Edited by: Stefania Paolillo, University of Naples Federico II, Italy
Reviewed by: Li Wang, Soochow University, China; Catherine Anne Elliot, Lincoln University, New Zealand
This article was submitted to Exercise Physiology, a section of the journal Frontiers in Physiology
†These authors have contributed equally to this work
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Chronic heart failure (HF) is a common symptom complex characterized by shortness of breath, fatigue, fluid retention and severe exercise intolerance. It is estimated that about 1–2% of the adult population in developed countries suffer from this disease, among people over 70 years even more than 10% (Kuznetsova et al.,
Recent data show that about 50% of all HF patients have normal or near-normal left ventricular ejection fraction – a condition named HF with preserved ejection fraction (HFpEF) (Owan et al.,
While the prognosis of patients with HFpEF is better than in HF with reduced ejection fraction (HFrEF) (Meta-Analysis Global Group in Chronic Heart,
Severe exercise intolerance is a hallmark of HF and a strong determinant of morbidity and mortality (Haykowsky and Kitzman,
In the process of designing training protocols, the optimal selection of training intensity plays a central role (Vanhees and Stevens,
While aerobic exercise is well-established for the treatment of HFrEF, no consensus exists for the management of HFpEF (Pina et al.,
Despite the superiority of HIT over MCT in patients with HFrEF, no differences were found between HIT and MCT in the OptimEx trial that compared the effects of these training modalities in patients with HFpEF. However, patients in the MCT group completed five training sessions per week compared to three training sessions in the HIT group. Moreover, training modalities were not balanced between groups with regards to energy consumption and weekly exercise time in the MCT group, which were about 80% higher in the MCT group than in the HIT group (Mueller et al.,
VO2peak is a valid and reproducible marker for exercise capacity and a strong predictor of prognosis and QoL (Haykowsky et al.,
Cardiac and vascular remodeling and fibrosis stimulated by chronic inflammation appear to be among the most important factors for the progression of HFpEF (Ponikowski et al.,
In addition to exercise testing, monitoring of habitual physical activity may offer further useful information to assess functional status, as it accounts for changes in activity behavior, which exercise testing does not (Jehn et al.,
The primary aim of this study is to investigate the impact of a supervised 12-week HIT on exercise capacity, measured as VO2peak, in patients with HFpEF, compared to an MCT. As secondary objectives, training-related effects on biomarkers as surrogates of systemic inflammation, endothelial function and clinical prognosis, cardiac and vascular structure and function, functional status, QoL, body composition, and habitual physical activity will be examined. An additional secondary objective is to examine the tHb-mass with regards to VO2peak.
The proposed study is a prospective, 2-arm randomized controlled trial. The intervention arm includes the investigation of a supervised 12-week HIT on exercise capacity, functional status and QoL in patients with HFpEF. A control group training with isocaloric MCT will serve as comparator. Patients will be randomized in a 1:1 ratio to either the intervention or control group using the computer-based system provided by Castor. Based on a sample size calculation, a total of 86 patients will be enrolled (43 patients in each group). The Department of Sport, Exercise and Health (DSBG) at the University of Basel will be the main study center collaborating with the Division of Cardiology at the Clinic Arlesheim, local general practitioners and cardiologists.
Members of the study team will screen the medical files of the DSBG and the Clinic Arlesheim, Division of Cardiology, for previously treated patients with an established diagnosis of HFpEF. Eligible patients will be contacted by phone and asked whether they are willing to take part in this study. If patients agree to participate, they will be invited to the DSBG for further information, giving informed consent, and initiating the study procedures. In addition, collaborating institutions will also provide patients with HFpEF with an information sheet about the goal and the conduct of the present study and refer them to the DSBG for final screening. In order to minimize bias that may be introduced by differences in executing and interpreting technical exams, questionnaires, or training sessions, all study related measurements and procedures will be performed in a standardized manner at the DSBG. Inclusion and exclusion criteria are presented in
Inclusion and exclusion criteria.
Informed consent as documented by signature | Planned cardiac interventions in the following 6 months |
NYHA functional class II-III | Unstable angina pectoris |
Signs and symptoms of chronic HF: | Uncontrolled brady- or tachyarrhythmia |
Dyspnea | Uncontrolled hypertonic blood pressure |
Paroxysmal nocturnal dyspnea | Severe uncorrected valvular heart disease |
Reduced exercise capacity | Paroxysmal atrial fibrillation |
Extended recovery after exercising | Clinically significant concomitant disease states (e.g., advanced renal failure, hepatic |
Fatigue | dysfunction, insulin-dependent diabetes) |
Peripheral edema (lower leg, ankle) | |
EF > 50% | COPD grades III-IV according to the GOLD classification |
Structural or functional changes in echocardiography: | |
LAVI > 34 ml·m−2 OR | On-going cancer treatment |
LVMI > 115 g·m−2 (men), >95 g·m−2 (women) OR | Significant musculoskeletal disease limiting exercise capacity |
E/E' ratio >13 AND mean E' septal and lateral wall <9 cm·s−1 | |
NT-proBNP > 125 pg·ml−1 | Active infection |
At least 4 weeks on stable medical treatment or without signs and symptoms of cardiac de-compensation | Immunosuppressive medical therapy |
Blood transfusion within the previous 30 days | |
Trainable: | Pregnancy or lactation |
VT > 40% of predicted VO2max AND | |
VO2peak > 10 ml·min−1·kg−1 at the screening visit | Known or suspected non-compliance, drug or alcohol abuse |
Inability to follow the procedures of the study, e.g., due to insufficient language skills, psychological disorders, dementia, etc. | |
Participation in another intervention study | |
Enrolment of the investigators, their family members, and other persons involved in |
|
Life-expectancy <6 months | |
Age <18 years |
The study procedures and assessments are shown in
Study organization. Flow chart for patients, showing screening, inclusion and exclusion, randomization, intervention, tests at baseline, post-intervention visit and follow-up. HIT, high-intensity interval training; MCT, moderate continuous training; HF, heart failure; ECG, electrocardiogram; PWV, pulse wave velocity; FMD, flow mediated dilation; CAR, carotid artery reactivity; VO2peak, maximal oxygen uptake; QoL, quality of life; SF-8, short form health survey 8; KCCQ, Kansas City cardiomyopathy questionnaire; MLWHFQ, Minnesota living with heart failure questionnaire; CMJ, counter movement jump.
The primary outcome will be the change in VO2peak as a marker for exercise capacity from the baseline visit to the end of the 12-week training intervention (post-intervention visit) following HIT with strength training compared to MCT with strength training. VO2peak will be determined by spiroergometry. Patients will undergo an incremental symptom-limited exercise test on an electronically operated bicycle ergometer (Ergoselect 200, Ergoline, Bitz, Germany) using one of two fixed ramp protocols, depending on the patient's fitness status (protocol 1: warm-up unloaded, then increase of 7 W·min−1; protocol 2: warm-up at 10 W, then increase of 10 W·min−1) (Task Force of the Italian Working Group on Cardiac et al.,
Secondary outcomes measured at baseline and follow-up will include the assessment of
tHb-mass by the optimized carbon monoxide (CO) rebreathing method. The measurement procedure includes inhaling a small amount of CO using the Detalo Performance™-rebreathing device (Detalo Health Aps, Denmark). CO completely binds to Hb in the blood, which enables the calculation of tHb-mass from the difference of CO-Hb concentration before and after inhalation (~5%) in capillary blood. Two to three capillaries will be taken at each time point and examined with a blood gas analyzer (Montero et al.,
NT-proBNP as a prognostic factor and the determination of the correlation of VO2peak and NT-proBNP. NT-pro-BNP will be analyzed on a cobas system (Roche Dignostics, Rotkreuz, Switzerland).
Further disease-specific biomarkers such as renin, angiotensin-II, urocortin-2, osteopontin, soluble suppression of tumorigenicity-2, galectin-3, growth differentiation factor-15, copeptin, big-endothelin-1, placental growth factor/soluble Fms-like tyrosine-kinase-1, high-sensitivity C-reactive protein, interleukin-6, insulin-like growth factor-binding protein-7, irisin, glycocalyx components, matrix metalloproteinases, -activity and inhibitor, nitric oxide, total (anti-) oxidative capacities, myeloperoxidase, oxidized low-density lipoprotein, circulating immature and mature endothelial cells, and reticulocytes (see
All parameters will be analyzed in serum. Irisin and glycocalyx components such as syndecan-1, heparan sulfate and hyaluronan concentrations will be determined by enzyme-linked immunosorbent assay (ELISA) kits (Phoenix Pharmaceuticals Inc, Burlingame, USA; Diaclone Research, Besancon, France; Seikagaku, Tokyo, Japan; Fa. Cusabio Art.Nr.: CSB-E09585h; Echelon Biosciences, Salt Lake City, USA; respectively). Total (anti-) oxidative capacities as well as oxidized low-density lipoprotein will be determined using the Labor Diagnostika Nord (Nordhorn, Germany) assays. Parameters of endothelial matrix remodeling (matrix metalloproteinases, -activity, and inhibitor) and myeloperoxidase will also be determined by ELISAs (R&D Systems Europe, United Kingdom). MMP activity will be analyzed using Fluorogenic Peptide Substrate (R&D Systems Europe, United Kingdom). Circulating immature and mature endothelial cells and reticulocytes will be analyzed by flow cytometry (CytoFLEX, Beckman Coulter and ADVIA, Siemens, Switzerland, respectively).
Echocardiographic parameters of the left ventricular structure, systolic and diastolic function measured using a Full HD Color Doppler Ultrasound Scanner UF-890AG (Fukuda Denshi, Tokyo, Japan) by experienced echocardiographers according to recent international guidelines and standards (Miljkovik and Spiroska,
Peak arterial-to-venous oxygen content difference (Da-vO2) calculated using the Fick Principle (Peak Da-vO2 = VO2peak/peak cardiac output) (De Cort et al.,
Brachial-ankle PWV measured oscillometrically in the supine position at the right and the left upper arm and ankles with the VaSera VS-1500N Vascular Screening System (Fukuda Denshi Co. Ltd, Tokyo, Japan). Central PWV calculated applying the ARCSolver algorithm (Wassertheurer et al.,
NYHA functional class.
QoL assessed by the questionnaires SF-8 (Yiengprugsawan et al.,
Body composition will be analyzed by four-segment bioelectrical impedance analysis using the InBody 720 (Inbody Co. Ltd., Seoul, South Korea). Body mass index (BMI = weight/height2) and waist-to-hip ratio (WHR = waist circumference/hip circumference).
Physical Activity Level (PAL), number of daily steps and time spent at different walking speeds (min·day−1) will be measured using a waterproof micro-electromechanical triaxial accelerometer worn on the non-dominant wrist (GeneActiv, Activinsights Ltd, Kimbolton, Cambridgeshire, UK) to assess physical activity intensity (light, moderate, vigorous) and periods of inactivity, sleep and wake over 7 days for 24 h per day (Esliger et al.,
Muscular strength measured with a so-called Mid-Thigh-Pull Test. This isometric test is conceptualized in order to rate whole body force; the force of all extensors of the leg e.g., M. rectus femoris and of the back such as M. erector spinae as well as muscles of the forearm and hand. The test is comparable to elevate a table by hand. Two factors will be measured: the ability of a patient to generate maximum force (‘Peak Force') and the second factor is measuring the increase of force over time (Rate of Force Development) (Scott et al.,
Measurement of muscle oxygen saturation. Muscle oxygen saturation is measured with near-infrared spectroscopy (NIRS, Portamon, Artinis Medical Systems, Elst, The Netherlands). NIRS relies mainly on two characteristics of human tissue. First, the relative transparency of tissue to light in the NIR range, and second, the oxygenation-dependent light absorbing characteristics of Hb. By using a number of different wavelengths, the relative changes in Hb concentration can be displayed continuously and saturation, respectively, absorption can be measured. If the absorption is known, the Lambert-Beer law can be used to calculate the chromophore's absorption (Scholkmann et al.,
Cardiac output with cardiac impedance measurements using the Physioflow device (Physioflow, Manatec Biomedical, Poissy, France) (Endes et al.,
The measurement of flow-mediated dilation (FMD), a principle to measure the integrity of the endothelia. It refers to dilation of an artery when blood flow increases in that artery. The primary cause of FMD is release of nitric oxide by endothelial cells through shear stress. FMD of the brachial arteries provides a non-invasive alternative to other measurement procedures. To determine FMD, brachial artery dilation following a transient period of forearm ischemia is measured using ultrasound (UNEFEX 38G 3.0, UNEX Co., Nagoya, Japan) (Konigstein et al.,
Static and dynamic retinal vessel analysis using the retinal vessel analysis system (IMEDOS Systems, Jena, Germany) and a fundus camera (450 FF; Carl Zeiss, Jena, Germany). We will take three valid static images and two dynamic videos of the retina to quantify retinal microvascular structure and function. Conventional eye drops (Tropicamide 0.5%) will be used for pupil dilation of one eye. Retinal vessel diameters will be calculated as central retinal arteriolar and venular equivalents. Three flicker cycles are applied and averaged to calculate the flicker light-induced dilatation in %change relative to baseline. Details of the standardized procedures have been published previously (Streese et al.,
Carotid artery reactivity (CAR), a parameter non-invasively assessed by transcutaneous ultrasound, to examine endothelial function following sympathetic stimulation produced by the cold pressor test. Right carotid artery diameter is recorded before and during 90 s of immersion of the hand up to the wrist in ice water (4°C). Images will be obtained using a high-resolution ultrasound machine (UF-760AG, 5-12 MHz linear array transducer, Fukuda Denshi Co. Ltd., Tokyo, Japan) (Liao et al.,
Overview of blood biomarkers. Laboratory parameters reflecting biomarkers of chronic heart failure; CRP, C-reactive protein.
In both groups (MCT and HIT) adverse events (AE) and serious AE will be recorded. All potential cardiovascular events will be considered to be serious AEs (e.g., atrial and ventricular arrhythmia, unstable angina pectoris, clinical worsening of HF requiring hospitalization or intensified diuretic therapy, cardiovascular death) and will be immediately treated, recorded and/or discussed in detail to plan further procedures. If clinically indicated, laboratory parameters (kidney and liver values, International Normalized Ratio, hematology, etc.) will be measured and the patient's personal cardiologist will be informed to discuss further procedures.
Safety and adherence after end of the study will be checked during follow-up calls. Patients will be referred to adequate long-term rehabilitation settings (e.g., ambulant cardiovascular rehabilitation, KARAMBA; or the Swiss Physical Activity Promotion in Primary Care System, PAPRICA).
Patients will be randomly allocated to either the intervention or the control group in a 1:1 ratio stratified by gender. Randomization will be done using a computer-based system implemented in the Castor software (Castor, Amsterdam, Netherlands). The randomization procedure includes a minimization algorithm. The randomization procedure is implemented in the online electronic-data-capture system provided by Castor. Only authorized study personnel will have access to the randomization tool and can assign a treatment to a new study patient. Due to the minimization algorithm, the allocation sequence is totally concealed until randomization is carried out.
It is not possible to blind patients or investigators performing the randomization and training sessions. Therefore, un-blinding procedures are not applicable. However, data analysis will be performed by investigators blinded for patient data and allocation to HIT, respectively, MCT.
All patients will perform a 12-week training program 3 times per week on a bicycle ergometer (Ergoselect 200, Ergoline, Bitz, Germany),
Overview of the endurance-strength training protocols. Training protocol for HIT and MCT during the 12-week intervention period. W-up, warm-up; HRpeak, heart rate peak; 5RM, 5-repetition maximum; RT, resistance training; HIT, high-intensity interval training, c-down, cool-down; MCT, moderate continuous training.
During the 12-week intervention period, we will use an adapted version of the classical 4 x 4 min protocol of Wisloff et al. (
Patients perform the same strength exercises and then cycle continuously over a period of 12 weeks, increasing from initial 17 to 26 min from week 1 to 8 and 44 min from week 9 to 12 at 60–70% of the HRpeak.
Workload of the bicycle ergometer will be continuously adjusted in order to maintain a constant relative exercise intensity throughout the 12-week training period according to the protocol (Wisloff et al.,
Based on the current epidemiological situation and regulations in Switzerland, it seems very likely to be able to offer continued training during another COVID-lockdown.
Good compliance will be ensured by friendly research staff, motivating training units and improving health outcomes, such as better QoL. In case a patient will miss a training session, it will be rescheduled, if possible. A compliance of 80% (29 sessions) will be required to be considered in the analysis of the per-protocol set (PPS).
Concomitant treatments or medication considered necessary by the treating physicians are permitted. However, if the ability to exercise is affected in case of worsening HF (e.g., cardiac decompensation, NYHA IV) or any new diagnosis of concomitant disease (e.g., cancer, pulmonary hypertension, trauma, and so on), patients will be excluded from the study and will not be allowed to resume the study procedures at a subsequent date.
We use a closed form formula to assess the sample size, as implemented in the R function “power.t.test()”. This provides us a sample size for comparison of two independent groups. Borm et al. (
Under these assumptions, with a 1:1 treatment allocation ratio, 86 patients need to be recruited in order to show a minimal clinically relevant difference of 2 ml min−1 kg−1, while maintaining power for the PPS with 77 evaluable patients.
The full analysis set (FAS) will include all patients randomized to the trial, regardless of compliance. The PPS will include all patients from the FAS, who had a 12-week post-intervention visit and who complied with their assigned treatment.
The primary analysis will be done on the FAS and based on the intention to treat principle. Patients will be analyzed according to the group to which they were assigned. Missing values in the primary endpoint will be imputed using multiple imputations via chained equations based on baseline characteristics. An ANCOVA approach will be taken: a linear regression model will be used to model patient VO2peak at post-intervention as outcome, with the baseline value as an independent covariate. As an explorative analysis, the association of tHb-mass and patient type, as well as their combined effects on VO2peak will be examined using multiple regression models or other methods for estimating causal effects. As a safety analysis, the number and type of AEs will be summarized and reported per group. In addition, the number of non-compliant patients and their baseline characteristics compared between treatment groups will be summarized.
Confounding variables such as baseline patient characteristics (e.g., age, BMI, smoking/drinking), comorbidities, medication, and baseline fitness level will be accounted for during regression analyses.
Data will be anonymized and stored in a pseudonymized manner by using the patient IDs. Personal data and contact information for the follow-up phone calls and assessment of physical activity will be entered into a separate excel file with restricted access for staff performing these calls. Collection and management of the clinical trial data will be done using an electronic data capture (EDC) system based on the software Castor. Source data, randomization and pseudonymization lists will be kept under lock and key. Password protection ensures that only authorized persons can enter the system to view, add or edit data according to their permissions. The EDC system will be locked after all data have been monitored and all raised queries have been resolved. Data will be archived at the DSBG. Independently from investigators, regulatory authorities can audit this trial. Direct access to source documents will be permitted for purposes of monitoring, audits and inspections. All involved parties must keep the participant data strictly confidential. Any results of this study will be published in an anonymized manner. The study protocol and dataset shall be accessible to any regulatory authority after publication for at least 10 years.
Heart failure, in particular heart failure with preserved ejection fraction (HFpEF) constitutes a large and increasing burden to the national and international healthcare systems. Nevertheless, the etiology of HFpEF has not yet been fully understood and symptoms are often misdiagnosed. Since mortality in patients with HFpEF remains unacceptably high with a 5-year survival rate of only 30%, new therapeutic strategies are urgently needed (Burkhoff,
Our analyses of biomarkers reflect the most important pathogenetic pathways, will provide further important insights into the pathophysiology of HFpEF, and might be helpful in predicting the response to exercise therapy in order to effectively guide risk stratification as well as preventive and therapeutic measures.
The trial will be conducted according to the protocol version 6 of 20th of June 2020. Patient recruitment started in September 2020 and is expected to run consecutively until November 2022. Data collection of the intervention phase is expected to be completed in February 2023 and data analysis in May 2023. Data collection of follow-up calls (6 months, 1, 2, and 3 years) will be completed in August 2023, February 2024, February 2025, and February 2026.
The original contributions presented in the study are included in the article, further inquiries can be directed to the corresponding author.
Ethics approval to conduct this trial has been granted by the local Ethics Committee for the Region of North-western and Central Switzerland (EKNZ, Project-ID: 2019-00188). The latest amendment to the study protocol (version 6) has been approved by the EKNZ on 20th of June 2020. The trial will meet the criteria and principles of the Declaration of Helsinki and has been registered in the clinicaltrials.gov database (Trial registration number: NCT03184311, Registered 9th of June 2017).
BG, JK, RS, MB, SB-Z, ET, JL, HH, AS-T, and TD were broadly involved in the conception and design of the study and drafted the manuscript. Furthermore, BG, JK, and RS will be responsible for the logistic preparation and protocol-conform implementation of the study, the recruitment of patients, and the performance of training sessions. BG, JK, RS, AS-T, and TD critically reviewed the manuscript. The co-principal investigators of this study are AS-T and TD. TD and JL initiated the trial in collaboration with MB, SB-Z, ET, and AS-T. All authors have read and approved the final version of the manuscript and gave their consent for publishing this study protocol.
This study is financially supported by the Swiss National Science Foundation (SNSF project No. 185217).
SB-Z is an employee of the funding institution (SNSF). However, her contribution to the study took place before her current employment at SNSF. The funder (SNSF) did not have any role in the study design, decision to publish, or preparation of the manuscript. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
We thank Mr. Jonas Mundwiler, and Dr. Benita Janisch for their valuable inputs to the study protocol as well as Dr. Michael Coslovsky who was responsible for the design of the statistical analysis plan.