Soleus Muscle as a Surrogate for Health Status in Human Heart Failure

Strategies to optimise machine learning classification performance when using biomechanical features

Building prediction models using biomechanical features is challenging because such models may require large sample sizes. However, collecting biomechanical data on large sample sizes is logistically very challenging. This study aims to investigate if modern machine learning algorithms can help overcome the issue of limited sample sizes on developing prediction models. This was a secondary data analysis two biomechanical datasets - a walking dataset on 2295 participants, and a countermovement jump dataset on 31 participants. The input features were the three-dimensional ground reaction forces (GRFs) of the lower limbs. The outcome was the orthopaedic disease category (healthy, calcaneus, ankle, knee, hip) in the walking dataset, and healthy vs people with patellofemoral pain syndrome in the jump dataset. Different algorithms were compared: multinomial/LASSO regression, XGBoost, various deep learning time-series algorithms with augmented data, and with transfer learning. For the outcome of weighted multiclass area under the receiver operating curve (AUC) in the walking dataset, the three models with the best performance were InceptionTime with x12 augmented data (0.810), XGBoost (0.804), and multinomial logistic regression (0.800). For the jump dataset, the top three models with the highest AUC were the LASSO (1.00), InceptionTime with x8 augmentation (0.750), and transfer learning (0.653). Machine-learning based strategies for managing the challenging issue of limited sample size for biomechanical ML-based problems, could benefit the development of alternative prediction models in healthcare, especially when time-series data are involved.

Monocular 3D Human Pose Markerless Systems for Gait Assessment

Gait analysis plays an important role in the fields of healthcare and sports sciences. Conventional gait analysis relies on costly equipment such as optical motion capture cameras and wearable sensors, some of which require trained assessors for data collection and processing. With the recent developments in computer vision and deep neural networks, using monocular RGB cameras for 3D human pose estimation has shown tremendous promise as a cost-effective and efficient solution for clinical gait analysis. In this paper, a markerless human pose technique is developed using motion captured by a consumer monocular camera (800 × 600 pixels and 30 FPS) for clinical gait analysis. The experimental results have shown that the proposed post-processing algorithm significantly improved the original human pose detection model (BlazePose)'s prediction performance compared to the gold-standard gait signals by 10.7% using the MoVi dataset. In addition, the predicted T² score has an excellent correlation with ground truth (r = 0.99 and y = 0.94x + 0.01 regression line), which supports that our approach can be a potential alternative to the conventional marker-based solution to assist the clinical gait assessment.

Heart Failure-Induced Skeletal Muscle Wasting

PURPOSE OF REVIEW

Heart failure (HF) is a structural or functional cardiac abnormality which leads to failure of the heart to deliver oxygen commensurately with the requirements of the tissues and it may progress to a generalized wasting of skeletal muscle, fat tissue, and bone tissue (cardiac cachexia). Clinically, dyspnea, fatigue, and exercise intolerance are some typical signs and symptoms that characterize HF patients. This review focused on the phenotypic characteristics of HF-induced skeletal myopathy as well as the mechanisms of muscle wasting due to HF and highlighted possible therapeutic strategies for skeletal muscle wasting in HF.

RECENT FINDINGS

The impaired exercise capacity of those patients is not attributed to the reduced blood flow in the exercising muscles, but rather to abnormal metabolic responses, myocyte apoptosis and atrophy of skeletal muscle. Specifically, the development of skeletal muscle wasting in chronic HF is characterized by structural, metabolic, and functional abnormalities in skeletal muscle and may be a result not only of reduced physical activity, but also of metabolic or hormonal derangements that favour catabolism over anabolism. In particular, abnormal energy metabolism, mitochondrial dysfunction, transition of myofibers from type I to type II, muscle atrophy, and reduction in muscular strength are included in skeletal muscle abnormalities which play a central role in the decreased exercise capacity of HF patients. Skeletal muscle alterations and exercise intolerance observed in HF are reversible by exercise training, since it is the only demonstrated intervention able to improve skeletal muscle metabolism, growth factor activity, and functional capacity and to reverse peripheral abnormalities.

Calf muscle pump function as a predictor of all-cause mortality

Calf muscle pump (CMP) promotes venous return from the lower extremity and contributes to preload and cardiac output. Impaired CMP function may reflect a measure of frailty or cumulative disease burden or may impede cardiac function. The study objective was to test the hypothesis that impaired CMP negatively impacts survival. Consecutive adult patients who underwent venous strain gauge plethysmography at the Mayo Clinic Gonda Vascular Laboratory (January 1, 1998 - December 31, 2011) were assessed for overall survival. Patients with venous incompetence, venous obstruction or unilateral calf pump dysfunction were excluded. Risk of mortality was assessed with Cox proportional hazard ratios and after adjusting for Charlson Comorbidity Index variables. Over the study period, 2728 patients were included in the analysis. Compared to patients with normal CMP, those with impaired CMP were older (p < 0.001), predominantly female (p = 0.01) and had higher mean Charlson scores (p < 0.001). Patients with impaired CMP had a higher mortality rate at 5 (8.9% vs 2.4%), 10 (17.5% vs 5.9%), and 15 years (22.8% vs 8.3%) compared to those with normal CMP (p < 0.001 for each comparison). Of patients with heart failure, those with impaired CMP had worse survival at each 5-year increment compared to those with normal CMP (p < 0.05 at each increment). In conclusion, impaired CMP appears to be an independent predictor of poor outcomes after adjusting for variables within the Charlson Comorbidity Index. The association between impaired CMP, heart failure, and mortality may represent a negative impact on circulatory function or a surrogate measure of frailty.

Skeletal Muscle Alterations Are Exacerbated in Heart Failure With Reduced Compared With Preserved Ejection Fraction: Mediated by Circulating Cytokines?

BACKGROUND

A greater understanding of the different underlying mechanisms between patients with heart failure with reduced (HFrEF) and with preserved (HFpEF) ejection fraction is urgently needed to better direct future treatment. However, although skeletal muscle impairments, potentially mediated by inflammatory cytokines, are common in both HFrEF and HFpEF, the underlying cellular and molecular alterations that exist between groups are yet to be systematically evaluated. The present study, therefore, used established animal models to compare whether alterations in skeletal muscle (limb and respiratory) were different between HFrEF and HFpEF, while further characterizing inflammatory cytokines.

METHODS AND RESULTS

Rats were assigned to (1) HFrEF (ligation of the left coronary artery; n=8); (2) HFpEF (high-salt diet; n=10); (3) control (con: no intervention; n=7). Heart failure was confirmed by echocardiography and invasive measures. Soleus tissue in HFrEF, but not in HFpEF, showed a significant increase in markers of (1) muscle atrophy (ie, MuRF1, calpain, and ubiquitin proteasome); (2) oxidative stress (ie, higher nicotinamide adenine dinucleotide phosphate oxidase but lower antioxidative enzyme activities); (3) mitochondrial impairments (ie, a lower succinate dehydrogenase/lactate dehydrogenase ratio and peroxisome proliferator-activated receptor-γ coactivator-1α expression). The diaphragm remained largely unaffected between groups. Plasma concentrations of circulating cytokines were significantly increased in HFrEF for tumor necrosis factor-α, whereas interleukin-1β and interleukin-12 were higher in HFpEF.

CONCLUSIONS

Our findings suggest, for the first time, that skeletal muscle alterations are exacerbated in HFrEF compared with HFpEF, which predominantly reside in limb, rather than in respiratory, muscle. This disparity may be mediated, in part, by the different circulating inflammatory cytokines that were elevated between HFpEF and HFrEF.

Exercise training in Tgαq*44 mice during the progression of chronic heart failure: cardiac vs. peripheral (soleus muscle) impairments to oxidative metabolism

Cardiac function, skeletal (soleus) muscle oxidative metabolism, and the effects of exercise training were evaluated in a transgenic murine model (Tgα_q*44) of chronic heart failure during the critical period between the occurrence of an impairment of cardiac function and the stage at which overt cardiac failure ensues (i.e., from 10 to 12 mo of age). Forty-eight Tgα_q*44 mice and 43 wild-type FVB controls were randomly assigned to control groups and to groups undergoing 2 mo of intense exercise training (spontaneous running on an instrumented wheel). In mice evaluated at the beginning and at the end of training we determined: exercise performance (mean distance covered daily on the wheel); cardiac function in vivo (by magnetic resonance imaging); soleus mitochondrial respiration ex vivo (by high-resolution respirometry); muscle phenotype [myosin heavy chain (MHC) isoform content; citrate synthase (CS) activity]; and variables related to the energy status of muscle fibers [ratio of phosphorylated 5'-AMP-activated protein kinase (AMPK) to unphosphorylated AMPK] and mitochondrial biogenesis and function [peroxisome proliferative-activated receptor-γ coactivator-α (PGC-1α)]. In the untrained Tgα_q*44 mice functional impairments of exercise performance, cardiac function, and soleus muscle mitochondrial respiration were observed. The impairment of mitochondrial respiration was related to the function of complex I of the respiratory chain, and it was not associated with differences in CS activity, MHC isoforms, p-AMPK/AMPK, and PGC-1α levels. Exercise training improved exercise performance and cardiac function, but it did not affect mitochondrial respiration, even in the presence of an increased percentage of type 1 MHC isoforms. Factors "upstream" of mitochondria were likely mainly responsible for the improved exercise performance.NEW & NOTEWORTHY Functional impairments in exercise performance, cardiac function, and soleus muscle mitochondrial respiration were observed in transgenic chronic heart failure mice, evaluated in the critical period between the occurrence of an impairment of cardiac function and the terminal stage of the disease. Exercise training improved exercise performance and cardiac function, but it did not affect the impaired mitochondrial respiration. Factors "upstream" of mitochondria, including an enhanced cardiovascular O₂ delivery, were mainly responsible for the functional improvement.

Muscle size explains low passive skeletal muscle force in heart failure patients

Background

Alterations in skeletal muscle function and architecture have been linked to the compromised exercise capacity characterizing chronic heart failure (CHF). However, how passive skeletal muscle force is affected in CHF is not clear. Understanding passive force characteristics in CHF can help further elucidate the extent to which altered contractile properties and/or architecture might affect muscle and locomotor function. Therefore, the aim of this study was to investigate passive force in a single muscle for which non-invasive measures of muscle size and estimates of fiber force are possible, the soleus (SOL), both in CHF patients and age- and physical activity-matched control participants.

Methods

Passive SOL muscle force and size were obtained by means of a novel approach combining experimental data (dynamometry, electromyography, ultrasound imaging) with a musculoskeletal model.

Results

We found reduced passive SOL forces (∼30%) (at the same relative levels of muscle stretch) in CHF vs. healthy individuals. This difference was eliminated when force was normalized by physiological cross sectional area, indicating that reduced force output may be most strongly associated with muscle size. Nevertheless, passive force was significantly higher in CHF at a given absolute muscle length (non length-normalized) and likely explained by the shorter muscle slack lengths and optimal muscle lengths measured in CHF compared to the control participants. This later factor may lead to altered performance of the SOL in functional tasks such gait.

Discussion

These findings suggest introducing exercise rehabilitation targeting muscle hypertrophy and, specifically for the calf muscles, exercise that promotes muscle lengthening.