The impact of diastolic dysfunction as a predictor of cardiovascular events: A systematic review and meta-analysis

Exploring machine learning strategies for predicting cardiovascular disease risk factors from multi-omic data

BACKGROUND

Machine learning (ML) classifiers are increasingly used for predicting cardiovascular disease (CVD) and related risk factors using omics data, although these outcomes often exhibit categorical nature and class imbalances. However, little is known about which ML classifier, omics data, or upstream dimension reduction strategy has the strongest influence on prediction quality in such settings. Our study aimed to illustrate and compare different machine learning strategies to predict CVD risk factors under different scenarios.

METHODS

We compared the use of six ML classifiers in predicting CVD risk factors using blood-derived metabolomics, epigenetics and transcriptomics data. Upstream omic dimension reduction was performed using either unsupervised or semi-supervised autoencoders, whose downstream ML classifier performance we compared. CVD risk factors included systolic and diastolic blood pressure measurements and ultrasound-based biomarkers of left ventricular diastolic dysfunction (LVDD; E/e' ratio, E/A ratio, LAVI) collected from 1,249 Finnish participants, of which 80% were used for model fitting. We predicted individuals with low, high or average levels of CVD risk factors, the latter class being the most common. We constructed multi-omic predictions using a meta-learner that weighted single-omic predictions. Model performance comparisons were based on the F1 score. Finally, we investigated whether learned omic representations from pre-trained semi-supervised autoencoders could improve outcome prediction in an external cohort using transfer learning.

RESULTS

Depending on the ML classifier or omic used, the quality of single-omic predictions varied. Multi-omics predictions outperformed single-omics predictions in most cases, particularly in the prediction of individuals with high or low CVD risk factor levels. Semi-supervised autoencoders improved downstream predictions compared to the use of unsupervised autoencoders. In addition, median gains in Area Under the Curve by transfer learning compared to modelling from scratch ranged from 0.09 to 0.14 and 0.07 to 0.11 units for transcriptomic and metabolomic data, respectively.

CONCLUSIONS

By illustrating the use of different machine learning strategies in different scenarios, our study provides a platform for researchers to evaluate how the choice of omics, ML classifiers, and dimension reduction can influence the quality of CVD risk factor predictions.

Analysis and recognition of post-exercise cardiac state based on heart sound features and cardiac troponin I

PURPOSE

Excessive intensity exercises can bring irreversible damage to the heart. We explore whether heart sounds can evaluate cardiac function after high-intensity exercise and hope to prevent overtraining through the changes of heart sound in future training.

METHODS

The study population consisted of 25 male athletes and 24 female athletes. All subjects were healthy and had no history of cardiovascular disease or family history of cardiovascular disease. The subjects were required to do high-intensity exercise for 3 days, with their blood sample and heart sound (HS) signals being collected and analysed before and after exercise. We then developed a Kernel extreme learning machine (KELM) model that can distinguish the state of heart by using the pre- and post-exercise data.

RESULTS

There was no significant change in serum cardiac troponin I after 3 days of load cross-country running, which indicates that there was no myocardial injury after the race. The statistical analysis of time-domain characteristics and multi-fractal characteristic parameters of HS showed that the cardiac reserve capacity of the subjects was enhanced after the cross-country running, and the KELM is an effective classifier to recognize HS and the state of the heart after exercise.

CONCLUSION

Through the results, we can draw the conclusion that this intensity of exercise will not cause profound damage to the athlete's heart. The findings of this study are of great significance for evaluating the condition of the heart with the proposed index of heart sound and prevention of excessive training that causes damage to the heart.

Association of Echocardiographic Parameter E/e' With Cardiovascular Events in a Diverse Population of Inpatients and Outpatients With and Without Cardiac Diseases and Risk Factors

BACKGROUND

The echocardiographic parameter E/e' has been associated with cardiovascular (CV) events. However, few studies have analyzed multiple associated CV outcomes using E/e' in a diverse population of both inpatients and outpatients with and without cardiac diseases and risk factors.

METHODS

Medical records of 75,393 patients without atrial fibrillation (AF) with first available E/e' were retrieved from our hospital database. Patients with mitral valve disease were excluded, and the remainder were studied in protocol 1 (70,819 patients). Patients with hypertension, diabetes mellitus, hyperlipidemia, CV diseases, prior CV events, CV surgeries, and left ventricular ejection fraction <50% or missing left ventricular ejection fraction were further excluded, and the remaining patients were studied in protocol 2 (14,665 patients). The study outcomes are major adverse CV events (MACE), which included myocardial infarction (MI), AF, ischemic and hemorrhagic stroke (IHS), hospitalization for heart failure (HHF), and cardiac death. The primary outcomes were MACE and each of the MACE components.

RESULTS

At the end of maximal 5-year follow-up (median 22.18 months with interquartile range 7.20-49.08 months for MACE in protocol 1 and 23.46 months with interquartile range 8.15-49.02 months for MACE in protocol 2), compared with an E/e' value of <8, an intermediate value of E/e' 8 to 15 and a high value of E/e' >15 were significantly associated with MACE, MI, AF, IHS, HHF, and cardiac death in protocol 1 (all P < .0001). In protocol 2, an intermediate E/e' value of 8 to 15 and a high value of E/e' >15 were significantly associated with MACE, MI, AF, IHS, HHF, and CV death (all P < .05), except an intermediate value E/e' 8 to 15 was not associated with AF.

CONCLUSIONS

In a diverse population of inpatients and outpatients with and without cardiac diseases and risk factors, the echocardiographic parameter E/e' was associated with CV events and is a useful marker of risk.

Left atrial strain evaluation to assess left ventricle diastolic dysfunction and heart failure with preserved ejection fraction: a guide to clinical practice : Left atrial strain and diastolic function

Although frequently unrecognized, left atrium (LA) function plays a key role in global cardiac performance as it modulates left ventricle (LV) filling through three main functions: the LA reservoir, conduit and booster pump. Given the interdependence between the LA and the LV, it is known that LA dimension and/or LA function can be used as a surrogate for LV diastolic function.Recently, LA deformation analysis using 2-dimensional (2D) speckle tracking echocardiography (STE) strain emerged as a promising non-invasive technique that can help to estimate LV filling pressures (LVFP) and diagnose heart failure with preserved ejection fraction (HFpEF), as recently proposed in the recent 2022 European Association of Cardiovascular Imaging (EACVI) consensus which included LA reservoir strain as an additive parameter to estimate LVFP.This article aims to review the latest evidence regarding the role of LA strain in the assessment and management of LV diastolic dysfunction (LVDD) and HFpEF, providing a clinical guide with tips and tricks to use LA strain as a new technique to help to estimate LVFP.

Effect of polygenic risk for schizophrenia on cardiac structure and function: a UK Biobank observational study

BACKGROUND

Cardiovascular disease is a major cause of excess mortality in people with schizophrenia. Several factors are responsible, including lifestyle and metabolic effects of antipsychotics. However, variations in cardiac structure and function are seen in people with schizophrenia in the absence of cardiovascular disease risk factors and after accounting for lifestyle and medication. Therefore, we aimed to explore whether shared genetic causes contribute to these cardiac variations.

METHODS

For this observational study, we used data from the UK Biobank and included White British or Irish individuals without diagnosed schizophrenia with variable polygenic risk scores for the condition. To test the association between polygenic risk score for schizophrenia and cardiac phenotype, we used principal component analysis and regression. Robust regression was then used to explore the association between the polygenic risk score for schizophrenia and individual cardiac phenotypes. We repeated analyses with fibro-inflammatory pathway-specific polygenic risk scores for schizophrenia. Last, we investigated genome-wide sharing of common variants between schizophrenia and cardiac phenotypes using linkage disequilibrium score regression. The primary outcome was principal component regression.

FINDINGS

Of 33 353 individuals recruited, 32 279 participants had complete cardiac MRI data and were included in the analysis, of whom 16 625 (51·5%) were female and 15 654 (48·5%) were male. 1074 participants were excluded on the basis of incomplete cardiac MRI data (for all phenotypes). A model regressing polygenic risk scores for schizophrenia onto the first five cardiac principal components of the principal components analysis was significant (F=5·09; p=0·00012). Principal component 1 captured a pattern of increased cardiac volumes, increased absolute peak diastolic strain rates, and reduced ejection fractions; polygenic risk scores for schizophrenia and principal component 1 were negatively associated (β=-0·01 [SE 0·003]; p=0·017). Similar to the principal component analysis results, for individual cardiac phenotypes, we observed negative associations between polygenic risk scores for schizophrenia and indexed right ventricular end-systolic volume (β=-0·14 [0·04]; p=0·0013, pFDR=0·015), indexed right ventricular end-diastolic volume (β=-0·17 [0·08]); p=0·025; pFDR=0·082), and absolute longitudinal peak diastolic strain rates (β=-0·01 [0·003]; p=0·0024, pFDR=0·015), and a positive association between polygenic risk scores for schizophrenia and right ventricular ejection fraction (β=0·09 [0·03]; p=0·0041, pFDR=0·015). Models examining the transforming growth factor-β (TGF-β)-specific and acute inflammation-specific polygenic risk scores for schizophrenia found significant associations with the first five principal components (F=2·62, p=0·022; F=2·54, p=0·026). Using linkage disequilibrium score regression, we observed genetic overlap with schizophrenia for right ventricular end-systolic volume and right ventricular ejection fraction (p=0·0090, p=0·0077).

INTERPRETATION

High polygenic risk scores for schizophrenia are associated with decreased cardiac volumes, increased ejection fractions, and decreased absolute peak diastolic strain rates. TGF-β and inflammatory pathways might be implicated, and there is evidence of genetic overlap for some cardiac phenotypes. Reduced absolute peak diastolic strain rates indicate increased myocardial stiffness and diastolic dysfunction, which increases risk of cardiac disease. Thus, genetic risk for schizophrenia is associated with cardiac structural changes that can worsen cardiac outcomes. Further work is required to determine whether these associations are specific to schizophrenia or are also seen in other psychiatric conditions.

FUNDING

National Institute for Health Research, Maudsley Charity, Wellcome Trust, Medical Research Council, Academy of Medical Sciences, Edmond J Safra Foundation, British Heart Foundation.

Postnatal cardiovascular morbidity following preterm pre-eclampsia: An observational study

OBJECTIVE

To explore the nature of postnatal cardiovascular morbidity following pregnancies complicated by preterm pre-eclampsia and investigate associations between pregnancy characteristics and maternal postnatal cardiovascular function.

STUDY DESIGN

This was an observational sub-study of a single-centre feasibility randomised double-blind placebo-controlled trial (https://www.

CLINICALTRIALS

gov; NCT03466333), involving women with preterm pre-eclampsia, delivering before 37 weeks. Eligible women underwent echocardiography, arteriography and blood pressure monitoring within three days of birth, six weeks and six months postpartum. Correlations between pregnancy and cardiovascular characteristics were assessed using Spearman's correlation.

MAIN OUTCOME MEASURES

The prevalence of cardiovascular dysfunction and remodelling six months following preterm pre-eclampsia.

RESULTS

Forty-four women completed the study. At six months, 27 (61 %) had diastolic dysfunction, 33 (75 %) had raised total vascular resistance (TVR) and 18 (41 %) had left ventricular remodelling. Sixteen (46 %) women had de novo hypertension by six months and only two (5 %) women had a completely normal echocardiogram. Echocardiography did not change significantly from six weeks to six months. Earlier gestation at delivery and lower birthweight centile were associated with worse six-month diastolic dysfunction (E/E': rho = -0.39, p = 0.001 & rho = -0.42, p = 0.005) and TVR (rho = -0.34, p = 0.02 & rho = -0.37, p = 0.01).

CONCLUSIONS

Preterm pre-eclampsia is associated with persistent cardiovascular morbidity-six months postpartum in the majority of women. These cardiovascular changes have significant implications for long-term cardiovascular health. The graded severity of diastolic dysfunction and TVR with worsening pre-eclampsia phenotype suggests a dose-effect. However, the mechanistic link remains uncertain.

Cardiac function and exercise capacity in patients with metabolic syndrome: A cross-sectional study

Background

Metabolic syndrome is a pre-diabetes condition that is associated with increased cardiovascular morbidity and mortality. We aimed to explore how exercise capacity, cardiac structure, and function were affected in patients with metabolic syndrome.

Methods

Outpatients with echocardiography and exercise stress test combined with impedance cardiography (ETT + ICGG) results available from Nov 2018 to Oct 2020 were retrospectively enrolled. Echocardiographic, ETT + ICG profiles, and exercise performance were compared between patients with metabolic syndrome and the ones without. Sensitivity analyses were performed excluding patients without established coronary heart disease and further 1:1 paired for age and gender, respectively. Multiple linear regression was used to find out related predictors for maximal metabolic equivalents (METs).

Results

Three hundred and twenty-third patients were included, among whom 97 were diagnosed as metabolic syndrome. Compared to patients without metabolic syndrome, echocardiography showed that patients with metabolic syndrome had a significantly lower E/A ratio (p &lt; 0.001). Besides, they have larger left atrium, larger right ventricle, and thicker interventricular septum (all p &lt; 0.001), but similar left ventricular ejection fraction (P = 0.443). ICG showed that patients with metabolic syndrome had significantly higher stroke volume at rest and maximum (p &lt; 0.001), higher left cardiac work index at rest and maximum (p = 0.005), higher systemic vascular resistance (SVR) at rest (p &lt; 0.001), but similar SVI (p = 0.888). During exercise, patients with metabolic syndrome had lower maximal METs (p &lt; 0.001), and a higher proportion suffering from ST-segment depression during exercise (p = 0.009). Sensitivity analyses yielded similar results. As for the linear regression model, 6 independent variables (systolic blood pressure, BMI, E/A ratio, the height of O wave, the peak value of LCWi, and the baseline of SVR) had statistically significant effects on the maximal METs tested in exercise (R = 0.525, R2 = 0.246, P &lt; 0.001).

Conclusion

Patients with metabolic syndrome had significant structural alteration, apparent overburden of left ventricular work index, pre-and afterload, which may be the main cause of impaired exercise tolerance.

Genetic and environmental determinants of diastolic heart function

Diastole is the sequence of physiological events that occur in the heart during ventricular filling and principally depends on myocardial relaxation and chamber stiffness. Abnormal diastolic function is related to many cardiovascular disease processes and is predictive of health outcomes, but its genetic architecture is largely unknown. Here, we use machine learning cardiac motion analysis to measure diastolic functional traits in 39,559 participants of the UK Biobank and perform a genome-wide association study. We identified 9 significant, independent loci near genes that are associated with maintaining sarcomeric function under biomechanical stress and genes implicated in the development of cardiomyopathy. Age, sex and diabetes were independent predictors of diastolic function and we found a causal relationship between genetically-determined ventricular stiffness and incident heart failure. Our results provide insights into the genetic and environmental factors influencing diastolic function that are relevant for identifying causal relationships and potential tractable targets.

E/E′ Is a New Independent Predictor of Recovered Ejection Fraction in Patients With Systolic Heart Failure Undergoing Ablation for Atrial Fibrillation

Aims: Catheter ablation should be considered in patients with atrial fibrillation (AF) and with heart failure (HF) with reduced ejection fraction (EF; HFrEF) to improve survival and reduce heart failure hospitalization. Careful patient selection for AF ablation is key to achieving similar outcome benefits. However, limited data exist regarding predictors of recovered ejection fraction. We aimed to evaluate the predictors of recovered ejection fraction in consecutive patients with HF undergoing AF ablation.

Methods and Results: A total of 156 patients [67.3% men, median age 63 (11)] with AF and HF underwent initial catheter ablation between September 2017 and October 2019 in the First Affiliated Hospital of Dalian Medical University. Overall, the percentage of recovered ejection fractions was 72.3%. Recovered EFs were associated with a 39% reduction in all-cause hospitalization compared to non-recovered EFs at the 1-year follow-up [23.8 vs. 62.8 (odds ratio) OR 2.09 (1.40–3.12), P < 0.001]. Univariate analysis for recovered EFs showed that diabetes (P = 0.083), prevalent HF (P = 0.014), prevalent AF (P = 0.051), LVEF (P = 0.022), and E/E′ (P = 0.001) were associated with EF improvement. Multivariate analysis showed that the only independent predictor of EF recovery was E/E′ [OR 1.13 (1.03–1.24); P = 0.011]. A receiver operating characteristic analysis determined that the suitable cut-off value for E/E′ was 15 (sensitivity 38.7%, specificity 89.2%, the area under curve 0.704).

Conclusions: Ejection fraction (EF) recovery occurred in 72.3% of patients, associated with a 39% reduction in all-cause hospitalization compared to the non-recovered EFs in our cohort. The only independent predictor of recovered EF was E/E′ < 15 in our series.

Accelerated diastolic dysfunction in premenopausal women with rheumatoid arthritis

Background

Disturbances of diastolic function precede systolic heart failure and, although clinically silent, represent the earliest sign of cardiac involvement. Diastolic dysfunction (DD) is associated with age, gender (female), and hypertension. However, little is known about the age-specific incidence rates and risk factors for DD in patients with rheumatoid arthritis (RA).

Methods

We used standard two-dimensional/Doppler echocardiography to screen for the presence of diastolic dysfunction in 61 patients with RA and 107 healthy subjects. All participants were premenopausal women with no history of hypertension. DD includes an impaired relaxation with or without increased left ventricular (LV) filling pressures, pseudonormal filling, and restrictive filling based on parameters measured using echocardiography.

Results

The two groups were similar with respect to age (P=0.269). Patients with RA had significantly higher LV mass index, LV filling pressure, and lower E/A velocity than controls. All patients had preserved ejection fraction (EF ≥50%). DD was more common in patients with RA at 47% compared to 26% in the controls (P=0.004). Women with RA in the 30- to 49-year age range were over 3.5 times more likely to have DD than those of similar age in the control group (OR=3.54; 95% CI 1.27 to 9.85). Among patients with RA, high CRP levels were independently associated with DD even after adjustment for cardiovascular risk factors (P=0.009).

Conclusions

In premenopausal women with RA, DD is much more common and the age of onset is reduced. Early screening of myocardial function may provide an opportunity for preventing future cardiovascular disease.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13075-021-02629-1.

Impaction of regurgitation jet on anterior mitral leaflet is associated with diastolic dysfunction in patients with bicuspid aortic valve and mild insufficiency: a cardiovascular magnetic resonance study

To assess the impact of regurgitant jet direction on left ventricular function and intraventricular hemodynamics in asymptomatic patients with bicuspid aortic valve (BAV) and mild aortic valve regurgitation (AR), using cardiac magnetic resonance (CMR) feature tracking and 4D flow imaging. Fifty BAV individuals were retrospectively selected: 15 with mild AR and posterior regurgitation jet (Group-PJ), 15 with regurgitant jet in other directions (Group-nPJ) and 20 with no regurgitation (Controls). CMR protocol included cine steady state free precession (SSFP) sequences and 4D Flow imaging covering the entire left ventricle (LV) cavity and the aortic root. Cine-SSFP images were analyzed to assess LV volumes, longitudinal and circumferential myocardial strain. Circumferential and longitudinal peak diastolic strain rate (PDSR) and peak diastolic velocity (PDV) were reduced in group PJ if compared to group nPJ and control group (PDSR = 1.10 ± 0.2 1/s vs. 1.34 ± 0.5 1/s vs. 1.53 ± 0.3 1/s, p:0.001 and 0.68 ± 0.2 1/s vs. 1.17 ± 0.2 1/s vs. 1.05 ± 0.4 1/s ; p < 0.001, PDV = − 101.6 ± 28.1 deg/s vs. − 201.4 ± 85.9 deg/s vs. − 221.6 ± 67.1 deg/s; p < 0.001 and − 28.1 ± 8 mm/s vs. − 38.9 ± 11.1 mm/s vs. − 43.6 ± 14.3 mm/s, p < 0.001, respectively), whereas no differences have been found in systolic strain values. 4D Flow images (available only in 9 patients) showed deformation of diastolic transmitral streamlines direction in group PJ compared to other groups. In BAV patients with mild AR, the posterior direction of the regurgitant jet may hamper the complete mitral valve opening, disturbing transmitral flow and slowing the LV diastolic filling.

Left ventricle diastolic dysfunction in idiopathic inflammatory myopathies: A meta-analysis and systematic review

OBJECTIVES

Recent studies have confirmed that heart failure is one of the most important causes of death in patients with idiopathic inflammatory myopathy (IIM). Left ventricle diastolic dysfunction (LVDD) is closely associated with heart failure. Our aim is to determine if the prevalence of LVDD is increased in IIM patients.

METHODS

We performed a time- and language-restricted literature search to identify studies conducted to compare the echocardiographic parameters in IIM patients and controls. Mean differences were used to calculate the effect sizes of the echocardiographic parameters.

RESULTS

A total of 13 studies met the inclusion criteria and comprised a total of 227 juvenile dermatomyositis (JDM) patients, 391 adult IIM patients, and 550 controls. The adult IIM patients had lower mitral annular early diastolic velocity (e') and peak of early diastolic flow velocity/peak of late diastolic flow velocity (E/A) ratio compared to controls. The mean left atrial dimension and E/e' ratio was higher in adult IIM patients as compared to controls. Similarly, in JDM patients, the decreased e' was also observed.

CONCLUSION

Patients with IIM were more likely to have echocardiographic parameters indicative of diastolic dysfunction. The early heart assessments should be performed in IIM patients.

Left atrial ejection fraction is an indicator of left ventricular diastolic function

Left atrial ejection fraction (LAEF) has been previously shown to accurately distinguish between patients with and without clearly defined left ventricle diastolic dysfunction (LVDD) by ASE/EACVI criteria, but indeterminate cases were excluded. We sought to determine if LAEF could accurately distinguish between normal, indeterminate and LVDD patients. A retrospective cohort of 125 patients who underwent transthoracic echocardiography was studied. Comprehensive echocardiographic study was performed with measurement of validated diastolic parameters. Subjects were assigned LVDF ASE/EACVI categories. ANOVA test was used to compare means between groups and binary logistic regression and ROC curves to assess diagnostic accuracy. Mean LAEF was statistically different between groups (p < 0.001): 56.3% ± 4.5 for normal patients, 50.2% ± 5.5 for indeterminate patients and 44% ± 8.5 for patients with LVDD. LAEF distinguished LVDD from patients without diastolic dysfunction (OR = 0.73, 95% CI 0.63-0.65, p < 0.001) and ROC curve reveals excellent discriminative power (AUC 0.91, 95% CI 0.84-0.97, p < 0.001). LAEF also distinguished indeterminate patients from LVDD (OR = 0.9, 95% CI 0.83-0.95, p < 0.001) and ROC curve revealed good discriminative power (AUC 0.72, 95% CI 0.62-0.82, p < 0.001). LAEF can accurately differentiate between normal, indeterminate and LVDD patients and could be considered as an additional parameter in the study of diastolic function.

Metformin in non-diabetic patients with metabolic syndrome and diastolic dysfunction: the MET-DIME randomized trial

PURPOSE

Metabolic syndrome (MetS) affects one out of 3 adults in the western world and is associated with preclinical diastolic dysfunction that impairs functional capacity and quality of life (QoL). This randomized trial was designed to evaluate if the addition of metformin to the standard treatment of non-diabetic patients with MetS improves diastolic dysfunction.

METHODS

Prospective, randomized, open-label, blinded-endpoint trial. Fifty-four non-diabetic adults with MetS and diastolic dysfunction were randomized to lifestyle counseling or lifestyle counseling plus metformin (target dose 1000 mg bid). The primary endpoint was the change in mean e' velocity (assessed at baseline, 6, 12 and 24 months). Secondary endpoints were improvements in insulin resistance, functional capacity and QoL. Linear mixed effects modeling was used for longitudinal data analysis using modified intention-to-treat (mITT) and per-protocol (PP) approaches.

RESULTS

Forty-nine patients were included in the mITT analysis (mean age = 51.8 ± 6.4; 55% males). Metformin treatment was associated with a significant decrease in HOMA-IR. There was a significantly different mean change in e' velocity during the study period between trial arms, both in the mITT (at 24 months, change of +0.67 ± 1.90 cm/s in metformin arm vs. -0.33 ± 1.50 cm/s in control arm) and PP populations (+0.80 ± 1.99 cm/s in metformin arm vs. -0.37 ± 1.52 cm/s in control arm), using a random intercept linear mixed model. There were no significant differences in peak oxygen uptake and SF-36 scores between trial arms.

CONCLUSIONS

Treatment with metformin of non-diabetic MetS patients with diastolic dysfunction, on top of lifestyle counseling, is associated with improved diastolic function.