Machine Learning based Model Reveals the Metabolites Involved in Coronary Artery Disease

Purpose The goals of the present study were to assess the genotypic and allelic distribution of Bsm-I (rs1544410) and Apa-I (rs7975232) polymorphisms of the vitamin D receptor (VDR) gene in coronary artery disease (CAD) patients in comparison to control patients of the same age without CAD and to determine whether these gene variants are associated with dyslipidemia. Materials and Methods Based on a case-control design, 302 hospitalized patients with CAD and 194 people of comparable age without CAD were enrolled in the study. The BsmI and ApaI polymorphisms of VDR gene were studied using polymerase chain reaction followed by restriction analysis. The allele digested by the restriction enzyme was denoted by a lower letter, whereas that not digested was indicated by a capital letter. Determination of the level of vitamin D and immunoreactive insulin in the blood serum was carried out using the immuno-enzyme method. Results The bb genotype of Bsm-I VDR gene polymorphism was detected more often in patients with CAD than in the comparison group with an increased risk of CAD by 1.52 times (p=0.006, OR=1.52(1.05÷2.2). The level of HDL cholesterol was higher in CAD patients − carriers of BB genotype compared to its level in Bb genotype carriers and bb genotype carriers (1,13±0,05 mmol/l, 1,01±0,03 mmol/l, 1,02±0,03 mmol/l respectively, p<0,05). The level of vitamin D was higher in patients with BB genotype compared to its level in bb genotype carriers (45.12±3.73 nmol / l and 34.16±1.95 nmol/l respectively, p=0.008). The occurrence of a allele of Apa-I VDR gene polymorphism was higher in patients with CAD than in the control group (p=0.02, OR=1.21(0.93÷1.57). HDL cholesterol level was higher in CAD patients - AA genotype carriers compared with carriers of Aa and aa genotypes (1.18±0.08 mmol / l, 1,02±0.02 mmol / l and 1.01±0.03 mmol/l respectively, p<0,05). Immunoreactive insulin level was significantly higher in CAD patients – aa genotype carriers. No differences in LDL cholesterol and triglycerides were found. Vitamin D level was lower in CAD patients - Aa and aa genotype carriers (33,8±33,9 nmol/l ,p=0,02 and 24,7±4,9 nmol/l, p=0,05 respectively in comparison to vitamin D level = 43,3 ±4,2 nmol/l in AA genotype carriers). Conclusion The bb genotype of Bsm-I VDR gene polymorphism is associated with an increased risk of CAD. A carriage of b allele in CAD patients is associated with lower level of vitamin D and HDL cholesterol. A carriage of a allele of Apa-I VDR gene polymorphism in CAD patients is associated with lower level of vitamin D and HDL cholesterol.

UNBIASED ASSESSMENT OF CIRCULATING MICRORNAS PROVIDES AN INDEPENDENT LINK BETWEEN THE CHOLESTEROL-REGULATING MIR-382-5P AND CORONARY ARTERY DISEASE IN HUMANS

To explore the effects of allitridi capsules on endothelial function and clinical prognosis in coronary artery disease (CAD) patients with obstructive sleep apnea hypopnea syndrome (OSAHS). A total of 80 CAD patients with OSAHS were randomly assigned to receive conventional treatment (control, n = 40) and additional allitridi treatment (120 mg/day, n = 40) for 6 months. Another 40 CAD patients without OSAHS and 30 healthy individuals were chosen as controls. Endothelial function was assessed by endothelium dependent flow-mediated dilation (FMD) with high-definition color Doppler ultrasound. Serum nitric oxide (NO) and plasma endothelin-1 (ET-1) levels were determined by enzyme-linked immunosorbent assay (ELISA) and radioimmunoassay. The duration of follow-up was 1 year. The baseline clinical characteristics were not different between control and allitridi groups (P > 0.05). Compared with CAD patients without OSAHS, FMD and serum NO level were significantly lower ((7.9 ± 1.5)% vs (11.2 ± 2.9)%, P = 0.011 and (71.11 ± 10.62) vs (86.28 ± 11.03) µmol/L, P = 0.007), plasma ET-1 level was markedly higher ((112.34 ± 17.22) vs (89.87 ± 11.56) ng/L, P = 0.025) in CAD patients with OSAHS. At Month 6 post-treatment, FMD and serum NO level were significantly higher ((12.1 ± 3.1)% vs (9.1 ± 1.6)%, P = 0.020 and (105.24 ± 17.01) vs (82.39 ± 11.12) µmol/L, P = 0.001) and plasma ET-1 level in the allitridi group was lower ((77.12 ± 9.65) vs (97.77 ± 11.04) ng/L, P = 0.001) than that in the control group. At Month 12 post-treatment, the incidence of MACE was lower in the allitridi group than that in the control group (8.3% vs 15.8%, P = 0.016). Allitridi capsules significantly improved endothelial function in CAD patients with OSAHS.

The aim of the study is to assess the effects of celecoxib and sulfasalazine on the risk of coronary artery disease (CAD) in patients with ankylosing spondylitis (AS).Using the claims data of Taiwan National Health Insurance (NHI) database, a nationally representative data that contain the medical records of 23 million Taiwan residents, we randomly selected 1 million cohort from the database, and then we enrolled only patients who were newly diagnosed with AS (n = 4829) between year 2001 and 2010, excluding patients who had CAD (ICD-9- CM codes: 410–414) before the diagnosis of AS (n = 4112). According to propensity score matched 1:2 on age, gender, AS duration, Charlson comorbidity index, hypertension, and hyperlipidemia, 236 and 472 patients were included in the case (AS with CAD) and control (AS without CAD) groups, respectively. We used the WHO defined daily dose (DDD) as a tool to assess the dosage of sulfasalazine and celecoxib exposure. Conditional logistic regression was used to estimate the crude and adjusted odds ratios (ORs) and 95% confidence interval (CI) for the risk of CAD associated with use of sulfasalazine and celecoxib.Among 4112 AS patients, 8.4% (346/4112) developed CAD. CAD in AS patients were positively associated with age of 35 to 65, Charlson comorbidities index (CCI), hypertension, and hyperlipidemia. There was no gender difference between case and control groups. After adjustment for age, gender, CCI, hypertension, and hyperlipidemia, sulfasalazine users with an average daily dose ≥ 0.5 DDD (0.5 gm/day) had negative association with CAD events as compared to sulfasalazine nonusers (OR 0.63; 95% CI, 0.40–0.99, P < 0.05). NSAIDs, including celecoxib, etoricoxib, but no naproxen and diclofenac were negatively associated with CAD. Celecoxib users, with an average daily dose > 1.5 DDD, were negatively associated with CAD events, compared to celecoxib nonusers (OR 0.34; 95% CI, 0.13–0.89; P < 0.05).In this 10-year population-based case-control study, 8.4% of AS patients developed CAD. Sulfasalazine usage at an average dose of ≥ 0.5 gm/day demonstrated negative association with CAD events in patients with AS.

Hypercholesterolaemia (HC) impairs arteriogenesis, i.e. collateral artery growth. Monocytes are crucial mediators of arteriogenesis. We investigated the impact of the cardiovascular risk factor HC on ligand-induced monocyte chemotaxis. The migratory response of monocytes towards the arteriogenic ligands vascular endothelial growth factor-A (VEGF-A) and monocyte chemoattractant protein-1 (MCP-1) in hypercholesterolaemic coronary artery disease (CAD) patients (n = 14), hypercholesterolaemic controls (n = 8) and age-matched healthy controls (n = 19) was analysed. Furthermore, the serum VEGF-A level was determined in all individuals. VEGF-A-induced monocyte chemotaxis was severely impaired in hypercholesterolaemic CAD patients when compared with age-matched healthy controls (P < 0.001). The same was true for the migratory response towards MCP-1 (P < 0.001). VEGF-A- and MCP-1-induced monocyte chemotaxis of hypercholesterolaemic controls was also decreased in comparison with the healthy control group, but not as severe as observed in the hypercholesterolaemic CAD patients. VEGF-A serum levels did not differ between the three study groups. Hypercholesterolaemia severely impairs monocyte function in hypercholesterolaemic CAD patients. Monocyte dysfunction is probably connected to impaired collateral artery growth. The duration of the cardiovascular risk factor HC seems to influence the extent of monocyte dysfunction, as there exists a continuum of diminished monocyte chemotaxis in the three study groups. Further trials are warranted in order to determine whether statins can reverse the negative influence of HC on cell function.

Background: Coronary artery disease (CAD) is the most common type of cardiovascular disease (CVD), it causes narrowing of the blood vessel lumen due to atherosclerotic plaque formation. Atherosclerosis, is strongly influenced by hypercholesterolemia (HC). Proteomics offers advantages over traditional studies that focus on individual biomarkers. This method provides a holistic view by examining the entire proteomic spectrum in a biological sample, making it an indispensable asset for spotting potential novel protein biomarkers directly linked to atherosclerosis occurrence, progression, and complications such as plaque instability or rupture. Methods: Slow Off-rate Modified Aptamer (SOMAmer)-based protein array was used to quantify proteins. Our cross-sectional study involved healthy controls (n=45), and patients diagnosed with HC (n=51) or CAD (n=32). Proteome data was analyzed using multiple statistical methods. Results: The orthogonal partial least square discriminant analysis (OPLS-DA) revealed a clear separation of the 3 study groups based on two principal components, with R2X= 0.267, R2Y= 0.9, and Q2= 0.28, indicating distinct protein alterations between groups. A total of 1,305 proteins were analyzed and 156 of them in CAD vs control, 140 of them in HC vs control, and 326 of the proteins in CAD vs HC comparisons, were significantly differentially expressed with a p-value&lt;0.05. Among those, q-value&lt;0.05 were determined. As a result, in CAD patients, 2 proteins exhibited significantly differential expressions compared to healthy controls. In contrast, HC patients had 15 significantly altered proteins compared to controls. Furthermore, a comparison between CAD and HC patients revealed 87 significantly regulated proteins. The receiver operating characteristic (ROC) was performed on proteins with distinct expressions ( q&lt;0.05) in pairwise group comparisons. The AUC was calculated to identify potential biomarker candidates for the disease groups. In the CAD vs. control analysis, both significant proteins (PCSK9, SDF-1) exhibited an AUC of 0.75. In the comparison between HC and control groups, out of 15 significantly altered proteins, 7 had an AUC of 0.75 or higher (LRP1B, Apo E, QORL1, Apo E3, Apo B, Transferrin, MMP-3). Among the 87 proteins significantly differentially expressed in CAD versus HC, 59 had an AUC ≥ 0.75. Sixteen of these proteins had an AUC ≥ 0.80. Conclusion: Distinct protein patterns associated with each condition were identified. CAD patients showed a significant increase in the cholesterol-metabolizing protein PCSK9 and varying levels of the angiogenesis-related protein SDF-1. Several other proteins (LRP1B, VAV, CSRP3, HSP 60, and TAK1-TAB1) also exhibited unique expressions in disease groups, suggesting potential roles in atherosclerosis development. Moreover, a statistically significant correlation between SDF-1, ApoB levels, and LDL levels was found in CAD patients but not in healthy controls. This discovery phase cohort study unveils potential biomarkers for atherosclerosis within the HC context. This is the full abstract presented at the American Physiology Summit 2024 meeting and is only available in HTML format. There are no additional versions or additional content available for this abstract. Physiology was not involved in the peer review process.

MotivationSelecting the optimal machine learning (ML) model for a given dataset is often challenging. Automated ML (AutoML) has emerged as a powerful tool for enabling the automatic selection of ML methods and parameter settings for the prediction of biomedical endpoints. Here, we apply the tree-based pipeline optimization tool (TPOT) to predict angiographic diagnoses of coronary artery disease (CAD). With TPOT, ML models are represented as expression trees and optimal pipelines discovered using a stochastic search method called genetic programing. We provide some guidelines for TPOT-based ML pipeline selection and optimization-based on various clinical phenotypes and high-throughput metabolic profiles in the Angiography and Genes Study (ANGES).ResultsWe analyzed nuclear magnetic resonance-derived lipoprotein and metabolite profiles in the ANGES cohort with a goal to identify the role of non-obstructive CAD patients in CAD diagnostics. We performed a comparative analysis of TPOT-generated ML pipelines with selected ML classifiers, optimized with a grid search approach, applied to two phenotypic CAD profiles. As a result, TPOT-generated ML pipelines that outperformed grid search optimized models across multiple performance metrics including balanced accuracy and area under the precision-recall curve. With the selected models, we demonstrated that the phenotypic profile that distinguishes non-obstructive CAD patients from no CAD patients is associated with higher precision, suggesting a discrepancy in the underlying processes between these phenotypes.Availability and implementationTPOT is freely available via http://epistasislab.github.io/tpot/.Supplementary information Supplementary data are available at Bioinformatics online.

CAD increases the long noncoding RNA PUNISHER in small extracellular vesicles and regulates endothelial cell function via vesicular shuttling

Non-Invasive Detection of Coronary Artery Disease in Patients With Left Bundle Branch Block Using 64-Slice Computed Tomography

Previous studies assessing effect of ischemia on ventricular repolarization are mostly directed toward patients with coronary artery disease (CAD); however, similar reports on cardiac syndrome X (CSX) are scarce. Whether microvascular dysfunction of CSX and ischemia induced by CAD produce comparable effect on ventricular repolarization is unclear and deserve further studies. In the present study, ECG measures of ventricular repolarization were compared between CAD and CSX patients (40 subjects in each group). Following evaluation of sociodemographic characteristics, medical and past medical history, a resting ECG was used to assess measurements of ventricular repolarization in each patient, namely, QT interval (QT), corrected QT interval (QTc), QT dispersion (QTd), corrected QT dispersion (QTcd), adjacent QT dispersion (AdQTd), QT dispersion ratio (QTdR), JT dispersion (JTd), and Corrected JT dispersion (JTcd). Results showed comparable QT intervals and QTd in CAD and CSX patients even after adjustment for the possible variations in gender, age and body mass index of the studied groups. Although JTd was increased in CSX subjects (26.6 ± 7.2 ms) compared with CAD patients (22.7 ± 6.5 ms, p = 0.019), statistical significance disappeared after correcting JT for variations in heart rate. QT and QTc were significantly below 440 ms in CAD as well as CSX patients (p < 0.001). In contrast, maximum QTd, maximum QTcd and AdQTd of CAD and CSX patients were significantly above 440 ms (p < 0.001). The means of JTd and JTcd were significantly above 22 ms and 24 ms respectively (p < 0.001, p = 0.001) in CSX but not CAD patients (p = 0.529, p = 0.281). The present findings clearly demonstrate comparable measures of ventricular repolarization in CAD and CSX patients and consequently an equal risk of cardiac events in both groups.

The prevalence of type 2 diabetes (T2DM) is higher in peripheral artery disease (PAD) than in coronary artery disease (CAD) patients, and PAD overall confers higher cardiovascular risk than CAD. How the incidence of major cardiovascular events compares between PAD and CAD patients when analyses are stratified by the presence of type 2 diabetes (T2DM) is unclear and is addressed in the present study. We prospectively recorded major cardiovascular events and death over 10.0±4.7 years in 923 patients with stable CAD, of whom 26.7% had T2DM and in 292 patients with PAD, of whom 42.1% had T2DM. Four groups were analyzed: CAD patients without diabetes (CAD/T2DM-; n=677), CAD patients with T2DM (CAD/T2DM+; n=246), PAD patients without diabetes (PAD/T2DM-; n=169) and PAD patients with T2DM (PAD/T2DM+; n=123). When compared to the incidence of MACE in CAD+/T2DM- patients (25.1%), it was significantly higher in CAD+/T2DM+ patients (35.4%; p&amp;lt;0.001), in PAD+/T2DM- patients (30.2%; p=0.022) and in PAD+/T2DM+ patients (47.2%; p&amp;lt;0.001). Patients with both PAD and T2DM in turn were at a higher risk than CAD+/T2DM+ or PAD+/T2DM- patients (p=0.001 and p&amp;lt;0.001, respectively). The incidence of MACE did not differ significantly between PAD+/T2DM- and CAD+/T2DM+ patients (p=0.413). Compared to patients with CAD, Cox regression analyses after multivariate adjustment showed an adjusted hazard ratio of 1.46 [1.14-1.87], p=0.002 for the presence of PAD. Conversely, T2DM increased the risk of MACE after multivariate adjustment in CAD and PAD patients (adjusted HR 1.58 [1.27-1.98], p&amp;lt;0.001). In conclusion, our data show that T2DM and the presence of PAD are mutually independent predictors of MACE. Patients with both PAD and T2DM are at an exceedingly high risk of MACE. Disclosure C. H. Saely: None. H. Drexel: None. A. Vonbank: None. B. Larcher: None. A. Mader: None. M. Maechler: None. L. Sprenger: None. B. Mutschlechner: None. A. Leiherer: None. A. Muendlein: None.

Introduction: The prevalence of type 2 diabetes (T2DM) is higher in peripheral artery disease (PAD) than in coronary artery disease (CAD) patients, and PAD overall confers higher cardiovascular risk than CAD. Hypothesis: We hypothesize that the incidence of major cardiovascular events compares between PAD and CAD patients when analyses are stratified by the presence of type 2 diabetes (T2DM). Methods: We prospectively recorded major cardiovascular events and death over 10.0±4.7 years in 923 patients with stable CAD, of whom 26.7% had T2DM and in 292 patients with PAD, of whom 42.1% had T2DM. Four groups were analyzed: CAD patients without diabetes (CAD/T2DM-; n=677), CAD patients with T2DM (CAD/T2DM+; n=246), PAD patients without diabetes (PAD/T2DM-; n=169) and PAD patients with T2DM (PAD/T2DM+; n=123). Results: When compared to the incidence of MACE in CAD+/T2DM- patients (25.1%), it was significantly higher in CAD+/T2DM+ patients (35.4%; p&lt;0.001), in PAD+/T2DM- patients (30.2%; p=0.022) and in PAD+/T2DM+ patients (47.2%; p&lt;0.001). Patients with both PAD and T2DM in turn were at a higher risk than CAD+/T2DM+ or PAD+/T2DM- patients (p=0.001 and p&lt;0.001, respectively). The incidence of MACE did not differ significantly between PAD+/T2DM- and CAD+/T2DM+ patients (p=0.413). Compared to patients with CAD, Cox regression analyses after multivariate adjustment showed an adjusted hazard ratio of 1.46 [1.14-1.87], p=0.002 for the presence of PAD. Conversely, T2DM increased the risk of MACE after multivariate adjustment in CAD and PAD patients (adjusted HR 1.58 [1.27-1.98], p&lt;0.001). Conclusions: In conclusion, our data show that T2DM and the presence of PAD are mutually independent predictors of MACE. Patients with both PAD and T2DM are at an exceedingly high risk of MACE.

Pre-test probability for coronary artery disease in patients with chest pain based on machine learning techniques

Background A correct and prompt diagnosis of coronary artery disease (CAD) is a crucial component of disease management to reduce the risk of death and improve the quality of life in patients with CAD. Currently, the American College of Cardiology (ACC)/American Heart Association (AHA) and the European Society of Cardiology (ESC) guidelines recommend selecting an appropriate pre-diagnosis test for an individual patient according to the CAD probability. The purpose of this study was to develop a practical pre-test probability (PTP) for obstructive CAD in patients with chest pain using machine learning (ML); also, the performance of ML-PTP for CAD is compared to the final result of coronary angiography (CAG). Methods We used a database from a single-center, prospective, all-comer registry designed to reflect real-world practice since 2004. All subjects underwent invasive CAG at our hospital in South Korea. We used logistic regression algorithms, random forest (RF), supporting vector machine, and K-nearest neighbor classification for the ML models. The dataset was divided into two consecutive sets according to the registration period to validate the ML models. ML training for PTP and internal validation used the first dataset registered between 2004 and 2012 (8,631 patients). The second dataset registered between 2013 and 2014 (1,546 patients) was used for external validation. The primary endpoint was obstructive CAD. Obstructive CAD was defined as having a stenosis diameter of greater than 70% on the quantitative CAG of the main epicardial coronary artery. Results We derived an ML-based model consisting of three different models according to the subject used to obtain the information, such as the patient himself (dataset 1), the community's first medical center (dataset 2), and doctors (dataset 3). The performance range of the ML-PTP models as the non-invasive test had C-statistics of 0.795 to 0.984 compared to the result of invasive testing via CAG in patients with chest pain. The training ML-PTP models were adjusted to have 99% sensitivity for CAD so as not to miss actual CAD patients. In the testing dataset, the best accuracy of the ML-PTP model was 45.7% using dataset 1, 47.2% using dataset 2, and 92.8% using dataset 3 and the RF algorithm. The CAD prediction sensitivity was 99.0%, 99.0%, and 98.0%, respectively. Conclusion We successfully developed a high-performance model of ML-PTP for CAD which is expected to reduce the need for non-invasive tests in chest pain. However, since this PTP model is derived from data of a single medical center, multicenter verification is required to use it as a PTP recommended by the major American societies and the ESC.

Utilizing machine learning in echocardiographic analysis to distinguish obstructive and non-obstructive coronary artery disease.