Authors: Michael J. Plunkett, Ana Luiza C. Sayegh, Tanya J. McWilliams, Sasiharan Sithamparanathan, Julian F.R. Paton, James P. Fisher
Categories: Research Letters, 7
Source: The European Respiratory Journal
Impairment of exercise capacity, predominantly limited symptomatically by dyspnoea [1], affects most patients with pulmonary arterial hypertension (PAH) despite current therapies [2], with significant implication for patients, adversely impairing health-related quality of life [3] and clinical prognosis [4]. However, the underpinning physiological mechanisms behind dyspnoea and exercise limitation remain incompletely understood. Skeletal muscle metabolic and microcirculatory deficits are present in PAH [2], and likely lead to earlier and more pronounced accumulation of metabolites during exercise. We hypothesised that this would augment the activation of group III/IV afferents responsive to metabolites present in exercising limb muscles (i.e. muscle metaboreflex), and provide a novel driver for hyperventilation [5], pulmonary arterial pressure [6] and sensations of dyspnoea [7] in PAH.
To the Editor:
Impairment of exercise capacity, predominantly limited symptomatically by dyspnoea [1], affects most patients with pulmonary arterial hypertension (PAH) despite current therapies [2], with significant implication for patients, adversely impairing health-related quality of life [3] and clinical prognosis [4]. However, the underpinning physiological mechanisms behind dyspnoea and exercise limitation remain incompletely understood. Skeletal muscle metabolic and microcirculatory deficits are present in PAH [2], and likely lead to earlier and more pronounced accumulation of metabolites during exercise. We hypothesised that this would augment the activation of group III/IV afferents responsive to metabolites present in exercising limb muscles (i.e. muscle metaboreflex), and provide a novel driver for hyperventilation [5], pulmonary arterial pressure [6] and sensations of dyspnoea [7] in PAH.
Ventilation, pulmonary artery pressure and perception of dyspnoea were determined during isolated muscle metaboreflex activation (MMA) in 14 PAH patients (nine females, mean±sd age 48.4±12.8 years) recruited, following provision of written informed consent, from the Greenlane Pulmonary Vascular Service, Auckland, New Zealand. Their responses were compared to 14 age- and sex-matched healthy controls (nine females, mean±sd age 49.4±14.2 years). Participants were free of other comorbidities that may limit exercise capacity or cause exertional dyspnoea. This study was approved by the Health and Disability Ethics Committee (HDEC 2022 FULL 12454), prospectively registered (Australia New Zealand Clinical Trials Registry, ACTRN12622000493741), and conducted according to the Declaration of Helsinki (2013).
MMA was achieved using a standard technique of post-exercise circulatory occlusion (PECO) following handgrip exercise, which involves inflation of a cuff on the exercising arm to supra-systolic pressures prior to cessation of handgrip, thereby trapping metabolites within the muscle during recovery [8]. Participants performed two trials of 8 min (MMA and control), in a random order. Each trial consisted of a 2-min baseline period, followed by 2 min of right-sided static handgrip (handgrip dynamometer; ADInstruments, Bella Vista, Australia) to 35% of maximal voluntary contraction, followed by 4 min of rest (final 2 min termed recovery period). In one trial (MMA trial) PECO was performed for the first 2 min of recovery (MMA period). In the control trial participants underwent recovery without MMA, with the first 2 min termed the “free-flow” period.
Minute ventilation (V̇E) and end-tidal gas partial pressures were measured with a pneumotachometer (Hans Rudolph, Shawnee, KS, USA) and gas analyser (ADInstruments) respectively, connected via a mouthpiece (with nose-clip). Subjective dyspnoea ratings for each trial period were assessed using the Borg 0–10 dyspnoea scale [9]. Transthoracic echocardiographic (Vivid S70, GE Healthcare, Chicago, IL, USA) assessment of right ventricular outflow tract (RVOT) acceleration time (AT), and RVOT velocity time integral (VTI) were performed and mean pulmonary artery pressure (mPAP) estimated from AT using Kitatabake's equation [10]. Assuming a constant RVOT cross-sectional area, percent changes from baseline right ventricular stroke volume were estimated from RVOT-VTI [11], and percent changes in cardiac output were estimated by multiplying RVOT-VTI by heart rate. Mean responses from baseline, for each variable, were calculated for the second minute of each trial period.
Time series data were examined using mixed models for repeated measures analysis. Normality was verified by visualisation. Where appropriate, multiple comparison post hoc analyses were performed with unpaired t-test and Bonferroni correction. Data are expressed as mean±sd. Statistical significance was defined as p<0.05. Statistical analysis was performed using SPSS version 29 (IBM, Armonk, NY, USA).
Of PAH aetiologies, nine were idiopathic, three connective tissue disease-associated, one drug-induced and one congenital heart disease-associated. World Health Organization functional class of PAH patients were I (n=3), II (n=8) and III (n=3). Mean parameters on most recent right heart catheterisation were mPAP 42.9±13.3 mmHg, pulmonary vascular resistance (PVR) 5.5±3.2 WU and cardiac output 6.4±1.6 L·min^−1^. Three patients were on monotherapy (one on phosphodiesterase type 5 inhibitor (PDE5i), two on calcium channel blockers (CCB)), eight on dual-combination therapy (PDE5i and endothelin receptor antagonist, alongside CCB in one), and three on triple-combination therapy with prostacyclin analogues.
MMA stimulated an excess ventilatory response with a four-fold larger increase in ventilation in PAH compared to healthy controls (ΔV̇E 2.23±1.90 versus 0.42±1.36 L·min^−1^, respectively; p=0.012) (figure 1a). While baseline end-tidal carbon dioxide tension (PETCO2~~) was lower in PAH (37±4.9 versus 41±3.3 mmHg respectively), there was also a more pronounced decline with MMA in PAH (ΔPETCO2~~ −2.43±1.98 versus −1.23±1.15 mmHg, respectively; group p=0.030). In contrast, without MMA, no between group difference was observed, with similar V̇E recovery to baseline in PAH and healthy controls occurring during the free-flow period (ΔV̇E 0.51±1.29 versus 0.85±1.55 L·min^−1^, respectively; group p=0.743) (figure 1a), and no difference in free-flow period PETCO2~~ responses (ΔPETCO2~~ −1.01±0.96 versus −0.52±0.67 mmHg, respectively; group p=0.469). Accordingly, the MMA ventilatory response (determined for each individual as the ventilatory response from baseline of the MMA period, minus the ventilatory response during the corresponding free-flow period of the control trial) was higher in PAH versus controls (1.72±1.42 versus −0.434±1.42 L·min^−1^, respectively; unpaired t-test p<0.001) (figure 1d).
MMA sustained the handgrip exercise-induced increase in mPAP in PAH but not healthy controls (ΔmPAP 10.34±9.49 versus 0.40±5.02% of baseline, respectively; group p=0.017) (figure 1b). This seemingly occurred through increased PVR, rather than a flow-mediated effect, as cardiac output recovered to baseline during MMA. Without MMA during free-flow recovery, mPAP returned to baseline similarly in PAH and healthy controls (figure 1b). The MMA mPAP response (calculated analogous to the ventilatory response) was numerically greater in PAH than healthy controls (11.51±12.84 versus 3.03±8.99%, respectively; unpaired t-test p=0.077) (figure 1d).
The impact of MMA on dyspnoea perception in PAH was marked, with doubling of Borg dyspnoea score responses compared to healthy controls (ΔBorg 1.89±1.21 versus 0.81±1.13 units, respectively; p<0.001) (figure 1c), with a clinically important mean difference of 1.08 units [12]. In PAH, the mean Borg dyspnoea score reported during handgrip from both trials was positively correlated with the magnitude of MMA ventilatory response (Pearson correlation coefficient r=0.581, p=0.029), though not with the MMA mPAP response (r=0.404, p=0.247).
Our results further elucidate the physiological mechanisms of dyspnoea in PAH, demonstrating the role of skeletal muscle metaboreflex activation. Classically, the predominant factors limiting exercise are understood to be impaired cardiac output response to exercise due to high right ventricular afterload, and excess ventilation stimulated primarily by high physiological dead-space ratio (VD/VT) and reduced arterial partial pressure of CO2 (PaCO2~~), causing sensations of dyspnoea [13], often exacerbated by impaired breathing mechanics augmenting the work of breathing [14]. We demonstrate the metaboreflex to be an additional driver of excess ventilation, and enhanced perception of dyspnoea [13]. Furthermore, by exacerbating already high PVR and right ventricular afterload, thereby limiting cardiac output, skeletal muscle underperfusion may result, potentially further heightening skeletal muscle dysfunction and fatigue, metabolite build-up and thus metaboreflex responses in a positive feedback loop.
A limitation of this study is use of non-invasive echocardiography rather than the gold standard right heart catheterisation. RVOT AT was used instead of tricuspid regurgitant velocity to estimate mPAP due to superior recoverability and correlation to invasively measured mPAP during exercise, as previously reported [15]. We recognise that PETCO2~~ may further dissociate from PaCO2~~ in PAH, and as PaCO2~~ was not measured in this study, we cannot definitely conclude the augmented ventilatory response to metaboreflex activation is excessive relative to CO2 production (i.e. lowers PaCO2~~). Additionally, the metabolic environment of the muscle interstitium during exercise was not measured, and thus we are unable to elucidate the mechanisms by which the ventilatory response to metaboreflex activation is augmented in PAH, whether by greater metabolite accumulation, higher metaboreflex sensitivity (peripherally or centrally), or a combination of these.
In summary, the novel findings of this study are that metaboreflex activation evokes exaggerated ventilatory and pulmonary vascular responses in PAH, and causes clinically meaningful increases in ratings of perceived dyspnoea. This study raises the possibility that metaboreflex afferents represent a novel physiological mechanism to target (e.g. pharmacologically or by exercise training) for improving exertional dyspnoea in PAH.
The authors wish to thank all the volunteers for their valuable participation in this study; Kelly Boegel, Nicola Edwards, Jonathan Mok and Dean Thomas of the Department of Cardiology, Te Toka Tumai Auckland, Te Whatu Ora Health New Zealand for support with echocardiography; and the Department of Respiratory Medicine, Te Toka Tumai Auckland, Te Whatu Ora, Health New Zealand, for their kind support.