Relevance of differential control of sympathorespiratory response magnitudes in clinical assessments

The peripheral chemoreflex is activated by hypoxaemia and hypercapnia, which is sensed mainly by the carotid bodies (CB) located near the carotid artery bifurcation. CB afferents have widespread synaptic connections within the brainstem, allowing for the peripheral chemoreflex to drive multiple efferent outputs, including respiratory motor and sympathetic nerve activity (Zoccal et al. 2014) (Fig. 1). The magnitude of response to a given level of hypoxia is often used to quantify peripheral chemoreflex sensitivity and displays a large interindividual variability. Whether a single output is a valid surrogate measure of other outputs downstream of the CB is unclear. Previous work in young healthy humans demonstrated an uncoupling between ventilatory and sympathetic response magnitude to chemoreflex activation (Keir et al. 2019). Whether this is also relevant in a clinical population is unknown. Given that CB hypersensitivity/tonicity may underlie sympathetic overdrive in cardiovascular disease (Zoccal et al. 2014), understanding whether chemoreflex-driven sympathetic responses can be accurately inferred through the more clinically accessible ventilatory response measurement will have important implications. The peripheral chemoreceptor afferents (light green arrows) input to the caudal nucleus tractus solitarius (NTS) and elicit an excitatory output to the ventral respiratory column (VRC) and rostral ventrolateral medulla (RVLM). The arterial baroreceptors (light blue arrows) and slow-adapting pulmonary stretch receptors (purple arrows) input to the intermediate NTS and elicit an inhibitory influence on the VRC (not shown) and the RVLM. Arterial baroreceptors are known to induce this RVLM inhibition via activation of the caudal ventrolateral medulla (CVLM), whereas the neural circuitry for the pulmonary stretch receptors remains unclear, and so for simplicity, a direct inhibition on the RVLM has been proposed. During chemoreflex stress, the predominant influence on the VRC is understood to be shifted to the peripheral chemoreceptor afferents, and thus the inhibitory influences on the VRC are not shown in the figure. Green boxes indicate points of afferent convergence/integration, lending potential for differential control of ventilatory and sympathetic responses (see text for detail). Arrow heads depict excitatory neural inputs and flat tops depict inhibitory neural inputs. In a recent study published in The Journal of Physiology, Prasad and collaborators (Prasad et al. 2020) quantified the relationship between ventilatory and sympathetic responses to (1) four 1 min bouts of hyperoxia (100% O2) separated by 3 min of room air breathing, and (2) three 3 min stages of eucapnic hypoxia (O2 and CO2 gas content manipulated to achieve arterial oxygen saturations () of 90, 85 and 80% and end-tidal partial pressures of CO2 () matched to baseline). Ventilation was recorded with concurrent muscle sympathetic nerve activity (MSNA) in 61 obstructive sleep apnoea (OSA) patients treated with continuous positive airway pressure (an additional 21 participants (82 total) were tested in the hypoxia trial without MSNA). Cardiovascular (heart rate, , blood pressure (BP)), respiratory (airflow, end-tidal and ), and sympathetic (MSNA via fibular nerve microneurography) variables were measured throughout. Ventilatory responses to hyperoxia were quantified as the mean of four single breath nadirs, with only one nadir breath being used from each of the four 1 min bouts. Similarly, to quantify the sympathetic responses to hyperoxia, the mean 15 s nadir from each of the four 1 min hyperoxic bouts was computed. Ventilatory and sympathetic responses to eucapnic hypoxia were quantified as the linear slope across four 3 min mean values: resting baseline and of 90, 85 and 80%. The hyperoxia test was intended to evaluate the contribution of the CB to resting ventilatory and sympathetic activity, as short term hyperoxia inhibits the peripheral chemoreflex. Therefore, hyperoxic reductions in ventilation and sympathetic activity implies a peripheral chemoreceptor tonicity to their outputs at rest. The eucapnic hypoxia test was intended to quantify the sensitivity of the peripheral chemoreflex. The authors observed a decrease in min ventilation (; −3 L/min; −35%) and MSNA burst frequency (−10 bursts/min; −42%) during hyperoxia (BP response not measured). The peak response to eucapnic hypoxia ( 80%) was quantified as an increase in (+7 L/min; +83%), MSNA burst frequency (+12 bursts/min; +55%), and mean arterial pressure (MAP; +4 mmHg; 4%). and MSNA response magnitudes were not related during hyperoxia or eucapnic hypoxia, though moderate-to-good correlations were observed between ventilatory and blood pressure responses to hypoxia ( r = 0.50; tidal volume, r = 0.61). To assess response reproducibility, a subgroup of participants (n = 17) were re-tested 6 weeks later. In calculating the coefficient of variation of the method error (CVME), the authors reported good reproducibility (≤24%) in hyperoxic and MSNA response and moderate reproducibility (25–39%) in eucapnic hypoxic and MSNA response. Collectively, the findings demonstrate a lack of relationship between ventilatory and sympathetic response magnitudes to hyperoxia and eucapnic hypoxia, which calls for ventilatory and sympathetic outputs of the peripheral chemoreflex to be specifically evaluated. The commendable sample size (n = 61) with concurrent ventilation and MSNA recordings permitted this important correlation analysis. In a subset of participants (n = 17), the good reproducibility of the hyperoxic test led the authors to recommend the use of acute hyperoxia testing to quantify the degree to which basal levels of ventilation and sympathetic activity are driven by peripheral chemoreceptors. This study advances our understanding of differential control of ventilation and sympathetic activity and highlights important clinical implications. Absent relationships between ventilatory and MSNA response magnitudes highlight that respiratory and sympathetic outputs can be differentially controlled (e.g. phrenic nerve activity can respond with a different magnitude than sympathetic nerve activity). Differential control may be due to heterogeneous afferent innervation of autonomic brainstem nuclei. Both respiratory premotor neurons within the ventral respiratory column (VRC) and sympathetic premotor neurons within the rostral ventrolateral medulla (RVLM) are activated by peripheral chemoreceptor afferents and inhibited by arterial baroreceptor and pulmonary stretch receptor afferents (Zoccal et al. 2014). During a chemoreflex stress in humans, control of ventilation likely shifts to being more predominantly influenced by the peripheral chemoreceptors, with the inhibitory afferents having a far lesser weighted influence. In contrast, sympathetic activity, though increased during chemoreflex stress via the peripheral chemoreceptor afferents, still demonstrates strong inhibitory regulation via the arterial baroreceptor and pulmonary stretch receptor afferents (Keir et al. 2019). This differential organization provides a logical neural substrate for differential control of ventilation and sympathetic activity. Indeed, in human testing, administration of a single stressor can rarely isolate a single afferent pathway. In the present study, the authors report eucapnic hypoxia to elicit variable changes in MAP (range: −8 to +23 mmHg) and ventilatory parameters (tidal volume: +0.05 to 2.78 L/breath; : +0.1 to +27.8 L/min; see the Supporting information), indicative of variable levels of sympathoinhibitory feedback from the arterial baroreceptor and pulmonary stretch receptor afferents, respectively. Thus, sympathetic responses to hypoxic stressors are not solely due to peripheral chemoreceptor afferents, but the integrative response to multiple afferent inputs. An awareness of differential control of ventilation and sympathetic activity when interpreting clinical assessments is imperative. The use of hypoxia challenges as a clinically relevant assessment tool to determine peripheral chemoreflex sensitivity/tonicity, particularly in patients with cardiovascular disease, is growing. Sympathetic overactivity is a hallmark characteristic of cardiovascular disease and augmented peripheral chemoreceptor sensitivity/tonicity is one mechanism proposed to underlie it (Zoccal et al. 2014). This has led to the ongoing development of CB resection surgery as a therapeutic approach to reduce sympathetic overactivity. However, it is unlikely that peripheral chemoreceptors contribute to sympathetic activation in all patients, which calls for careful eligibility screening. In 2016, safety and feasibility of unilateral CB resection was trialled in 15 patients with resistant hypertension (Narkiewicz et al. 2016). Eight patients were denoted as ‘responders’ based on an arterial BP drop ≥10 mmHg at 3-month follow-up. A baseline comparison of responders vs. non-responders found responders had higher hypoxic ventilatory responses, suggesting the use of ventilatory responses to pre-select patients for CB modulation therapy. In 2017, unilateral (n = 4) and bilateral (n = 6) CB resections were performed to reduce sympathetic activity in patients with systolic heart failure (Niewinski et al. 2017). Indeed, part of the inclusion criteria was augmented peripheral chemosensitivity (>0.6 L/min/), expressed as the hypoxic ventilatory response. The findings from Prasad et al. (2020) and others Keir et al. (2019) challenge the assumption that increased ventilatory responses to hypoxia reflects exaggerated peripheral chemoreflex-mediated sympathoexcitation. These results strongly argue that ventilatory response magnitudes should not be used to infer sympathetic response magnitudes. Given the relatively large sample size (n = 61) of the present study, these findings should be weighted greater than the responder/non-responder analysis by Narkiewicz et al (2016). The study by Prasad and colleagues (2020) is not without important methodological considerations. Most importantly, the authors recommend the use of acute hyperoxia as a reliable diagnostic of tonic sympatho-respiratory modulation by the peripheral chemoreceptors. However, the authors compared single breath and 15 s nadir responses in ventilation and MSNA, respectively, to a 3 min baseline mean. Quantifying the hyperoxic response to single breath and 15 s nadir values observable at baseline in room air would be a more accurate representation of the hyperoxia response magnitudes for both ventilation and MSNA. This is especially important when (i) short nadir responses are analysed and (ii) the outcome measurements of interest (ventilation and MSNA) are highly fluctuant. Indeed, we have demonstrated a lack of validity and reliability in 15 s MSNA measurement epochs (Notay et al. 2016) and suspect a similar lack of validity and reliability for single breath ventilatory measurements. Therefore, we suspect that the level of ventilatory and sympathetic inhibition is vastly overestimated in the present study. Regarding the statistical tests, the authors calculated CVME for ventilatory and MSNA variables to assess day-to-day reproducibility during baseline, hyperoxia and hypoxia. To supplement, other measures of reliability, such as the intraclass correlation coefficient test to assess the day-to-day absolute agreement and consistency, would be a beneficial adjunct, bringing significant insights related to clinical implications. Moreover, reporting differences between males and females, although not the study focus, could be valuable, notably for future meta-analyses, as it is known that most cardiovascular diseases have important sex-based differences in epidemiology and clinical outcomes. Finally, whether the uncoupled ventilatory and MSNA response magnitudes in treated OSA patients is relevant to other disease states requires future investigation. Prasad and colleagues (2020) have strongly demonstrated a dissociation of ventilatory and sympathetic response magnitudes to hyperoxia and eucapnic hypoxia. These data demonstrate the importance of differential control when interpreting physiological outcomes in human testing. As a result, the authors recommend ventilation and sympathetic responses to gas challenges be specifically evaluated. It is important to acknowledge that direct measures of sympathetic activity (i.e. microneurography) are invasive and can be difficult to obtain in ageing and clinical populations. Therefore, future directions call for the development of accurate, non-invasive assessments to quantify contributions of the peripheral chemoreceptors to resting and reflexive sympathetic activity that can be routinely used in clinical and research environments. None. All authors have read and approved the final version of this manuscript and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All persons designated as authors qualify for authorship, and all those who qualify for authorship are listed. A.V.I. is supported by a Canadian Institute of Health Research (CIHR) Fellowship. J.B. is supported by the Libin Cardiovascular Institute Postdoctoral Scholarship in Women's Cardiovascular Health and Natural Sciences and Engineering Research Council of Canada (NSERC) Brain CREATE. N.O.B. is supported by SIDS Calgary Society. L.T.L. is supported by Alberta Innovates Health Solutions. The authors recognize that not all relevant articles were cited due to reference limitations. The authors appreciate the critical review and guidance of Drs Richard J. A. Wilson (University of Calgary) and Daniel A. Keir (Western University) in preparation of the manuscript.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.