval (CI)): 0.989 (0.982-0.997) (p = 0.002) and OR (95% CI): 0.988 (0.978-0.998) (p = 0.009) respectively). Conclusion Salusin alpha has a negative correlation with SCORE2 and PCE risk scores. It has significant diagnostic power in the prediction of CV risk groups and is an independent variable that has a risk-reducing effect in the distribution of risk groups.

Adherence to lifestyle advice and its related cardiovascular disease risk among US adults with high cholesterol.

Little is known about adherence to lifestyle advice from health professionals among people with high blood cholesterol, and its relationship with cardiovascular disease (CVD). We aimed to examine the proportion of adherence to lifestyle advice and its effect on lipid profile and CVD among people with high cholesterol. Within the National Health and Nutrition Examination Survey 1999-2010, the study included adults aged ≥20 years who were recommended to improve lifestyles in diet (N=6645), bodyweight (N=4797), or exercise (N=5594) due to their high cholesterol. Adherent status was self-reported through questionnaires. Lipid measurements were collected from laboratory tests. 10-year Atherosclerotic CVD (ASCVD) risk was estimated by using pooled cohort risk equations for participants aged 40-79 years. CVD mortality up to December 31, 2019 was obtained from the National Death Index. The percentages of adherents were 80.9%, 80.7%, and 72.7% for eating less fat, controlling weight, and increasing exercise, respectively. The percentages of adherents for controlling weight and increasing exercise significantly increased from 1999 to 2000 to 2009-2010 (both Ptrend <0.05). Adherents had a higher high-density lipoprotein cholesterol (HDL-C) and lower total cholesterol/HDL-C ratio than non-adherents (Pfor difference <0.05 for all). Participants adhering to increasing exercise had a lower odds ratio of 10-year intermediate-to-high ASCVD risk (Odd ratio 0.73, 95% confidence interval [CI] 0.56-0.95) and a lower CVD mortality (Hazard ratio 0.70, 95% CI 0.51-0.97) than non-adherents. Lifestyle advice from health staff obtained relatively high and increasing acceptability. Given the better lipid profile and lower CVD risk of adherents, health professionals should be encouraged to recommend lifestyle modifications for adults with high cholesterol in clinical practice.

Sex-specific performance of the ASCVD pooled cohort equation risk calculator as a correlate of coronary artery calcium in Kampala, Uganda

Introduction: The prevalence of cardiovascular disease (CVD) is rising in Sub-Saharan Africa, but it is not known whether current risk assessment tools predict coronary atherosclerosis in the region. Furthermore, sex-specific performance and interaction with HIV serostatus has not been well studied. Methods: This cross-sectional study compared ASCVD risk scores and detectable coronary artery calcium (CAC>0) by sex in Kampala, Uganda (n = 200). The cohort was enriched for persons living with HIV, and all participants had at least one CVD risk factor. We fit log binomial regression models and constructed ROC curves to assess the correlation between ASCVD scores and CAC>0. Results: The mean age was 56. 62% were female and 50% of both men and women were living with HIV. The median 10-year ASCVD risk score was significantly higher in men (11.0%, IQR 7.6–19.4%) than in women (5.1%, IQR 3.2–8.7%), although the prevalence of CAC>0 was similar (8.1 vs 10.5%, p = 0.63). Each 10% increase in ASCVD risk was associated with increased risk of CAC>0 in men (PR 1.59, 95% CI 1.00–2.55, p = 0.05) but not women (PR 1.15, 95% CI 0.44–3.00, p = 0.77). ROC curves demonstrated an AUC of 0.57 for women vs 0.70 for men. Adjustment for HIV serostatus improved the predictive value of ASCVD in women only (AUC 0.78, p = 0.02). Conclusions: ASCVD risk score did not correlate with the presence of CAC in women. When HIV status was added to the ASCVD risk score, correlation with CAC was improved in women but not in men.

P165 CARDIOVASCULAR RISK FACTORS AND HYPOGONADISM INFLUENCE ON CARDIAC OUTCOMES IN AN AGING POPULATION OF BETA–THALASSEMIA PATIENTS: LOOKING AT THE HEART OF THE PROBLEM

Abstract Background β–thalassemia major (β–TM) is a hereditary genetic disease hindered by many comorbidities due to transfusion–related iron. Despite iron chelation therapy increased the life expectancy up to 50 years today, Iron–related heart disease is still a leading complication, with an expected change in cardiac disease drivers with the ageing of these patients. This study aims to evaluate the distribution and prevalence of cardiovascular risk factors in a population of thalassemia major patients, and their relationship with observed cardiovascular events and potential modifying factors. Methods and Results One–hundred fifty–nine β–TM patients older than 18 years of age were included in this study. Mean age was 40.9 ± 8.4 years. Low serum lipid levels with low HDL levels were noted, with 28% having diabetes mellitus and 62% with hypogonadism; Splenectomy was reported in 70%. During the observed period, 3.8% of patients had at least one episode of heart failure, 35.9% showed early signs of heart failure, 22% received a diagnosis of diastolic dysfunction, 38% had a left ventricular ejection fraction &amp;lt;55%, and 21.4% showed supraventricular arrhythmias. Cardiovascular risk was then assessed using two algorithms (CUORE and Pooled Cohort Risk Equation – PCRE) and was generally low despite the high burden of cardiac–related events. Patients with hypogonadism (who showed lower cardiac T2* value than those without; p &amp;lt; 0.001) showed a statistically significant correlation with the occurrence of cardiovascular events. Discussion The β–TM population has a relatively low mean age, but shows a particular metabolic profile associated with numerous comorbidities: an increased prevalence of diabetes mellitus, low HDL values and frequent hypogonadism, which tends to be associated with increased iron deposition in the myocardium. The cardiovascular risk estimated by specific algorithms (CUORE and PCRE) was generally low, probably due to the young age of the cohort and derivation pitfalls when applied to this specific population, but the prevalence of cardiac events was not negligible. Conclusions The chronic accumulation of iron in the heart and the specific metabolic profile, mainly observed in patients with hypogonadism, allows us to define β–TM as a condition with high cardiovascular risk from many points of view (iron–related myopathy, atherosclerosis and arrhythmias), which requires better stratification tools and specific follow–up program.

Automating and improving cardiovascular disease prediction using Machine learning and EMR data features from a regional healthcare system

BackgroundThe ACC/AHA Pooled Cohort Equations (PCE) Risk Calculator is widely used in the US for primary prevention of atherosclerotic cardiovascular disease (ASCVD), but may under- or over-estimate risk in some populations. We therefore designed an automated, population-specific ASCVD risk calculator using machine-learning (ML) methods and electronic medical record (EMR) data, and compared its predictive power with that of the PCE calculator. Methods and FindingsWe collected data from 101,110 unique EMRs of living patients from January 1, 2009 to April 30, 2020. ML techniques were applied to patient datasets that included either only cross-sectional (CS) features, or CS combined with longitudinal (LT) features derived from vital statistics and laboratory values. We compared the utility of the models using a proposed new cost measure (Screened Cases Percentage @ Sensitivity level).All ML models tested achieved better predictive power than the PCE risk calculator. The random forest ML technique (RF) applied on the combination of CS and LT features (RF-LTC) produced the best area under curve (AUC) score of 0.902 (95% confidence interval (CI), 0.895–0.910). To detect 90% of all positive ASCVD cases, the best ML model required screening only 43% of patients, while the PCE risk calculator required screening 69% of patients. ConclusionsPrediction models built using ML techniques improved ASCVD prediction and reduced the number of screenings required to predict ASCVD when compared with the PCE calculator, alone. Combining LT and CS features in the ML models significantly improved ASCVD prediction compared with using CS features, alone.

Abstract P217: Cardiovascular Events Among Adults Who Qualify For Statin Therapy Based On Age Alone Alone: A Reasons For Geographical And Racial Differences In Stroke (REGARDS) Study

Background: Predicted 10-year atherosclerotic cardiovascular disease (ASCVD) risk in the Pooled Cohort equation increases substantially with age. Some older adults have estimated ASCVD risk ≥7.5% without elevated risk factors; it is unclear whether the observed ASCVD risk is high in this group. We compared the incidence of ASCVD between adults with an estimated 10-year ASCVD risk ≥7.5% based on age alone vs those with elevated risk factors. Methods: Adults ages 45-79 enrolled in the REasons for Geographic and Racial Differences in Stroke (REGARDS) study without a history of ASCVD, diabetes mellitus, or statin use were included (i.e, a primary prevention subset). Ten-year ASCVD risk was estimated using the Pooled Cohort Risk Equation. Participants were categorized into 3 groups according to the predicted ASCVD risk: &lt;7.5%, “age-only” ≥7.5%, and “elevated factors” ≥7.5%. Participants with “age-only” risk ≥7.5% were nonsmokers with untreated SBP &lt;130 mmHg, total cholesterol &lt;200 mg/dL, and HDL-C ≥50 mg/dL for women or ≥40 mg/dL for men. All other participants with predicted ASCVD ≥7.5% were considered “elevated factors” ≥7.5%. Incident ASCVD was defined as ischemic or hemorrhagic stroke, nonfatal MI, or CHD death. Results: Among 10,742 participants, 891 ASCVD events occurred over a median follow-up of 11.1 (7.1 - 13.0) years. The “age-only” risk ≥7.5% group included 352 (3.3%) participants with mean age 69.9 ± 5.0. The risk &lt;7.5%, and “elevated factors” risk ≥7.5% had a mean age of 56.3 ± 5.9, and 66.3 ± 7.1 years, respectively. Compared to the risk &lt;7.5% group, the “age-only” and “elevated factors” risk ≥7.5% group experienced 2.6-fold and 3.6-fold higher event rates, respectively (Table). Conclusions: Few adults in this primary prevention population had predicted risk ≥7.5% based on age alone. Even without elevated risk factors, adults with predicted ASCVD risk ≥7.5% based on age alone experienced substantial event rates consistent with their predicted risk.

Update on hormone therapy for the management of postmenopausal women.

Hormone therapy (HT) has been used in postmenopausal women for decades in clinical practice. With further analysis and newer studies, the benefits and risks of HT have been repeatedly verified and discussed. HT is recommended for the treatment of vasomotor symptoms (VMS), genitourinary syndrome of menopause (GSM) and the prevention of osteoporosis. However, the precise association between HT and the risks of cardiovascular diseases, venous thromboembolism, neurodegenerative diseases, breast cancer, and endometrial cancer remains controversial. Therefore, determining how to take advantage of and control the risks of HT by adjusting the initiation time, regimen, and duration is crucial. Recent studies have indicated that HT is not related to the risk of all-cause, cardiovascular, or breast cancer mortality although it might increase the incidence of some chronic diseases. For symptomatic postmenopausal women under the age of 60 without contraindications, early initiation of HT is safe and probably has a mortality benefit over the long term. Initiating HT close to menopause at the lowest effective dose is more likely to have maximal benefits and the lowest risks. Transdermal and vaginal HT may have a lower risk, but recent evidence suggests additional clinical benefits of oral HT formulations in relieving VMS and preventing osteoporosis. The pooled cohort risk equation for atherosclerotic cardiovascular disease (ASCVD) and the free app named Menopro can be used to perform individual risk assessments. In addition, Chinese herbal medicines have benefits in alleviating hot flashes, depression, and menopausal symptoms, although further data are needed to strongly support their efficacy. Acupuncture and electroacupuncture have definite efficacy in the treatment of postmenopausal symptoms with few adverse effects, so they are a reasonable option as an alternative therapy for high-risk women. This review discusses the history of, guidelines on, and strategies for HT in order to make suggestions based on the most up-to-date evidence for the management of postmenopausal women.

Lifetime Cardiovascular Disease Risk by Coronary Artery Calcium Score in Individuals With and Without Diabetes: An Analysis From the Multi-Ethnic Study of Atherosclerosis.

To assess lifetime cardiovascular disease (CVD) risk by coronary artery calcium (CAC) score in individuals with diabetes from the Multi-Ethnic Study of Atherosclerosis (MESA) and compare risk with that in individuals without diabetes. We developed a microsimulation model with well, diabetes, post-CVD, and death health states using multivariable time-dependent Cox regression with age as time scale. We initially used 10-year follow-up data of 6,769 MESA participants, including coronary heart disease (CHD) (n = 272), heart failure (n = 201), stroke (n = 186), and competing death (n = 619) and assessed predictive validity at 15 years. We externally validated the model in matched National Health and Nutrition Examination Survey (NHANES) participants. Subsequently, we predicted CVD risk until age 100 years by diabetes, 10-year pooled cohort equations risk, and CAC score category (0, 1-100, or 100+). The model showed good calibration and discriminative performance at 15 years, with discrimination indices 0.71-0.78 across outcomes. In the NHANES cohort, predicted 15-year mortality risk corresponded well with Kaplan-Meier risk, especially for those with diabetes: 29.6% (95% CI 24.9-34.8) vs. 32.4% (95% CI 27.2-37.2), respectively. Diabetes increased lifetime CVD risk, similar to shifting one CAC category upward (from 0 to 1-100 or from 1-100 to 100+). Patients with diabetes and CAC score of 0 had a lifetime CVD risk that overlapped with that of individuals without diabetes who were at low 10-year pooled cohort equations risk (<7.5%). Patients with diabetes carry a spectrum of CVD risk. CAC scoring may improve decisions for preventive interventions for patients with diabetes by better delineating lifetime CVD risk.

Body Shape Index and Cardiovascular Risk in Individuals With Obesity.

Introduction: Cardio-metabolic risks are tested to show various anthropometric measurements. This study aimed to evaluate a body shape index (ABSI) of individuals with obesity to determine the role of these measurements in cardiovascular risk prediction.Methods: This cross-sectional study included 368 patients who were evaluated by the researcher in the polyclinic. Sociodemographic information was obtained, and anthropometric measurements were made. Body mass index (BMI), waist circumference, and ABSI were evaluated in all patients. The patient’s risk of developing cardiovascular disease was calculated from the pooled cohort equations risk calculator (PRCAE), Framingham risk score, and systematic coronary risk evaluation (SCORE) risk calculation systems.Results: Of the 368 patients in the study, 302 (82.1%) were females, and 66 (17.9%) were males. The average age of participants was 46.2 ± 12.0 years. The median BMI of participants was 37.5 (34.0-42.4) kg/m2. The median ABSI of participants was 0.0816 (0.0775-0.0849). A positive correlation was found between ABSI and Framingham risk score and PRCAE risk score (r = 0.297, p = 0.000 and r = 0.305, p = 0.000, respectively). A significant relationship was found between ABSI and Framingham, PRCAE, and SCORE risk groups (p = 0.000, p = 0.000, and p = 0.000, respectively).Conclusions: Our study results revealed a significant association of ABSI with Framingham, PRCAE, and SCORE risk calculation systems, which helps predict cardiovascular risk.