Influence of Yoga-based cardiac rehabilitation on gut-microbiome diversity and Gut–Heart–Brain axis (YoGH-Biome) in heart failure: a study protocol

Abstract

Heart failure (HF) is a chronic and progressive cardiovascular condition associated with significant morbidity, mortality and healthcare burden. Increasing evidence points to a critical role of gut dysbiosis and the gut–heart–brain axis in HF pathophysiology. Altered gut microbiota may influence systemic inflammation, neurohormonal activity and cardiac function through gut-derived metabolites such as trimethylamine N-oxide (TMAO) and short-chain fatty acids (SCFAs). Yoga-based cardiac rehabilitation (Yoga-CaRe) is a cost-effective intervention that has been shown to improve quality of life, exercise capacity and cardiovascular outcomes in cardiac patients. However, the mechanism underlying its benefits remains unclear. Furthermore, its effect on gut microbiota diversity and the downstream impact on the gut–heart–brain axis in HF remains largely unexplored. This study outlines a prospective, randomised, open-label, blinded-endpoint trial investigating the effects of a 12-week Yoga-CaRe intervention versus enhanced standard care in 60 HF patients with reduced ejection fraction. Participants will be randomly assigned in a 1:1 ratio to either the Yoga-CaRe or the control group. The Yoga-CaRe group will participate in 20 supervised yoga sessions, complemented by guided daily home practice, while the control group will receive enhanced standard care. The trial will assess changes in gut microbiota composition, levels of gut-derived metabolites (TMAO and SCFAs), inflammatory biomarkers (TNF-α and high-sensitivity C reactive protein), heart rate variability, 6 min walk test (6MWT) and echocardiography. Biological samples and clinical data will be analysed using integrated bioinformatics and statistical approaches to evaluate intervention efficacy and identify potential mechanistic pathways. The YoGH-Biome study has received ethical clearance from the Institutional Ethics Committee of the SDM College of Medical Sciences and Hospital, India (SDMIEC/2025/1073). It is registered with the Clinical Trials Registry of India. Study results will be disseminated via scientific publications, conferences and stakeholder forums to inform integrative strategies for HF management. Trial registration number: CTRI/2023/12/060757.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Heart failure is associated with gut dysbiosis and altered gut-derived metabolites that contribute to systemic inflammation, autonomic dysfunction and disease progression through the gut–heart–brain axis. Yoga-based cardiac rehabilitation improves functional capacity and quality of life in cardiac populations, but its mechanistic effects in heart failure remain unclear.

WHAT THIS STUDY ADDS

• This study will be the first mechanistic randomised controlled trial to evaluate the effects of yoga-based cardiac rehabilitation on gut microbiome diversity, gut-derived metabolites, inflammation, autonomic function and clinical outcomes in patients with heart failure with reduced ejection fraction.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• By clarifying microbiome-mediated pathways underlying yoga-based rehabilitation, this study may inform integrative, scalable and low-cost strategies for heart failure management, and importantly, guide future research targeting the gut–heart–brain axis.

Introduction

Heart failure (HF) is a global health issue affecting over 26 million people worldwide, resulting from the heart’s inability to pump blood efficiently.¹ As the population ages, cardiovascular (CV) risk factors like obesity, diabetes and hypertension increase, leading to the worsening of the disease.² Coronary artery disease (CAD), myocardial infarction (MI), hypertension, atrial fibrillation, valvular heart disease, certain infections and cardiomyopathies are some of the most common causes of HF.³ The 2021 European Society of Cardiology (ESC) guidelines divided HF into three categories based on left ventricular ejection fraction (LVEF), that is, HF with reduced EF ≤40% (HFrEF), HF with mildly reduced EF, ie, 41%–49%, and HF with preserved EF ≥50% (HFpEF). HF is a costly disease with significant economic impacts, particularly in low-income and middle-income countries.³ HFrEF exerts a greater impact on functional capacity because of its more pronounced impairment in the heart’s ability to pump blood effectively.⁴ An effective therapeutic approach for HF remains a challenging unmet need. Early diagnosis and individualised treatment strategies are crucial for improving patients’ quality of life (QoL) and functional capacity, and decreasing hospitalisation and mortality rates.⁵ The prognosis of chronic HF is poor with current standard therapy, with a survival rate of 80%–90% at 1 year, 50%–60% at 5 years and 30% at 10 years.^{3 4} Moreover, the pathophysiology of HF remains unclear and needs to be explored to identify new targets for therapies to reduce the risk of HF.

Cardiac rehabilitation (CR) is the process of restoring desirable levels of physical, social and psychological functioning after the onset of CV illness.^{6 7} CR aims to optimise functional capacity, enhance patient QoL and minimise the risk of recurrent cardiac events.⁸ Yoga is an ancient Indian mind-body medicine that works in all aspects of a person. Yoga covers most of the elements of a comprehensive CR programme: improved physical fitness, stress reduction and lifestyle improvement.⁹ Recently, we evaluated the effects of Yoga-CaRe on CV events and QoL in MI patients in a multicentre randomised controlled trial (RCT). The trial found Yoga-CaRe to be safe and to improve QoL in patients postacute MI.¹⁰ Other clinical studies have also found that Yoga-CaRe programmes improve CV risk factors in patients with CAD.¹¹ However, the mechanism underlying its benefits remains unclear and is yet to be explored. Moreover, the quality RCT of Yoga-CaRe on heart health in HF patients is unavailable, and its underlying mechanism remains unclear.

A bilateral relationship among the gut, heart and brain has been described as the gut–heart–brain axis.^{12 13} Hippocrates, the ‘father of modern medicine’, claimed that ‘all disease begins in the gut’. In contrast, ancient Indian science (Ayurveda and Yoga) believes that all sickness begins in the gut and mind. The human gut contains more microbial cells (more than 10¹⁴) than human cells, and these microbes play a critical role in the host’s homeostasis. Emerging evidence suggests that gut dysbiosis, characterised by an imbalance in the composition or diversity of the gut microbiota, may contribute to the development and progression of HF by modulating immune responses and inflammatory pathways.^{14 15} Alterations such as reduced production of short-chain fatty acids (SCFAs), elevated circulating levels of trimethylamine N-oxide (TMAO), increased intestinal permeability (‘leaky gut’), and disrupted bile acid metabolism have been associated with immune dysregulation and systemic inflammation implicated in HF pathogenesis.¹⁶ Furthermore, mental stress is well recognised as a significant contributor to both gut dysfunction and CV diseases, including HF.¹⁷ While yoga has been shown to reduce stress and enhance QoL, its impact on gut microbiome diversity and the gut–heart–brain axis remains understudied and warrants further investigation. Therefore, we aimed to investigate the impact of Yoga-CaRe on gut microbiome diversity and the gut–heart–brain axis to explore the potential effects and underlying mechanisms of yoga on heart health in HF patients, particularly those with reduced EF.

Methods and analysis

Design

Prospective, Randomised, Open Label, Endpoint, Blinding Mechanistic study (PROBE).

Study setting

The study will take place at a tertiary care centre equipped with advanced CV diagnostic and rehabilitation facilities.

Study timeline

Participant recruitment began in January 2025, and the study is expected to be completed by February 2026, including data collection and follow-up.

Sample size determinations

The sample size was determined based on the primary outcome, i.e., the change in Shannon Diversity Index of the gut microbiome from baseline to 12 weeks. Using pilot data and prior studies on microbiome-based interventions, a mean difference of 0.3 units in the Shannon index (SD=0.4) was considered clinically meaningful. To detect this difference with a two-tailed α=0.05 and 80% power, the minimum required sample size was 25 participants per group. Accounting for an estimated 20% dropout rate, a total of 60 participants (30 per group) will be enrolled. This sample size is expected to provide sufficient power to detect biologically meaningful changes in microbial diversity (α diversity) following the 12-week Yoga-CaRe intervention compared with enhanced standard care.

Study participants

Inclusion criteria

Inclusion criteria for the study eligibility will be male and female patients aged 18 years and above, diagnosed with HFrEF based on ESC guidelines, must be classified as New York Heart Association functional class II/III, have an EF of less than 40%, and be physically capable of participating in a hospital-based CR programme.

Exclusion criteria

Exclusion criteria include the presence of comorbid conditions that significantly limit physical activity, recent major CV procedures, or planned elective CV interventions or hospitalisations within the next 3 months. Patients with HF secondary to a significant correctable or reversible cause, such as primary valvular disease or congenital heart defects amenable to correction, will also be excluded. Additional exclusion criteria include current participation in other clinical trials or engagement in regular yoga or structured exercise training (more than once per week at moderate-to-vigorous intensity) within the past 6 weeks.

Randomisation and allocation concealment

Eligible and consenting patients will be randomised individually in a 1:1 ratio to either the Yoga-CaRe intervention group or the Enhanced Standard Care group. Randomisation will be carried out using a computer-generated algorithm via an Interactive Web Response System after entry of required baseline data. Variable block sizes will be used, stratified by clinical site, gender and age category (<60 years or ≥60 years), to ensure equitable distribution.

Planned interventions

Yoga CaRe group

Participants will attend 20 supervised yoga sessions over 12 weeks, complemented by daily guided home practice. The Yoga-CaRe programme will include a tailored combination of asanas (physical postures), pranayama (breathing exercises), dhyana (meditation) and dietary modifications designed to accommodate the functional capacity and clinical requirements of HF patients (see online supplemental tabls 1–4 for details). Caregivers or spouses of participants will be invited to these sessions.

Enhanced standard care group

Patients in the control arm will receive enhanced standard care comprising three structured educational sessions delivered over 3 months (45–60 min each), covering HF management, medication adherence, dietary recommendations, safe physical activity, stress reduction, sleep hygiene and early symptom recognition. Participants will receive printed educational materials summarising key content in the hospital before discharge (session 1), followed by two sessions offering standard educational lifestyle advice and nutritional counselling at weeks 5 and 12 (sessions 2, 3). A different team member (ie, not the yoga physician) will deliver these sessions to avoid contamination. The sessions will be conducted by qualified health science professionals trained in standardised patient education, using a detailed intervention manual to ensure uniformity.

Routine clinical practice will remain unchanged in both the intervention and control groups. Apart from the study intervention, standard clinical procedures, such as prescribing secondary prevention drugs, will be maintained identically in either arm.

Outcome measures

• Primary outcome: The primary outcome of the study will be the change in gut microbiome alpha diversity, assessed by the Shannon Diversity Index, from baseline to 12 weeks of intervention. This index reflects both species richness and evenness and is widely recognised for evaluating microbial diversity in gut microbiome studies.

• Secondary outcomes: The secondary outcomes will include changes in gut microbiome - beta diversity (Bray–Curtis and UniFrac distances), gut-derived metabolites (TMAO and SCFAs), systemic inflammatory markers (TNF-α, IL-6 and high-sensitivity C reactive protein (hs-CRP)), cardiac function (LVEF, E/A ratio, E/e′), functional exercise capacity (6MWD) and autonomic function (heart rate variability (HRV) indices).

Expected outcome

The Yoga-CaRe programme is anticipated to outperform enhanced standard care by improving gut microbiome diversity and modulating the gut–heart–brain axis, enhancing functional capacity, sympathovagal balance and SCFA, while reducing TMAO levels and inflammation. These changes are expected to translate into improved HF outcomes, including better EF and reduced CV events.

Study protocol

Eligible patients will be screened for inclusion based on predefined criteria. A detailed medical history, including prior medication use, demographic data and relevant personal history, will be documented. A clinical examination, necessary investigations and measurement of baseline parameters, such as brachial blood pressure and heart rate, will follow. On confirmation of eligibility, baseline data will be collected, and participants will be randomised to their assigned study arms. The intervention group will receive the Yoga-CaRe programme, while the control group will receive enhanced standard care comprising structured educational sessions on heart-healthy living (figure 1).

Figure 1. Study flow chart. Yoga-CaRe, Yoga-based cardiac rehabilitation.

Trial timeline, assessments and follow-up

1. Timeline: The clinical trial will be conducted over 24 months, with the first 15 months for trial preparation and recruitment, and an additional 3 months for participant follow-up and 6 months for data analysis and reporting.

2. Baseline assessment: At baseline, demographic and clinical data, including age, height, weight, body mass index (BMI), blood pressure, medical history, medication use and dietary habits, will be recorded. Blood (2 mL) will be collected in EDTA and plain tubes for biochemical investigations, including gut-derived metabolites and inflammatory markers. Stool samples will be collected in sterile containers and stored at –80°C for subsequent analyses of gut microbiome composition, diversity and richness. Functional exercise capacity, echocardiography, autonomic nervous system (ANS) function, Food Frequency Questionnaire (FFQ) and QoL will be assessed.

3. Assessment at 3 months: Immediately after the last intervention session in both groups, a comprehensive reassessment of all baseline outcome measures will be conducted at the 3-month follow-up to determine the intervention’s impact. A 15-day window period will be available for completing these follow-up assessments.

Adherence monitoring, intervention fidelity and participant engagement

The Yoga-CaRe intervention’s fidelity and participant adherence will be monitored using multiple standardised procedures. Attendance for all supervised yoga sessions (20 sessions over 12 weeks) will be recorded, and adherence will be defined as participation in ≥80% of sessions. The recorded sessions will be assessed for intervention fidelity against the standardised quality kit by the Yoga experts. Home-based practice will be tracked using daily practice logs, reviewed weekly by the study yoga instructor, and supplemented with random telephonic or video check-ins; optional video/photo logs and periodic in-person demonstrations will further verify compliance. Certified yoga instructors/physicians will deliver the intervention strictly according to the Yoga-CaRe manual, and session checklists will document adherence to prescribed sequences, timings and allowed modifications. Missed sessions, adverse events and reasons for withdrawal will be documented systematically. Fidelity of the intervention will also be ensured through structured training for Yoga instructors/physicians and 6 monthly refresher sessions. Regular direct observations and video recordings of sessions will complement these. Together, these approaches ensure high intervention fidelity, minimise behavioural and lifestyle-related confounding, and strengthen the internal validity and reproducibility of the study’s findings on gut microbiota and clinical outcomes.

Methods of data collection

1. Demographic characteristics: Baseline demographic and clinical information, including age, height, weight, BMI, blood pressure, medical history, medication usage and dietary habits, will be obtained from all participants through structured interviews and medical record review.

2. Assessment of gut microbiome composition

   1. Collection of stool sample: Participants will be provided with sterile, prelabelled stool collection containers along with clear instructions to avoid contamination during collection. Approximately 2–5 g of stool will be obtained and returned to the study site within 4–6 hours of collection (not exceeding 24 hours), after which samples will be stored at –80°C to preserve microbial integrity.

   2. DNA extraction: Genomic DNA will be extracted from the stool sample using the Real Gene Stool Genomic DNA extraction kit, with purity and concentration assessed via UV spectrophotometry (A260/A280 & A260/A230 ratios), and integrity confirmed by agarose gel electrophoresis. Only high-quality DNA will be used for downstream analyses.

   3. 16S rRNA metagenomic sequencing: The gut microbiome composition will be assessed through next-generation sequencing of the bacterial 16S rRNA gene, focusing on the hypervariable V3 and V4 regions. DNA amplification will be followed by PCR, and sequencing libraries will be prepared according to the Illumina 16S Metagenomic sequencing protocol. The sequencing will be performed on the Illumina MiSeq platform, providing insights into the microbial diversity and composition of the gut microbiota in patients with HF, thereby aiding understanding of the potential relationship between changes in gut microbiome composition and clinical outcomes.^{18 19}

3. Assessment of gut-derived metabolites: Serum metabolites such as TMAO and SCFAs will be quantified using specific ELISA kits to assess their presence and levels in serum or plasma samples, thereby further elucidating metabolic alterations associated with the condition.

4. Assessment of inflammatory markers: Serum or plasma levels of TNF-α, IL-6 and hs-CRP will be measured using ELISA-based assays. These markers provide critical insights into systemic inflammation in HF patients, aiding in understanding the biochemical pathways involved in disease progression and treatment response.

5. Assessment of functional exercise capacity: It will be assessed using 6MWT, a well-established and validated method for HF patients. Participants will be asked to walk continuously along a designated corridor for 6 min, during which the total distance travelled will be measured. The test will be conducted under clinical supervision following standard protocols, with vital signs including blood pressure, heart rate, oxygen saturation and perceived exertion recorded before and after the test.²⁰

6. Assessment of systolic and diastolic function of the heart: A comprehensive echocardiographic assessment will be conducted to evaluate systolic and diastolic function. Key measurements will include LVEF, end-diastolic and end-systolic volumes, E/A ratio and E/e’ ratio, using two-dimensional, M-mode and Doppler imaging techniques following American Society of Echocardiography guidelines.²¹

7. Assessment of ANS function: Sympathetic and parasympathetic tone will be assessed by measuring HRV.^{22 23} A 5 min short-term ECG will be recorded in the standard limb lead II configuration using a digital polygraph (Niviqure, India). The data acquired (R-R intervals) will be analysed using the frequency-domain method with the HRV analysis programme developed by the Biomedical Signal Analysis Group at the University of Kuopio, Finland. Low frequency (LF) reflects sympathetic activity, high frequency (HF) reflects parasympathetic activity, and the LF/HF ratio indicates autonomic balance.

8. FFQ: Participants’ habitual dietary intake will be evaluated using a structured FFQ. The questionnaire will assess the frequency of consumption of key food groups, including fruits, green leafy vegetables, deep-fried foods, egg/meat/poultry products, milk and milk products, and sugary drinks.

Assay validation and quality control procedures

All biochemical assays, including serum TMAO, SCFAs (acetate, propionate, butyrate) and inflammatory biomarkers (TNF-α, IL-6, hs-CRP), will be quantified using validated ELISA kits following standardised operating procedures and manufacturer-recommended guidelines. Calibration curves, assay linearity (R² >0.98), detection limits, and intra-assay and interassay variation will be documented. Each sample will be analysed in duplicate/triplicate, with those showing >10% variation reassayed, and a subset (10%) included as blinded duplicates to verify reproducibility. Standards, internal controls and external references will be incorporated in every plate to monitor batch consistency. All samples will be processed within 2 hours of collection, aliquoted, stored at –80°C and thawed only once, with anonymised barcoding to reduce operator bias. Trained personnel blinded to group allocation will perform all analyses, and raw OD values along with standard curve data will be archived for audit and quality assessment, ensuring high analytical precision and data validity across all endpoints.

Confounding control

To minimise potential confounding influences on gut microbiota and study outcomes, several standardised control measures will be implemented. Dietary intake will be assessed at baseline and post-intervention using a structured FFQ, with changes in key food groups (fibre, fermented foods, animal proteins and fats) incorporated as covariates in analyses. Comprehensive medication histories, including β-blockers, ACE inhibitors/ARBs, statins, diuretics and anti-diabetic agents, will be recorded at each visit, with participants instructed to maintain stable regimens; any changes will be documented and adjusted for statistical analysis. Individuals using antibiotics, probiotics or prebiotic supplements within 4 weeks before enrolment or during the study will be excluded or deferred to avoid microbiome-altering effects. Lifestyle factors such as smoking, alcohol use and physical activity will be recorded through standardised questionnaires, monitored for stability and included as covariates in regression models where relevant.

Data analysis

The data analysis for this mechanistic RCT will follow the intention-to-treat principle, with per-protocol analysis conducted as a sensitivity check. Descriptive statistics will be used to summarise both quantitative and qualitative data. Continuous variables will be presented as means±SD or medians with IQRs, depending on distribution, while categorical variables (eg, gender, comorbidities) will be expressed as frequencies and percentages. Data normality will be tested using the Shapiro-Wilk or Kolmogorov-Smirnov tests. Baseline characteristics between groups will be compared using independent t-tests or Mann-Whitney U tests for continuous variables and χ² or Fisher’s exact tests for categorical variables. Within-group changes from baseline to postintervention will be assessed using paired t-tests or Wilcoxon signed-rank tests, depending on data distribution. Between-group differences in outcomes will be evaluated using analysis of covariance (ANCOVA), with baseline values as covariates. Microbiome sequencing data will be processed and analysed using a validated bioinformatics pipeline implemented in QIIME2 and DADA2. Raw sequences will undergo demultiplexing, quality filtering, denoising, chimaera removal and generation of amplicon sequence variants for high-resolution taxonomic classification using the SILVA (V.138) reference database.²⁴ To address the compositional nature of microbiome data, abundance tables will be transformed using the centred log-ratio transformation before statistical testing. Beta diversity analyses will be based on Aitchison distance, while alpha diversity indices (Shannon, Simpson and Observed OTUs) will be compared within and between groups. The primary outcome, change in Shannon Diversity Index from baseline to 12 weeks, will be compared between the Yoga-CaRe and control groups using ANCOVA, adjusting for baseline Shannon index, age, BMI and medication use as covariates. In secondary analyses, beta diversity will be assessed using Bray-Curtis and weighted/unweighted UniFrac distances, visualised through principal coordinates analysis. Differential abundance of taxa will be identified using DESeq2 and ANCOM-BC2, with Benjamini-Hochberg false discovery rate (FDR) correction (q<0.05) for multiple comparisons. Batch effects arising from sequencing runs or DNA extraction batches will be evaluated using ComBat normalisation methods and included as covariates if significant. Associations between microbial features and biochemical/clinical parameters (TMAO, SCFAs, IL-6, TNF-α, hs-CRP, HRV, 6MWT) will be examined using Spearman correlations and multivariable regression models. Multiomics integration of microbiome, metabolite and clinical data will be performed using sparse Partial Least Squares regression and visualised as correlation networks.^{24 25}

Reporting standards and transparency

This study protocol has been developed and reported in accordance with internationally recognised guidelines to ensure methodological rigour, transparency and reproducibility. The trial design and documentation adhere to the SPIRIT 2013 (Standard Protocol Items: Recommendations for Interventional Trials) statement, ensuring inclusion of all essential components of a clinical trial protocol. The Yoga-CaRe intervention is described in compliance with the TIDieR (Template for Intervention Description and Replication) checklist, with detailed specifications on intervention content, delivery, duration, frequency, instructor qualifications and fidelity monitoring to support replication. Microbiome-related procedures, including sampling, sequencing, data processing and statistical analyses, follow the Strengthening the Organising and Reporting of Microbiome Studies guidelines to maintain consistency and improve reporting quality in microbiome research. Furthermore, the protocol conforms to Good Clinical Practice standards and adheres to the ethical principles outlined by the Indian Council of Medical Research (ICMR) and the Declaration of Helsinki. Supplementary SPIRIT and TIDieR checklists will be submitted to support transparent reporting and confirm adherence to these standards.

Ethics and dissemination

Yoga CaRe-HF is an ongoing multicentre randomised controlled trial, and YoGH-Biome is an embedded mechanistic study investigating the biological effects of the yoga intervention. The Data Monitoring Committee (DMC) will oversee the study to ensure participant safety, monitor adverse events and maintain ethical compliance. The DMC will regularly review trial progress and adherence to protocols, safeguarding participant confidentiality and anonymity. The trial is registered with Clinical Trials Registry India (CTRI/2023/12/060757). Data confidentiality will be strictly maintained and accessed only by the principal and co-principal investigators.

Data availability statement

Data are available on reasonable request.