Clinical and biochemical predictors of longitudinal changes in left atrial structure and function: A general population study.

PURPOSE: There is a need for better understanding the factors that modulate left atrial (LA) dysfunction. Therefore, we determined associations of clinical and biochemical biomarkers with serial changes in echocardiographic indexes of LA function in the general population. METHODS: We measured LA maximal and minimal volume indexes (LAVImax and LAVImin) by echocardiography and LA reservoir strain (LARS) by two-dimensional speckle-tracking in 627 participants (mean age 50.8 years, 51.2% women) at baseline and after 4.8 years. RESULTS: During follow-up, LARS decreased significantly in men (-.90%, P = .033) but not in women (-.23%, P = .60). In stepwise regression analysis, stronger decrease in LARS over time was associated with male sex, a higher age, body mass index (BMI), mean arterial pressure (MAP) and serum insulin at baseline and with a greater increase in BMI and MAP over time (P ≤ .018). Similarly, an increased risk of developing or retaining abnormal LARS was observed in older participants, in subjects with a higher baseline BMI, MAP, heart rate (HR), troponin T and ΔMAP, and in those who used ß-blockers at baseline. Both LAVImax and LAVImin increased significantly over time (P ≤ .0007). This increase was associated with a higher baseline age, pulse pressure and a lower HR at baseline and a greater increase in pulse pressure over time (P ≤ .029). Higher serum insulin and D-dimer were independently associated with a stronger increase in LAVImin (P ≤ .0034). CONCLUSION: Subclinical worsening in LA dysfunction was associated with older age, hypertension, obesity, insulin resistance and troponin T levels. Cardiovascular risk management strategies may delay LA deterioration.

Feature-based clustering of the left ventricular strain curve for cardiovascular risk stratification in the general population.

Objective: Identifying individuals with subclinical cardiovascular (CV) disease could improve monitoring and risk stratification. While peak left ventricular (LV) systolic strain has emerged as a strong prognostic factor, few studies have analyzed the whole temporal profiles of the deformation curves during the complete cardiac cycle. Therefore, in this longitudinal study, we applied an unsupervised machine learning approach based on time-series-derived features from the LV strain curve to identify distinct strain phenogroups that might be related to the risk of adverse cardiovascular events in the general population. Method: We prospectively studied 1,185 community-dwelling individuals (mean age, 53.2 years; 51.3% women), in whom we acquired clinical and echocardiographic data including LV strain traces at baseline and collected adverse events on average 9.1 years later. A Gaussian Mixture Model (GMM) was applied to features derived from LV strain curves, including the slopes during systole, early and late diastole, peak strain, and the duration and height of diastasis. We evaluated the performance of the model using the clinical characteristics of the participants and the incidence of adverse events in the training dataset. To ascertain the validity of the trained model, we used an additional community-based cohort (n = 545) as external validation cohort. Results: The most appropriate number of clusters to separate the LV strain curves was four. In clusters 1 and 2, we observed differences in age and heart rate distributions, but they had similarly low prevalence of CV risk factors. Cluster 4 had the worst combination of CV risk factors, and a higher prevalence of LV hypertrophy and diastolic dysfunction than in other clusters. In cluster 3, the reported values were in between those of strain clusters 2 and 4. Adjusting for traditional covariables, we observed that clusters 3 and 4 had a significantly higher risk for CV (28% and 20%, P ≤ 0.038) and cardiac (57% and 43%, P ≤ 0.024) adverse events. Using SHAP values we observed that the features that incorporate temporal information, such as the slope during systole and early diastole, had a higher impact on the model's decision than peak LV systolic strain. Conclusion: Employing a GMM on features derived from the raw LV strain curves, we extracted clinically significant phenogroups which could provide additive prognostic information over the peak LV strain.

Serum proteomic profiling of carotid arteriopathy: A population outcome study.

BACKGROUND AND AIMS: Circulating proteins reflecting subclinical vascular disease may improve prediction of atherosclerotic cardiovascular disease (ASCVD). We applied feature selection and unsupervised clustering on proteomic data to identify proteins associated with carotid arteriopathy and construct a protein-based classifier for ASCVD event prediction. METHODS: 491 community-dwelling participants (mean age, 58 ± 11 years; 51 % women) underwent carotid ultrasonography and proteomic profiling (CVD II panel, Olink Proteomics). ASCVD outcome was collected (median follow-up time: 10.2 years). We applied partial least squares (PLS) to identify proteins linked to carotid intima-media thickness (cIMT). Next, we assessed the association between future ASCVD events and protein-based phenogroups derived by unsupervised clustering (Gaussian Mixture modelling) based on proteins selected in PLS. RESULTS: PLS identified 19 proteins as important, which were all associated with cIMT in multivariable-adjusted linear regression. 8 of the 19 proteins were excluded from the clustering analysis because of high collinearity. Based on the 11 remaining proteins, the clustering algorithm subdivided the cohort into two phenogroups. Compared to the first phenogroup (n = 177), participants in the second phenogroup (n = 314) presented: i) a more unfavorable lipid profile with higher total cholesterol and triglycerides and lower HDL cholesterol (p ≤ 0.014); ii) higher cIMT (p = 0.0020); and iii) a significantly higher risk for future ASCVD events (multivariable-adjusted hazard ratio (95 % CI) versus phenogroup 1: 2.05 (1.26-3.52); p = 0.0093). The protein-based phenogrouping supplemented ACC/AHA 10-year ASCVD risk scoring for prediction of a first ASCVD event. CONCLUSIONS: Focused protein-based phenogrouping identified individuals at high risk for future ASCVD and may complement current risk stratification strategies.

Cardiorespiratory fitness components in relation to clinical characteristics, disease state and medication intake: A patient registry study.

BACKGROUND: Interpretation of cardiopulmonary exercise testing (CPET) results requires thorough understanding of test confounders such as anthropometrics, comorbidities and medication. Here, we comprehensively assessed the clinical determinants of cardiorespiratory fitness and its components in a heterogeneous patient sample. METHODS: We retrospectively collected medical and CPET data from 2320 patients (48.2% females) referred for cycle ergometry at the University Hospital Leuven, Belgium. We assessed clinical determinants of peak CPET indexes of cardiorespiratory fitness (CRF) and its hemodynamic and ventilatory components using stepwise regression and quantified multivariable-adjusted differences in indexes between cases and references. RESULTS: Lower peak load and peak O2 uptake were related to: higher age, female sex, lower body height and weight, and higher heart rate; to the intake of beta blockers, analgesics, thyroid hormone replacement and benzodiazepines; and to diabetes mellitus, chronic kidney disease, non-ST elevation myocardial infarction and atrial fibrillation (p < 0.05 for all). Lower peak load also correlated with obstructive pulmonary diseases. Stepwise regression revealed associations of hemodynamic and ventilatory indexes (including heart rate, O2 pulse, systolic blood pressure and ventilation at peak exercise and ventilatory efficiency) with age, sex, body composition and aforementioned diseases and medications. Multivariable-adjusted differences in CPET metrics between cases and controls confirmed the associations observed. CONCLUSION: We described known and novel associations of CRF components with demographics, anthropometrics, cardiometabolic and pulmonary diseases and medication intake in a large patient sample. The clinical implications of long-term noncardiovascular drug intake for CPET results require further investigation.

Integrative Interpretation of Cardiopulmonary Exercise Tests for Cardiovascular Outcome Prediction: A Machine Learning Approach.

Integrative interpretation of cardiopulmonary exercise tests (CPETs) may improve assessment of cardiovascular (CV) risk. Here, we identified patient phenogroups based on CPET summary metrics and evaluated their predictive value for CV events. We included 2280 patients with diverse CV risk who underwent maximal CPET by cycle ergometry. Key CPET indices and information on incident CV events (median follow-up time: 5.3 years) were derived. Next, we applied unsupervised clustering by Gaussian Mixture modeling to subdivide the cohort into four male and four female phenogroups solely based on differences in CPET metrics. Ten of 18 CPET metrics were used for clustering as eight were removed due to high collinearity. In males and females, the phenogroups differed significantly in age, BMI, blood pressure, disease prevalence, medication intake and spirometry. In males, phenogroups 3 and 4 presented a significantly higher risk for incident CV events than phenogroup 1 (multivariable-adjusted hazard ratio: 1.51 and 2.19; p ≤ 0.048). In females, differences in the risk for future CV events between the phenogroups were not significant after adjustment for clinical covariables. Integrative CPET-based phenogrouping, thus, adequately stratified male patients according to CV risk. CPET phenomapping may facilitate comprehensive evaluation of CPET results and steer CV risk stratification and management.

Unsupervised Time-Series Clustering of Left Atrial Strain for Cardiovascular Risk Assessment.

BACKGROUND: Early identification of individuals at high risk for developing cardiovascular (CV) events is of paramount importance for efficient risk management. Here, the authors investigated whether using unsupervised machine learning methods on time-series data of left atrial (LA) strain could distinguish clinically meaningful phenogroups associated with the risk for developing adverse events. METHODS: In 929 community-dwelling individuals (mean age, 51.6 years; 52.9% women), clinical and echocardiographic data were acquired, including LA strain traces, at baseline, and cardiac events were collected on average 6.3 years later. Two unsupervised learning techniques were used: (1) an ensemble of a deep convolutional neural network autoencoder with k-medoids and (2) a self-organizing map to cluster spatiotemporal patterns within LA strain curves. Clinical characteristics and cardiac outcome were used to evaluate the validity of the k clusters using the original cohort, while an external population cohort (n = 378) was used to validate the trained models. RESULTS: In both approaches, the optimal number of clusters was five. The first three clusters had differences in sex distribution and heart rate but had a similar low CV risk profile. On the other hand, cluster 5 had the worst CV profile and a higher prevalence of left ventricular remodeling and diastolic dysfunction compared with the other clusters. The respective indexes of cluster 4 were between those of clusters 1 to 3 and 5. After adjustment for traditional risk factors, cluster 5 had the highest risk for cardiac events compared with clusters 1, 2, and 3 (hazard ratio, 1.36; 95% CI, 1.09-1.70; P = .0063). Similar LA strain patterns were obtained when the models were applied to the external validation cohort, and clinical characteristics revealed similar CV risk profiles across all clusters. CONCLUSION: Unsupervised machine learning algorithms used in time-series LA strain curves identified clinically meaningful clusters of LA deformation and provide incremental prognostic information over traditional risk factors.

Improving predictive performance in incident heart failure using machine learning and multi-center data.

Objective: To mitigate the burden associated with heart failure (HF), primary prevention is of the utmost importance. To improve early risk stratification, advanced computational methods such as machine learning (ML) capturing complex individual patterns in large data might be necessary. Therefore, we compared the predictive performance of incident HF risk models in terms of (a) flexible ML models and linear models and (b) models trained on a single cohort (single-center) and on multiple heterogeneous cohorts (multi-center). Design and methods: In our analysis, we used the meta-data consisting of 30,354 individuals from 6 cohorts. During a median follow-up of 5.40 years, 1,068 individuals experienced a non-fatal HF event. We evaluated the predictive performance of survival gradient boosting (SGB), CoxNet, the PCP-HF risk score, and a stacking method. Predictions were obtained iteratively, in each iteration one cohort serving as an external test set and either one or all remaining cohorts as a training set (single- or multi-center, respectively). Results: Overall, multi-center models systematically outperformed single-center models. Further, c-index in the pooled population was higher in SGB (0.735) than in CoxNet (0.694). In the precision-recall (PR) analysis for predicting 10-year HF risk, the stacking method, combining the SGB, CoxNet, Gaussian mixture and PCP-HF models, outperformed other models with PR/AUC 0.804, while PCP-HF achieved only 0.551. Conclusion: With a greater number and variety of training cohorts, the model learns a wider range of specific individual health characteristics. Flexible ML algorithms can be used to capture these diverse distributions and produce more precise prediction models.