Breathing Exercise Against Dyspnoea in Heart Failure Patients to Improve Chemosensitivity and Ventilatory Efficiency - a Randomized Controlled Single-centre Trial
Breathing Exercise Against Dyspnoea in Heart Failure Patients to Improve Chemosensitivity and Ventilatory Efficiency - a Randomized Controlled Single-centre Trial
An exaggerated ventilatory response (minute ventilation, V̇E) to exercise relative to exhaled carbon dioxide (V̇CO2) is common in heart failure (HF) patients with reduced as well as preserved left ventricular ejection fraction (HFrEF, HFpEF). Severity of this exaggerated response is associated with poor prognosis. This response may be triggered by pulmonary congestion and peripheral muscle myopathy. A vicious circle is fuelled by hypersensitivity of chemoreceptors to hypercapnia and sympathetic nervous hyperactivity, resulting in hyperventilation (low PaCO2). Low PaCO2 is predictive of mortality in these patients. PaCO2 can be increased acutely, e.g. by apnoea. Also, nasal breathing has been found to reduce the V̇E/V̇CO2 slope during exercise compared to oral breathing. Three previous slow breathing studies in HFrEF patients have had encouraging results with regard to reducing sympathetic activity, reflected in lowered arterial (pulmonary) blood pressure and increased EF. The investigators hypothesise that a 12-week training with nasal slow breathing followed by end-expiratory apnoea based on education, centre-based introduction and home-based 15 min/day breathing training will be effective at reducing the exaggerated ventilatory response to exercise. A total of 68 patients with stable HF seen at the HF clinics of the Inselspital (34 HFrEF, 34 HFpEF) will be randomised to the breathing intervention or usual care. Primary outcome will be V̇E/V̇CO2 slope at 12 weeks. If breathing training successfully ameliorates the exaggerated ventilatory response and perception of dyspnea during exercise, it offers an attractive tele-health based add-on therapy that may add to or even amplify the beneficial effects of exercise training.
BACKGROUND
Ventilatory inefficiency, most commonly quantified as an increased ventilation (V̇E) to carbon dioxide exhalation (V̇CO2) slope during exercise, is a landmark of heart failure patients both with reduced and preserved ejection fraction (HFrEF, HFpEF).[1] Numerous studies have found higher V̇E/V̇CO2 slopes to be associated with poorer prognosis.[2-4] The components of the V̇E/V̇CO2 slope are the arterial CO2 partial pressure (PaCO2), that is affected by hyperventilation, and the pulmonary dead space/tidal volume ratio (VD/VT) that is affected by pulmonary perfusion abnormalities.[5] The exaggerated response in ventilation of HFrEF patients may be caused by hypersensitivity of chemoreceptors to CO2,[6] and/or a sympathetic nervous hyperactivity commonly found in HFrEF patients, based on an increased activation of metaboreceptors in peripheral muscles response to increased anaerobic metabolism.[7] Chronic sympathetic nervous hyperactivity has been suggested to decrease aerobic capacity of skeletal muscles based on reduced capillarisation[8] and reduced red blood cell flux[9] leading to a shift in muscle fibre type towards a lower content on type I fibres.[10] The ensuing anaerobic muscle metabolism leads to increased muscle fatiguability[11] and acidosis already at low levels of exercise, which trigger exaggerated responses in ventilation.[12] Hyperventilation, on the other hand, is well known to stimulate sympathetic nervous activity, and so the vicious circle of sympathetic nervous activity driving hyperventilation and hyperventilation activating sympathetic nervous activity continues.[13] This suggests that hyperventilation may not only be a consequence of poor left ventricular (LV) function, but also a driver.
Besides pharmaceutical therapies and electrophysiological interventions, exercise therapy has been found to have beneficial effects on hemodynamic and ventilatory parameters in HFrEF[14] and HFpEF patients alike.[15] The main mechanisms of exercise are thought to be reduced peripheral resistance and hence cardiac afterload by improvement of endothelial function, increased capillarisation leading to improved oxygenation of skeletal muscles and improved aerobic metabolism.[16] Despite the beneficial effects of exercise training in both, centre-based and home-based settings,[17, 18] adherence to physical activity has been found to be poor amongst HFrEF patients.[19] Surprisingly, few studies have targeted ventilation directly with therapeutic approaches. Only three studies have assessed the effects of slow-breathing training on cardiorespiratory function.[20, 21] These studies found improved physical function, reduced blood and pulmonary arterial pressure, increased ejection fraction (EF),[20, 22] improved ventilatory efficiency[20] and reduced sleep apnoea.[22] Further, they found improved regulation of the autonomic nervous system by reducing sympathetic drive and increasing vagal activity.[23] It is unknown whether slow breathing may increase PaCO2 sufficiently to change the sensitivity or set point of chemoreceptors. On the other hand, apnoea training has been found to lead to large changes in PaCO2 levels tolerated by chemoreceptors at rest and during exercise.[24, 25] However, to date there are no published studies that have implemented apnoea into a breathing training in HF patients. Further, previous studies have not investigated whether the effect of slow breathing on improving the V̇E/V̇CO2 slope was due to a chronic increase in PaCO2 or a decrease in ventilatory dead space.
HYPOTHESIS
The investigators hypothesise that a 12-week training with nasal slow breathing followed by end-expiratory apnoea based on education, centre-based introduction and home-based 15 min/day breathing training will be effective at reducing the exaggerated ventilatory response to exercise.
METHODS
Study design
Prospective randomised controlled study. Eligible patients are identified during their yearly check-up at the Heart Failure Clinic and Preventive Cardiology of the Inselspital in Bern. Patients will be randomised 1:1 (stratified for HFrEF/HFpEF and sex) to an intervention and control group. Patients in the intervention group perform the breathing training additionally to standard care and those in the control group receive standard care and are offered the breathing training after the end of the study. The study design and breathing intervention have been developed with direct input by a patient group (from pilot study).
In an additional cross-sectional substudy, the same measurements of the RCT are performed in a group of 15 patients after acute or chronic coronary syndrome (ACS/CCS) with inefficient ventilation, 15 healthy age-matched and 15 healthy young controls. Further, the cross-sectional substudy compares pulmonary gas exchange and breathing patterns during 5 min of oral versus 5 min of nasal breathing during submaximal continuous exercise in the HF, ACS/CCS, old healthy, and young healthy groups.
Breathing intervention
The respiratory pattern modulation training is performed at home for 12 weeks twice daily for 15 min per session and consists of three components: 1) education on abnormal ventilation in heart failure, the effect of ventilation on PaCO2 and the autonomous nervous system, and chemoreceptor sensitivity; 2) 1-3 sessions of guided and monitored face-to-face training with slow nasal abdominal breathing and intermittent apnoea supported by the Healer vest (L.I.F.E., Milan, Italy) measuring electrocardiogram (ECG), and chest excursions at the level of the xiphoid, thoracic manubrium, and abdomen; 3) independent home-based apnoea training supported by hand-outs, videos and weekly phone calls to monitor progress and adherence, answer questions and encourage further progression with duration of breath-hold.
Measurements
Measurements are performed during visit 1 before and visit 2 at the end of the intervention period.
Cardiopulmonary exercise testing (CPET)
CPETs are performed on a cycle ergometer according to the recommendations of the American Heart Association.[38] Ramp tests are performed as previously described.[31] O2 consumption and CO2 production will be measured continuously in an open spirometric system (Quark, Cosmed, Rome, Italy) and registered as average values over 8 breaths. Every 2 min, patients are asked about their perception of dyspnoea on the modified Borg scale. V̇E/V̇CO2 slope from rest to ventilatory threshold 2 (VT2), peak V̇O2 and V̇O2 at VT1 are determined as previously described.[31]
Blood analyses
Blood samples are obtained from the antecubital vein for analysis of haemoglobin and NT-proBNP. Arterialized blood are extracted from the ear lobe at rest and peak exercise for analysis of PaCO2, oxygen (PaO2), bicarbonate and pH.
Sensitivity of chemoreceptors
The sensitivity of chemoreceptors is measured by a rebreathing protocol.[39] The subjects are resting supine and breathe through a mouthpiece of an open spirometric system (Innocor, Cosmed, Rome, Italy). With the 3-way-valve open to room air, the test begins with 2-5 min of hyperventilation, allowing end-tidal partial CO2 pressure (PETCO2) to drop. Following hyperventilation, the subject breathes comfortably, while the 3-way-valve is switched to the rebreathing bag. Equilibration of PCO2 in bag, lungs, and arterial blood to mixed venous blood is achieved by taking three deep breaths. During the following minutes, PETCO2 is allowed to rise, while PETO2 is clamped at 150 mmHg during hyperoxic testing, and at 50 mmHg during a second, hypoxic test run by feeding 100% O2 into the circuit by a port at the rebreathing bag. Central and peripheral chemoreflex responses to CO2 are estimated by the difference between hyperoxic and hypoxic ventilatory response.[40, 41]
Patient reported outcomes
The Kansas City Cardiomyopathy Questionnaire (KCCQ) are filled in during visit 1 and visit 2 to assess quality of life and dyspnoea. During visit 2, a structured interview is performed with the patient to assess feasibility and barriers with the breathing training. Adherence to training is monitored based on verbal information by the patients during the weekly phone calls.
Heart rate variability (HRV) and breathing frequency (BF)
HRV is measured from 24-hour ECG recorded with the Healer vest (L.I.F.E., Milan, Italy) and analysed from a segment during a deep sleep phase as previously described by the investigators' group.[42] Low-frequency power (LF, ms2, 0.04-0.15 Hz), high-frequency power (HF, ms2, 0.15-0.4 Hz), and the LF/HF are analysed [43]. BF is measured by strain gauges from Healer vest.
Submaximal tests of cross-sectional substudy
After a 15 min break following the CPET ramp test, patients cycle 5 min at 50% of peak power output as assessed during the ramp test with exclusively nasal, and 5 min with exclusively oral breathing in randomised order, with a 15 min break between the two tests. Pulmonary gas exchange parameters and breathing pattern are measured and compared between the two breathing modes and between the four groups (HF, ACS/CCS, old healthy, young healthy).
OUTCOMES
Primary outcome is V̇E/V̇CO2 slope analysed by ANCOVA with repeated measures corrected for baseline values and EF and sex.
Secondary outcomes are the nadir of the V̇E/V̇CO2 ratio, breathing pattern, VD/VT, peak V̇O2, V̇O2 at VT1, resting PETCO2, peripheral and central chemoreceptor sensitivity, arterial blood gases, NT-proBNP, heart rate, HRV, ventricular premature beats from 24-hour ECG, KCCQ, feasibility and adherence.
Outcomes of cross-sectional study are VE/VCO2 ratio, VE, VO2, BF, VT, PetCO2, PetO2, O2/HR and rapid shallow breathing index between nasal and oral breathing.
REFERENCES
31. Marcin T, Trachsel LD, Dysli M, et al. Effect of self-tailored high-intensity interval training versus moderate-intensity continuous exercise on cardiorespiratory fitness after myocardial infarction: A randomized controlled trial. Ann Phys Rehabil Med 2021: 101490.
38. Fletcher GF, Ades PA, Kligfield P, et al. Exercise standards for testing and training: a scientific statement from the American Heart Association. Circulation 2013; 128(8): 873-934.
39. Duffin J. Measuring the respiratory chemoreflexes in humans. Respir Physiol Neurobiol 2011; 177(2): 71-9.
40. Duffin J, Mohan RM, Vasiliou P, et al. A model of the chemoreflex control of breathing in humans: model parameters measurement. Respir Physiol 2000; 120(1): 13-26.
41. Guyenet PG. Regulation of breathing and autonomic outflows by chemoreceptors. Compr Physiol 2014; 4(4): 1511-62.
42. Herzig D, Eser P, Omlin X, et al. Reproducibility of Heart Rate Variability Is Parameter and Sleep Stage Dependent. Frontiers in physiology 2017; 8: 1100.
43. Heart rate variability. Standards of measurement, physiological interpretation, and clinical use. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. European heart journal 1996; 17(3): 354-81.
Inclusion Criteria:
Exclusion Criteria: