Nutritional Interventions in Heart Failure: Challenges and Opportunities

Machine Learning-Based Risk Factor Analysis and Prediction Model Construction for the Occurrence of Chronic Heart Failure: Health Ecologic Study

BACKGROUND

Chronic heart failure (CHF) is a serious threat to human health, with high morbidity and mortality rates, imposing a heavy burden on the health care system and society. With the abundance of medical data and the rapid development of machine learning (ML) technologies, new opportunities are provided for in-depth investigation of the mechanisms of CHF and the construction of predictive models. The introduction of health ecology research methodology enables a comprehensive dissection of CHF risk factors from a wider range of environmental, social, and individual factors. This not only helps to identify high-risk groups at an early stage but also provides a scientific basis for the development of precise prevention and intervention strategies.

OBJECTIVE

This study aims to use ML to construct a predictive model of the risk of occurrence of CHF and analyze the risk of CHF from a health ecology perspective.

METHODS

This study sourced data from the Jackson Heart Study database. Stringent data preprocessing procedures were implemented, which included meticulous management of missing values and the standardization of data. Principal component analysis and random forest (RF) were used as feature selection techniques. Subsequently, several ML models, namely decision tree, RF, extreme gradient boosting, adaptive boosting (AdaBoost), support vector machine, naive Bayes model, multilayer perceptron, and bootstrap forest, were constructed, and their performance was evaluated. The effectiveness of the models was validated through internal validation using a 10-fold cross-validation approach on the training and validation sets. In addition, the performance metrics of each model, including accuracy, precision, sensitivity, F1-score, and area under the curve (AUC), were compared. After selecting the best model, we used hyperparameter optimization to construct a better model.

RESULTS

RF-selected features (21 in total) had an average root mean square error of 0.30, outperforming principal component analysis. Synthetic Minority Oversampling Technique and Edited Nearest Neighbors showed better accuracy in data balancing. The AdaBoost model was most effective with an AUC of 0.86, accuracy of 75.30%, precision of 0.86, sensitivity of 0.69, and F1-score of 0.76. Validation on the training and validation sets through 10-fold cross-validation gave an AUC of 0.97, an accuracy of 91.27%, a precision of 0.94, a sensitivity of 0.92, and an F1-score of 0.94. After random search processing, the accuracy and AUC of AdaBoost improved. Its accuracy was 77.68% and its AUC was 0.86.

CONCLUSIONS

This study offered insights into CHF risk prediction. Future research should focus on prospective studies, diverse data, advanced techniques, longitudinal studies, and exploring factor interactions for better CHF prevention and management.

Prognostic Assessment with the Malnutrition Universal Screening Tool in Heart Transplant Recipients: A Pilot Study and a Single-Center Experience

Introduction and aims: Malnutrition is associated with increased morbidity and mortality in patients who undergo cardiac surgery. Nevertheless, objective assessment of malnourished patients undergoing heart transplantation (HT) is limited. We aimed to analyze the relationship between the malnutrition status and the early and late clinical outcomes of patients undergoing HT using a novel semi-quantitative tool. Methods: All patients aged ≥18 years who underwent HT between January 2015 and July 2020 in a single center were retrospectively evaluated and included in the study. The semi-quantitative Malnutrition Universal Screening Tool (MUST) score (already validated in heart failure) was calculated for each patient at the time of transplantation to assess their nutritional status. A propensity score weighting approach was performed to evaluate the association between the increase in MUST score and the risk of early complications and in-hospital mortality. A Cox regression analysis was performed to assess follow-up mortality. Results: A total of 168 HT patients (median age 58.4 years, IQR 49.5-65.2, men n = 128, 76%) were included within the study period. Their median preoperative BMI was 24.0 kg/m2 (IQR 21.2-27.9). Preoperative MUST scores of 0, 1, and ≥2 were found in 92 (55%), 24 (14%), and 52 (31%) patients, respectively. The median preoperative eGFR was 64.3 mL/min (IQR 49.0-83.2). An increase in MUST score (from 0 to 2) was not significantly related to major postoperative complications or in-hospital mortality. An analogous increase in MUST score was associated with increased follow-up mortality risk (hazard ratio 1.28, 95% CI 1.04-1.83, p = 0.024). Conclusions: Malnutrition assessed with the MUST score seems not to be associated with increased in-hospital mortality or major postoperative complications in patients who undergo HT, but according to our preliminary data it is related to patients' long-term mortality.

Range of plasma brain natriuretic peptide (BNP) levels in hemodialysis patients at a high risk of 1-year mortality and their relationship with the nutritional status: a retrospective cohort study in one institute

Background

Brain natriuretic peptide (BNP) levels are used as a marker of heart failure, which is the leading cause of morbidity and mortality in dialysis patients. BNP levels increase as renal function declines. The range of BNP levels associated with satisfactory longevity in dialysis patients currently remains unknown.

Methods

In total, 660 patients receiving maintenance hemodialysis were enrolled. BNP levels were measured at the end of the year and in a follow-up to assess 1-year mortality between 2008 and 2012. Patients were divided into six groups according to BNP levels: < 50 (reference), 50 to < 100, 100 to < 300, 300 to < 500, 500 to < 1000, and ≥ 1000 pg/mL. One-year mortality at each BNP level was analyzed using Cox’s proportional hazards model after adjustments for confounding factors.

Results

During the follow-up period, 78 (11.8%) deaths were recorded. After adjustments for confounding factors, such as gender, age, hemodialysis vintage, and primary disease, the risk of 1-year mortality was significantly high with BNP levels of 500 to < 1000 (hazard ratio [HR] 3.010; 95% confidence interval [CI] 1.065–10.729; P = 0.037) and more than 1000 pg/mL (HR 5.291; 95%CI 2.014–18.170; P = 0.0003). After adjustments for Kt/V, the risk of 1-year mortality was also significantly high with BNP levels of 500 to < 1000 (HR 3.045; 95%CI 1.065–10.929; P = 0.037) and more than 1000 pg/mL (HR 5.221; 95%CI 1.943–18.165; P = 0.0006). Following further adjustments for nutritional factors, such as albumin levels, total cholesterol levels, the normalized protein catabolic rate (nPCR), body mass index (BMI), and percent creatinine generation rate (%CGR), BNP levels of 500–1000 (HR 1.990; 95%CI 0.639–7.570; P = 0.244), and more than 1000 pg/mL (HR 2.100; 95%CI 0.663–8.105; P = 0.213) were no longer risk factors.

Conclusion

In dialysis patients, a BNP level ≥ 500 pg/mL is a risk factor for 1-year mortality. The risk associated with high BNP levels is reduced by nutritional factors, which suggests a relationship between high BNP levels and the nutritional status. In conclusion, efforts are needed to maintain BNP levels at lower than 500 pg/mL and improve the nutritional status.

High-fat diet affects skeletal muscle mitochondria comparable to pressure overload-induced heart failure

In heart failure, high-fat diet (HFD) may exert beneficial effects on cardiac mitochondria and contractility. Skeletal muscle mitochondrial dysfunction in heart failure is associated with myopathy. However, it is not clear if HFD affects skeletal muscle mitochondria in heart failure as well. To induce heart failure, we used pressure overload (PO) in rats fed normal chow or HFD. Interfibrillar mitochondria (IFM) and subsarcolemmal mitochondria (SSM) from gastrocnemius were isolated and functionally characterized. With PO heart failure, maximal respiratory capacity was impaired in IFM but increased in SSM of gastrocnemius. Unexpectedly, HFD affected mitochondria comparably to PO. In combination, PO and HFD showed additive effects on mitochondrial subpopulations which were reflected by isolated complex activities. While PO impaired diastolic as well as systolic cardiac function and increased glucose tolerance, HFD did not affect cardiac function but decreased glucose tolerance. We conclude that HFD and PO heart failure have comparable effects leading to more severe impairment of IFM. Glucose tolerance seems not causally related to skeletal muscle mitochondrial dysfunction. The additive effects of HFD and PO may suggest accelerated skeletal muscle mitochondrial dysfunction when heart failure is accompanied with a diet containing high fat.