Coronary Heart Disease Risk Prediction Research Articles

Enhancing Coronary Heart Disease Risk Prediction with Machine Learning and Deep Learning

Enhancing Coronary Heart Disease Risk Prediction with Machine Learning and Deep Learning

Predictive model and risk analysis for coronary heart disease in people living with HIV using machine learning

ObjectiveThis study aimed to construct a coronary heart disease (CHD) risk-prediction model in people living with human immunodeficiency virus (PLHIV) with the help of machine learning (ML) per electronic medical records (EMRs).MethodsSixty-one medical characteristics (including demography information, laboratory measurements, and complicating disease) readily available from EMRs were retained for clinical analysis. These characteristics further aided the development of prediction models by using seven ML algorithms [light gradient-boosting machine (LightGBM), support vector machine (SVM), eXtreme gradient boosting (XGBoost), adaptive boosting (AdaBoost), decision tree, multilayer perceptron (MLP), and logistic regression]. The performance of this model was assessed using the area under the receiver operating characteristic curve (AUC). Shapley additive explanation (SHAP) was further applied to interpret the findings of the best-performing model.ResultsThe LightGBM model exhibited the highest AUC (0.849; 95% CI, 0.814–0.883). Additionally, the SHAP plot per the LightGBM depicted that age, heart failure, hypertension, glucose, serum creatinine, indirect bilirubin, serum uric acid, and amylase can help identify PLHIV who were at a high or low risk of developing CHD.ConclusionThis study developed a CHD risk prediction model for PLHIV utilizing ML techniques and EMR data. The LightGBM model exhibited improved comprehensive performance and thus had higher reliability in assessing the risk predictors of CHD. Hence, it can potentially facilitate the development of clinical management techniques for PLHIV care in the era of EMRs.

Parental genetically predicted liability for coronary heart disease and risk of adverse pregnancy outcomes: a cohort study

BackgroundAdverse pregnancy outcomes (APO) may unmask or exacerbate a woman’s underlying risk for coronary heart disease (CHD). We estimated associations of maternal and paternal genetically predicted liability for CHD with lifelong risk of APOs. We hypothesized that associations would be found for women, but not their male partners (negative controls).MethodsWe studied up to 83,969‬ women (and up to 55,568‬ male partners) from the Norwegian Mother, Father and Child Cohort Study or the Trøndelag Health Study with genotyping data and lifetime history of any APO in their pregnancies (1967–2019) in the Medical Birth Registry of Norway (miscarriage, stillbirth, hypertensive disorders of pregnancy, gestational diabetes, small for gestational age, large for gestational age, and spontaneous preterm birth). Maternal and paternal genetic risk scores (GRS) for CHD were generated using 148 gene variants (p-value < 5 × 10−8, not in linkage disequilibrium). Associations between GRS for CHD and each APO were determined using logistic regression, adjusting for genomic principal components, in each cohort separately, and combined using fixed effects meta-analysis.ResultsOne standard deviation higher GRS for CHD in women was related to increased risk of any hypertensive disorders of pregnancy (odds ratio [OR] 1.08, 95% confidence interval [CI] 1.05–1.10), pre-eclampsia (OR 1.08, 95% CI 1.05–1.11), and small for gestational age (OR 1.04, 95% CI 1.01–1.06). Imprecise associations with lower odds of large for gestational age (OR 0.98, 95% CI 0.96–1.00) and higher odds of stillbirth (OR 1.04, 95% CI 0.98–1.11) were suggested. These findings remained consistent after adjusting for number of total pregnancies and the male partners’ GRS and restricting analyses to stable couples. Associations for other APOs were close to the null. There was weak evidence of an association of paternal genetically predicted liability for CHD with spontaneous preterm birth in female partners (OR 1.02, 95% CI 0.99–1.05), but not with other APOs.ConclusionsHypertensive disorders of pregnancy, small for gestational age, and stillbirth may unmask women with a genetically predicted propensity for CHD. The association of paternal genetically predicted CHD risk with spontaneous preterm birth in female partners needs further exploration.

Serum lipoprotein(a) and reclassification of coronary heart disease risk; application of prediction in a cross-sectional analysis of an ongoing Iranian cohort

IntroductionRecent studies have introduced elevated lipoprotein(a) (Lp(a)) as a risk factor for coronary heart disease (CHD). This study investigated whether the addition of Lp(a) as a novel biomarker to the Framingham Risk Score (FRS) model improves CHD risk prediction.MethodsThe study included 1101 Iranian subjects (443 non-diabetic and 658 diabetic patients) who were followed for 10 years (2003–2013). Lp(a) levels and CHD events were recorded for each participant.ResultsThe Net Reclassification Index (NRI) after adding Lp(a) to the FRS model was 19.57% and the discrimination slope was improved (0.160 vs. 0.173). The Akaike Information Criterion (AIC), a measure of model complexity, decreased significantly after adding Lp(a) to the FRS model (691.9 vs. 685.4, P value: 0.007).ConclusionsThe study concluded that adding Lp(a) to the FRS model improves CHD risk prediction in an Iranian population without making the model too complex. This could help clinicians to better identify individuals who are at risk of developing CHD and to implement appropriate preventive measures.

Development and Validation of a Novel Model for Predicting Coronary Heart Disease in Snoring Hypertensive Patients with Hyperhomocysteinemia.

Hypertensive patients with snoring and elevated plasma homocysteine levels are common. When these factors are combined, the risk of coronary heart disease (CHD) is high. Herein, we developed and validated an easy-to-use nomogram to predict high-risk CHD in snoring hypertensive patients with elevated plasma homocysteine.Snoring patients (n = 1,962) with hyperhomocysteinemia and hypertension were divided into training (n = 1,373, 70%) and validation (n = 589, 30%) sets. We extracted CHD predictors using multivariate Cox regression analysis, then constructed a nomogram model. Internal validation using 1,000 bootstrap resampling was performed to assess the consistency and discrimination of the predictive model using the area under the receiver operating characteristic curve (AUC) and calibration plots.We constructed a nomogram model with the extracted predictors, including age, waist-height ratio, smoking, and low-density lipoprotein cholesterol levels. The AUCs of the training and validation cohorts at 80 months were 0.735 (95% CI: 0.678-0.792) and 0.646 (95% CI: 0.547-0.746), respectively. The consistency between the observed CHD survival and the probability of CHD survival in the training and validation sets was acceptable based on the calibration plots. A total of more than 151 points in the nomogram can be used in the identification of high-risk patients for CHD among snoring hypertensive patients with elevated plasma homocysteine.We developed a CHD risk prediction model for snoring hypertension patients with hyperhomocysteinemia. Our findings provide a useful clinical tool for the rapid identification of high-risk CHD at an early stage.

The predictive accuracy of coronary heart disease risk prediction models in rural Northwestern China

Cardiovascular risk models developed may have limitations when applied to rural Chinese. This study validated and compared the Framingham Risk Score (FRS) and Prediction for Atherosclerotic Cardiovascular Disease Risk in China (PAR) models in predicting 10-year risk of coronary heart disease (CHD) in a rural cohort in Ningxia, China from 2008 to 2019. The FRS and PAR models were validated by estimating predicted events, C index, calibration χ2 and plots. 1381 adults without CHD at baseline were followed up for 9.75 years on average. 168 CHD cases were observed. The FRS and PAR underestimated CHD events by 22 % and 46 % for the total population, while overestimated for males by 152 % and 78 %, respectively. The C index was slightly higher for PAR than FRS. Both models showed weak calibration with chi-square values above 20 (p < 0.001). Bland-Altman plots indicated FRS predicted higher CHD risk than PAR, lacking consistency. Overall, FRS and PAR demonstrated limited performance in predicting 10-year CHD risk in this rural population. PAR had slightly better discrimination than FRS, but require further improvement in calibration and individual risk estimation to suit the rural population in Northwest China.

Abstract 11652: The Addition of Social Determinants of Health to Coronary Heart Disease Risk Prediction

Introduction: Coronary heart disease (CHD) remains a leading cause of death worldwide, with an unequal distribution of this burden placed on individuals of lower socioeconomic status. Individual social determinants of health (SDoH) have been shown to be associated with CHD; however, whether the addition of SDoH can improve current CHD risk prediction algorithms has not been delineated. Hypothesis: We hypothesize that the addition of SDoH to the Multi-Ethnic Study of Atherosclerosis (MESA) risk score improves the prediction of 10-year CHD events. Methods: In 6,286 individuals from the MESA cohort, we evaluated the additional predictive value of SDoH beyond the traditional factors included in the existing MESA Risk Score. SDoH considered were employment, family income, house tenure, financial strain, education, marital status, social support, chronic stress, discrimination, neighborhood characteristics, healthcare access, and health insurance, which were organized into five composite domains. A penalized Cox proportional hazards model was used to select SDoH variables with the potential to improve prediction, then an unpenalized Cox regression model was fitted with traditional risk factors plus the selected SDoH variables. Results: Statistically significant differences in the distribution of SDoH were observed between racial and ethnic groups. Notably, unfavorable SDoH were concentrated in Black and Hispanic/Latino persons. Economic stability, education, and community and social context were associated with CHD events. The area under the curve (AUC) for prediction of a 10-year CHD event was 0.816 without SDoH factors and 0.818 with the addition of SDoH. Conclusions: Racial and ethnic differences in social determinants of health were obsserved. Economic stability, education, and community and social context were significantly associated with cardiovascular events, however the addition of SDoH did not significantly improve CHD risk prediction.

Genetics and Ischemic Heart Disease: Should We Opt for Genetic Testing for Primary Prevention?

Cardiovascular diseases (CVD) are a prevalent health concern within the general population of Pakistan, where the average lifespan is notably lower than the global average, with men typically living to 67 years and women to 69 years. According to the 2019 Global Burden of Disease study, Pakistan had an estimated age-standardized incidence rate of CVD at 918.18 per 100,000 (compared to the global rate of 684.33 per 100,000), along with an age-standardized death rate of 357.88 per 100,000 (globally, this rate is 239.85 per 100,000).1 Coronary heart disease (CHD), as revealed by the Framingham Heart Study focusing on individuals aged 40 to 94 without prior heart disease, displayed a lifetime risk of 49% for men and 32% for women when reaching the age of 40.2 There has been a declining trend in death rates in the United States attributed to CVD, CHD, and stroke since 1975. Data from 2000 to 2008 also indicate a decline in CHD mortality.3 Worryingly, the World Health Organization (WHO) reports a concerning rise in CHD-related fatalities in Pakistan. In 2020, 240,720 individuals died due to CHD, accounting for 16.49% of all deaths. This highlights an escalating trend of CHD-related mortality in Pakistan. It's important to note that most individuals presenting with cardiac events have one or more established or borderline risk factors aside from age and gender.4-6 While some essential risk factors are discernible, others may remain elusive. The screening of these risk factors and the evidence for targeted therapeutic interventions are still emerging and require further exploration.7 The starting point for assessing CVD risk factors is variables used to predict major cardiovascular events. These include age, sex, blood pressure, cholesterol levels, diabetes mellitus, and smoking status. Although risk assessment tools like the Pooled Cohort Equation in 2014 and Astro-CHARM have been developed, they have yet to provide satisfactory assessments for potential new CVD risk factors.8 CHD is recognized as a multifactorial disorder resulting from genetic and environmental factors interplay. Environmental risk factors have been identified in approximately 80% of CHD cases.9 Several risk scores, such as the Framingham Risk Score, PROCAM, Reynolds Risk Score, and QRISK 2, have been proposed to guide the use of statins in high-risk groups.10-14 Yet, these risk scores often lack precision and may either overestimate or underestimate future CHD events.15,16 The variation in disease susceptibility among individuals with similar environmental factors and conventional coronary artery disease risk factors (CRFs) may be attributed to genetic variations.17 Genetic analysis can potentially enhance risk discrimination beyond the consideration of CRFs alone. Family history of heart disease, accounting for more than 40% of risk estimation, has long been considered a part of CRFs.18 Candidate gene studies have been conducted to identify common variants in genes associated with disease pathways.19 Single-nucleotide polymorphisms (SNPs) have been employed as markers of genetic diversity. Among these SNPs, those located on the 9p21 locus have shown the strongest association with CHD risk to date.20,21 However, despite the clear link between these variants and incident CHD, 9p21 locus SNPs have not definitively improved the prediction or classification of CHD risk compared to traditional risk factors.22-24 It is important to note that most genetic studies on CHD have predominantly focused on European/Caucasian populations, and their applicability to the South Asian population, including Pakistan, requires further investigation.25,26 In this context, the Pakistani population, much like other Asian countries, is underrepresented in genetic research on CHD. Shahid SU et al. did some work in this respect,27 showing 21 SNPs risk score for genetic risk analysis in the Pakistani population. In conclusion, while different risk assessment tools have been developed for the Pakistani population aged 40 years and above, there is an urgent need to expand cardiac risk evaluation by identifying genetic markers related to CHD, particularly in the younger population. This will be crucial for advancing our understanding of CHD risk factors and developing more effective prevention and intervention strategies.

Prediction of Coronary Heart Disease Events in Blood or Marrow Transplantation Recipients.

The long-term risk of coronary heart disease (CHD) and clinical models that predict this risk remain understudied in blood or marrow transplantation (BMT) recipients. This study sought to examine the risk of CHD after BMT and identify the associated risk factors. Participants included patients transplanted between 1974 and 2014 at City of Hope, University of Minnesota, or University of Alabama at Birmingham and those who survived≥2 years after BMT. Multivariable logistic regression models assessed CHD risk in BMT survivors compared with a sibling cohort. A self-reported questionnaire and medical records provided information regarding sociodemographics, comorbidities, and therapeutic exposures, which were used to develop a CHD risk prediction nomogram. Overall, 6,677 BMT recipients participated; the mean age at BMT was 43.9 ± 17.7 years, 58.3% were male, and 73.3% were non-Hispanic Whites. The median length of follow-up was 6.9 years (range: 2-46.2 years) from BMT. CHD was reported in 249 participants, with a 20-year cumulative incidence of 5.45% ± 0.39%. BMT survivors had a 1.6-fold greater odds of CHD compared with a sibling cohort (95% CI: 1.09-2.40). A nomogram was then developed to predict the risk of CHD at 10 and 20 years after BMT including age at BMT (HR: 1.06/y; 95% CI: 1.04-1.08), male sex (HR: 1.89; 95% CI: 1.15-3.11), a history of smoking (HR: 1.61; 95% CI: 1.01-2.58), diabetes (HR: 2.45; 95% CI: 1.23-4.89), hypertension (HR: 2.02; 95% CI: 1.15-3.54), arrhythmia (HR: 1.90; 95% CI: 0.89-4.06), and pre-BMT chest radiation (yes vs no: HR: 2.83; 95% CI: 1.20-6.67; unknown vs no: HR: 0.88; 95% CI: 0.34-2.28). The C-statistic was 0.77 in the test set (95% CI: 0.70-0.83). This study identified BMT recipients at high risk for CHD, informing targeted screening for early detection and aggressive control of risk factors.

Polygenic risk score in comparison with C-reactive protein for predicting incident coronary heart disease

Background and aimsDespite interest in the use of polygenic risk scores (PRS) for predicting coronary heart disease (CHD) risk, the clinical utility of PRS compared to conventional risk factors has not been demonstrated. We compared the performance of PRS with that of high-sensitivity C-reactive protein (hsCRP) in two well-established cohorts. MethodsThe study population included individuals of European ancestry free of baseline CHD from ARIC (N = 13,113) and the Framingham Offspring Study (FHS) (N = 2,696). The primary predictors included a validated PRS consisting of >6.6 million single nucleotide polymorphisms and hsCRP. The outcome was incident CHD, defined as non-fatal or fatal myocardial infarction. We compared the performance of both predictors after adjusting for the Pooled Cohort Equations in multivariable-adjusted Cox regression models. We assessed discrimination and reclassification using c-statistics and net reclassification improvement. ResultsIncident CHD occurred in 565 ARIC and 153 FHS participants. In multivariable-adjusted models, both PRS and hsCRP were associated with incident CHD (p < 0.05 in both cohorts). In models incorporating both predictors, strengths of association were similar. For instance, in ARIC, the hazard ratio per SD increment was 1.38 (95% CI, 1.27–1.50, p = 2.94 × 10−14) for PRS and 1.41 (1.30–1.55, p = 3.10 × 10−15) for hsCRP. Neither predictor significantly increased model discrimination or net reclassification when compared with models containing the Pooled Cohort Equations alone. ConclusionsIn two independent cohorts, PRS performed similarly to hsCRP for the prediction of CHD risk. These findings suggest PRS does not have unique clinical utility beyond this widely-available, inexpensive measure of risk in unselected middle-aged populations.

Levels and clinical significance of the m6A methyltransferase METTL14 in patients with coronary heart disease.

To investigate the association of methyltransferase-like protein 14 (METTL14) expression with coronary heart disease (CHD). Three hundred and sixteen patients who attended Henan Provincial People's Hospital between June 2019 and February 2021 with principal symptoms of pain or tightness in the chest and who underwent coronary angiography for definitive diagnosis were enrolled. The uric acid, TG, TC, LDL-C, HDL-C, apolipoprotein A1, free fatty acid, lipoprotein a, homocysteine, CRP, and SAA levels were examined. The levels of METTL14, TNF-α, MCP-1, VCAM-1, ICAM-1, and IL-6 were evaluated by ELISA. Patients with CHD had significantly higher m6A methyltransferase activity. In addition, the incidence of diabetes and hypertension, as well as the concentrations of TC, CRP, and SAA were higher in CHD patients. Patients with coronary lesion branches also had significantly increased TG, LDL-C, CRP, and SAA levels. TNF-α, MCP-1, VCAM-1, ICAM-1, and IL-6 expression was also markedly increased in the CHD group (P < 0.001) as was the expression of METTL14 (P < 0.001). The METTL14 expression levels also differed significantly in relation to the number of branches with lesions (P < 0.01) and were correlated with SAA, VCAM-1, ICAM-1, IL-6, and the Gensini score. ROC curve analyses of METTL14 in CHD indicated an AUC of 0.881 (0.679, 0.894) with a cut-off value of 342.37, a sensitivity of 77%, and a specificity of 84%. MCP-1, VCAM-1, IL-6, SAA, and METTL14 were found to independently predict CHD risk. METTL14 levels were found to be positively associated with inflammatory markers and to be an independent predictor of CHD risk.

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

Coronary artery calcium score and polygenic risk score have each separately been proposed as novel markers to identify risk of coronary heart disease (CHD), but no prior studies have directly compared these markers in the same cohorts. To evaluate change in CHD risk prediction when a coronary artery calcium score, a polygenic risk score, or both are added to a traditional risk factor-based model. Two observational population-based studies involving individuals aged 45 years through 79 years of European ancestry and free of clinical CHD at baseline: the Multi-Ethnic Study of Atherosclerosis (MESA) study involved 1991 participants at 6 US centers and the Rotterdam Study (RS) involved 1217 in Rotterdam, the Netherlands. Traditional risk factors were used to calculate CHD risk (eg, pooled cohort equations [PCEs]), computed tomography for the coronary artery calcium score, and genotyped samples for a validated polygenic risk score. Model discrimination, calibration, and net reclassification improvement (at the recommended risk threshold of 7.5%) for prediction of incident CHD events were assessed. The median age was 61 years in MESA and 67 years in RS. Both log (coronary artery calcium+1) and polygenic risk score were significantly associated with 10-year risk of incident CHD (hazards ratio per SD, 2.60; 95% CI, 2.08-3.26 and 1.43; 95% CI, 1.20-1.71, respectively), in MESA. The C statistic for the coronary artery calcium score was 0.76 (95% CI, 0.71-0.79) and for the polygenic risk score, 0.69 (95% CI, 0.63-0.71). The change in the C statistic when each was added to the PCEs was 0.09 (95% CI, 0.06-0.13) for the coronary artery calcium score, 0.02 (95% CI, 0.00-0.04) for the polygenic risk score, and 0.10 (95% CI, 0.07-0.14) for both. Overall categorical net reclassification improvement was significant when the coronary artery calcium score (0.19; 95% CI, 0.06-0.28) but was not significant when the polygenic risk score (0.04; 95% CI, -0.05 to 0.10) was added to the PCEs. Calibration of the PCEs and models with coronary artery calcium and/or polygenic risk scores was adequate (all χ2<20). Subgroup analysis stratified by the median age demonstrated similar findings. Similar findings were observed for 10-year risk in RS and in longer-term follow-up in MESA (median, 16.0 years). In 2 cohorts of middle-aged to older adults from the US and the Netherlands, the coronary artery calcium score had better discrimination than the polygenic risk score for risk prediction of CHD. In addition, the coronary artery calcium score but not the polygenic risk score significantly improved risk discrimination and risk reclassification for CHD when added to traditional risk factors.

Relationship between serum high-sensitivity C-reactive protein (hs-CRP) and glycated hemoglobin (HbA1c) in South Indian population

Introduction and Aim: Cardiovascular disease are a leading cause of death in the world. Atherosclerosis plays the most important role in its pathophysiology. It is a slowly progressing chronic inflammatory process resulting in increased levels of inflammatory markers in serum. hs-CRP is a novel biomarker of coronary heart disease risk prediction. Type 2 diabetes mellitus individuals are at an increased risk of developing cardiovascular diseases. In this study, we aim to find the association of HbA1c with hs-CRP in the South Indian population. Materials and Methods: This was a retrospective study. The study participants were categorised into 3 groups based on their serum hs-CRP levels: Group 1 with hs-CRP &lt; 1.0 mg/L; group 2 with hs-CRP 1.0 – 3.0 mg/L and group 3 with hs-CRP &gt; 3.0 mg/L. The sample size was estimated to be 40 for each group. Age, fasting blood glucose, postprandial blood glucose, HbA1c, hs-CRP, total count, and ESR were collected. Statistical analyses were performed with SPSS version 16.0. Results: Age showed no significant difference among the groups. FBS, PPBS, HbA1c, TLC, and ESR showed a significant difference among the 3 groups. TLC, FBS, PPBS, and HbA1c had a significant positive correlation with hs-CRP. Conclusion: hs-CRP has a statistically significant positive correlation with HbA1c, fasting blood sugar, and postprandial blood sugar Additionally, we also observed that total leukocyte count had a significant positive correlation with hs-CRP and HbA1c. Hence, our results show that impaired glycemic control is associated with an increased risk of developing cardiovascular diseases.

Coronary Artery Calcium Density and Cardiovascular Events by Volume Level: The MESA.

The Agatston coronary artery calcium (CAC) score provides robust cardiovascular disease risk prediction but upweights plaque area by a density factor. Density, however, has been shown to be inversely associated with events. Using CAC volume and density separately improves risk prediction, but it is unclear how to apply this method clinically. We aimed to evaluate the association between CAC density and cardiovascular disease across the spectrum of CAC volume to better understand how to incorporate these metrics into a single score. We performed an analysis of MESA (Multi-Ethnic Study of Atherosclerosis) participants with detectable CAC to evaluate the association between CAC density and events by level of CAC volume using multivariable Cox regression models. In a cohort of 3316 participants, there was a significant interaction (P<0.001) between CAC volume and density for coronary heart disease (CHD) risk (myocardial infarction, CHD death, resuscitated cardiac arrest). Models using CAC volume and density resulted in improvement in the C-index (0.703, SE 0.012 versus 0.687, SE 0.013) and a significant net reclassification improvement (0.208 [95% CI, 0.102-0.306]) compared with the Agatston score for CHD risk prediction. Density was significantly associated with lower CHD risk at volumes ≤130 mm3 (hazard ratio, 0.57 per unit of density [95% CI, 0.43-0.75]), but the inverse association at volumes >130 mm3 was not significant (hazard ratio, 0.82 per unit of density [95% CI, 0.55-1.22]). The lower risk for CHD associated with higher CAC density varied by level of volume, and volume ≤130 mm3 is a potentially clinically useful cut point. Further study is needed to integrate these findings into a unified CAC scoring method.

Sex Hormone-Binding Globulin and Risk of Coronary Heart Disease in Men and Women.

The role of sex hormone-binding globulin (SHBG) levels in clinical risk stratification and intervention for coronary heart disease (CHD) remains uncertain. We aimed to examine whether circulating levels of SHBG are predictive of CHD risk in men and women. We investigated the association between SHBG and the risk of incident CHD in 128 322 men and 135 103 women free of CHD at baseline in the prospective United Kingdom Biobank (UKB) cohort. The unconfounded associations were estimated using Mendelian randomization (MR) analysis. We further conducted a meta-analysis to integrate currently available prospective evidence. CHD events included nonfatal and fatal myocardial infarction and coronary revascularization. In the UKB, during a median of 11.7 follow-up years, 10 405 men and 4512 women developed CHD. Serum levels of SHBG were monotonically associated with a decreased risk of CHD in both men (adjusted hazard ratio [HR] per log nmol/L increase in SHBG: 0.88 [0.83-0.94]) and women (HR: 0.89 [0.83-0.96]). MR-based analyses suggested causality and a dose-response relationship of SHBG with CHD risk. A cumulative meta-analysis including 216 417 men and 138 282 women from 11 studies showed that higher levels of SHBG were prospectively associated with decreased CHD risk in men comparing the highest with the lowest quartile: pooled relative risk (RR) 0.81 (0.74-0.89) and women (pooled RR: 0.86 [0.78-0.94]). Higher circulating SHBG levels were directly and independently predictive of lower CHD risk in both men and women. The utility of SHBG for CHD risk stratification and prediction warrants further study.

PP37 A Systematic Review of Machine Learning and Statistical Models For Predicting Coronary Heart Disease In Diabetic Patients

IntroductionRisk prediction models, using either machine learning or statistical algorithms, can act as inputs of a cost-effectiveness model when predicting costs and effectiveness of an intervention. This systematic review has two objectives: to evaluate methodological quality of the published models to predict diabetic coronary heart disease (CHD) risk; to evaluate whether the models were sufficiently reported to judge their applicability to the cost-effectiveness modelling.MethodsA targeted review of journal articles published in English, Dutch, Chinese, or Spanish was undertaken in PubMed, Embase, Scopus, Web of Science, and IEEE Explore from 1 January, 2016 to 31 May, 2021. To assess the methodological quality and reporting of the models, we used PROBAST (Prediction model Risk Of Bias Assessment Tool), CHARMS (a Checklist for critical Appraisal and data extraction for systematic Reviews of prediction Modelling Studies), and a checklist (Betts 2019) summarizing the application of cardiovascular risk prediction models to health technology assessment.ResultsOur search retrieved 6,579 hits, of which 18 models were eligible for inclusion. Among them, four studies developed machine learning models (2 recurrent neural networks, 1 random forest models, and 1 multi-task learning model) while 14 studies developed statistical models (8 Cox models, 5 logistic models, and 1 microsimulation model). More than 70 percent of models were of high methodological quality in aspects of participants (89%), predictors (72%), and outcomes (72%), while only five models (28%) in aspects of statistical analysis. For the reporting, only two models provided sufficient evidence in all aspects (i.e., participants, predictors, and outcomes) for judging their applicability to the cost-effectiveness modelling. Most models were reported sufficiently regarding participants (78%) and outcomes (72%), but only three models regarding predictors (17%).ConclusionsTo apply the CHD risk prediction models to cost-effectiveness modelling, concerns remain regarding the potential risk of bias due to inappropriate use of analysis methods, and regarding insufficient reporting on how to measure and assess the predictors.

Race-Dependent Association of High-Density Lipoprotein Cholesterol Levels With Incident Coronary Artery Disease

BackgroundPlasma lipids are risk factors for coronary heart disease (CHD) in part because of race-specific associations of lipids with CHD. ObjectivesThe purpose of this study was to understand why CHD risk equations underperform in Black adults. MethodsBetween 2003 and 2007, the REGARDS (REasons for Geographic and Racial Differences in Stroke) cohort recruited 30,239 Black and White individuals aged ≥45 years from the contiguous United States. We used Cox regression models adjusted for clinical and behavioral risk factors to estimate the race-specific hazard of plasma lipid levels with incident CHD (myocardial infarction or CHD death). ResultsAmong 23,901 CHD-free participants (57.8% White and 58.4% women, mean age 64 ± 9 years) over a median 10 years of follow-up, 664 and 951 CHD events occurred among Black and White adults, respectively. Low-density lipoprotein cholesterol and triglycerides were associated with increased risk of CHD in both races (P interaction by race >0.10). For sex-specific clinical HDL-C categories: low HDL-C was associated with increased CHD risk in White (HR: 1.22; 95% CI: 1.05-1.43) but not in Black (HR: 0.94; 95% CI: 0.78-1.14) adults (P interaction by race = 0.08); high HDL-C was not associated with decreased CHD events in either race (HR: 0.96; 95% CI: 0.79-1.16 for White participants and HR: 0.91; 95% CI: 0.74-1.12 for Black adults). ConclusionsLow-density lipoprotein cholesterol and triglycerides modestly predicted CHD risk in Black and White adults. Low HDL-C was associated with increased CHD risk in White but not Black adults, and high HDL-C was not protective in either group. Current high-density lipoprotein cholesterol–based risk calculations could lead to inaccurate risk assessment in Black adults.

Rs420137, rs386360 and rs7763726 polymorphisms in fibronectin type III domain containing 1 are associated with susceptibility to coronary heart disease: Analysis in the Han population.

BackgroundNumerous genetic studies have shown that genes are related to the pathogenesis of coronary heart disease (CHD). The main aim of this study was to confirm whether fibronectin type III domain containing 1 (FNDC1) polymorphisms correlate with the risk of CHD.MethodsIn this study, in order to assess the association between three FNDC1 single nucleotide polymorphisms (SNPs) and the risk of CHD, we conducted a case-control study involving 630 patients with CHD and 568 healthy controls using Agena MassARRAY (Agena Bioscience, San Diego, CA, USA). Genotype distribution in case and control groups was analyzed by Chi square test. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated by logistic regression models adjusted for age, sex, smoking, and alcohol consumption to assess the correlation between SNPs and CHD risk.ResultsOur results indicated that FNDC1-rs420137, -rs386360, and -rs7763726 played important roles in enhancing the risk of CHD. Subgroup analysis revealed that rs420137 increased the susceptibility to CHD in males, smokers, and patients aged ≤62 years. Rs360 had an increased risk of CHD in males, patients at aged ≤62 years, smokers, and non-drinkers. Furthermore, the association of rs7763726 with increased CHD risk was also observed in males, patients aged ≤62 years, smokers, and drinkers. Last but not least, these three SNPs we selected were protective factors against hypertension in CHD individuals.ConclusionOur research suggest that FNDC1-rs420137, -rs386360, and -rs7763726 variants may be regarded as novel biomarkers for predicting CHD risk and other specific mechanisms of action of CHD need to be further studied.

Development and Validation of a Coronary Heart Disease Risk Prediction Model in Snorers with Hypertension: A Retrospective Observed Study.

To develop and validate a risk prediction model for coronary heart disease (CHD) in snorers with hypertension, including traditional and new risk factors. Twenty factors were evaluated in the records of 2810 snorers with hypertension. Training (70%) and validation (30%) sets were created by random allocation of data, and a new nomogram model was developed. The model's discrimination and calibration were measured by calculating the area under the receiver operating curve (AUC) and creating calibration charts. The performance of the nomogram model was compared with that of the Prediction for ASCVD Risk in China (China-PAR) and Framingham models by decision curve analysis. An optimal cutoff point for the risk score in the training set was computed to stratify patients. In the nomogram model, the AUCs for predicting CHD at 5, 7 and 9 years in the training set were 0.706 (95% confidence interval [CI] 0.649-0.763), 0.703 (95% CI 0.655-0.751) and 0.669 (95% CI 0.593-0.744), respectively. The respective AUCs were 0.682 (95% CI 0.607-0.758), 0.689 (95% CI 0.618-0.760) and 0.664 (95% CI 0.539-0.789) in the validation set. The calibration chart showed that the predicted events from the nomogram score were close to the observed events. Decision curve analysis indicated that the nomogram score was slightly better than the Prediction for ASCVD Risk in China (China-PAR) and Framingham models for predicting the risk of CHD in snorers with hypertension. A cutoff point was identified for being CHD-free (a nomogram score of ≤121), which could be helpful for the early identification of individuals at high-risk of CHD. The nomogram score predicts the risk probability of CHD in snorers with hypertension at 5, 7 and 9 years, and shows good capability in terms of discrimination and calibration. It may be a useful tool for identifying individuals at high risk of CHD.

Marital status and genetic liability independently predict coronary heart disease incidence.

Married individuals have a lower coronary heart disease (CHD) risk than non-married, but the mechanisms behind this are not fully understood. We analyzed whether genetic liability to CHD may affect these associations. Marital status, a polygenic score of CHD (PGS-CHD), and other risk factors for CHD were measured from 35,444 participants (53% female) in Finnish population-based surveys conducted between 1992 and 2012. During the register-based follow-up until 2020, there were 2439 fatal and non-fatal incident CHD cases. The data were analyzed using linear and Cox regression models. Divorced and cohabiting men and women had a higher genetic risk of CHD than married individuals, but the difference was very small (0.023-0.058 standard deviation of PGS-CHD, p-values 0.011-0.429). Both marital status and PGS-CHD were associated with CHD incidence, but the associations were largely independent. Adjusting for behavioral and metabolic risk factors for CHD explained part of these associations (11-20%). No interaction was found between marital status and PGS-CHD for CHD incidence. We showed minor differences between the marital status categories in PGS-CHD and demonstrated that marital status and genetic liability predicted CHD incidence largely independently. This emphasizes the need to measure multiple risk factors when predicting CHD risk.