Development of hypertension in chronic intermittent hypoxia: is it driven by cardiac output rather than by peripheral resistance?

In this issue of Experimental Physiology, Lucking et al. (2014) present a range of different investigations of the hypertension that develops in rats exposed to chronic intermittent hypoxia (CIH). Chronic intermittent hypoxia is the most commonly used experimental model of obstructive sleep apnoea (OSA), which probably occurs in ∼25% of the adult population, with greater prevalence in men. Obstructive sleep apnoea is associated with increased cardiovascular risk, particularly hypertension, but also heart failure. Obstructive sleep apnoea is also estimated to be present in up to 80% of hypertensives. The mechanisms that link CIH, OSA and hypertension are therefore attracting much interest. It is well known that CIH and OSA are accompanied by increased sympathetic nerve activity (SNA) to the major vascular beds. Indeed, it is widely accepted that the hypertension is caused by increased vascular SNA and peripheral vasoconstriction. Lucking et al. (2014) now draw the provocative conclusion that the hypertension of CIH is caused by a sympathetically driven increase in cardiac output (CO), not by peripheral vasoconstriction. They used male rats exposed to CIH for only 2 weeks, whereas many studies have focused on 4–7 weeks of CIH. Thus, their findings are particularly relevant to the development of hypertension. At 2 weeks, CIH had already induced right ventricular (RV) and left ventricular (LV) hypertrophy as judged by the tissue weight/body weight ratio, while echocardiography on the pulmonary artery indicated an increase in CO. This contrasts with published evidence that CIH lasting 4–5 weeks led to RV and LV hypertrophy but a fall in CO (Chen et al. 2005). Their new results indicate that the heart initially shows an adaptive response to CIH before myocardial dysfunction develops. But, whereas echocardiography showed no detectable change in LV wall thickness or cavity diameter, there was a substantial increase in thickness of the interventricular septum (IVS). This is particularly interesting because in newly diagnosed OSA patients and patients with dilated cardiomyopathy and OSA, the outstanding feature was IVS thickening (Usui et al. 2006). These new findings raise the possibility that IVS thickening is an early effect of CIH, whether induced experimentally or by OSA. The mechanisms underlying this phenomenon require investigation. Baseline heart rate (HR) was increased by 2 weeks of CIH and presumably contributed to the increased CO. The tachycardia was attenuated by the autonomic ganglion blocker hexamethonium and ascribed to increased cardiac sympathetic activity, consistent with the known facilitatory effects of CIH on peripheral chemoreceptor sensitivity and the chemoreceptor reflex. Raised sympathetic drive would exert β-adrenoceptor-mediated hypertrophic effects on the myocardium, while hypoxia-induced pulmonary vasoconstriction could explain RV hypertrophy; however, neither would explain a preferential effect on IVS thickness. In OSA, it has been argued that the exaggerated negative intrathoracic pressure swings that occur during apnoeas increase RV filling, while hypoxia-induced pulmonary vasoconstriction increases RV afterload. Thus, the IVS is pushed towards the LV, increasing the tension of the IVS and acting as a remodelling stimulus (Usui et al. 2006). It is an attractive possibility that something similar occurs in CIH, particularly if the breathing instability caused by CIH promotes spontaneous apnoeas during sleep (see Mateika & Narwani, 2009). The conclusion that hypertension in CIH is not caused by sympathetically mediated peripheral vasoconstriction is more contentious. Lucking et al. (2014) calculated total peripheral resistance (TPR) from CO measured by echocardiography and arterial pressure (ABP) recorded subsequently, after carotid artery cannulation. A change in CO or ABP over time could have affected the calculated TPR and evaluation of the effect of CIH. However, if we accept that TPR is not chronically raised by CIH, then TPR is eliminated as a potential stimulus for LV hypertrophy. The question then arises of whether ventricular hypertrophy was dominated by β-adrenoceptor stimulation or whether the measured increase in LV mass mainly reflected thickening of the IVS. Lucking et al. (2014) support their conclusion with the finding that baseline hindlimb vascular conductance was not altered by CIH. Furthermore, hexamethonium caused similar vasodilatation in the hindlimbs of control rats and those exposed to 2 weeks of CIH, suggesting no difference in sympathetic vasoconstrictor tone. To strengthen this result, direct evidence would be required that muscle SNA was genuinely abolished by their dose of hexamethonium. Nevertheless, they also showed that CIH did not significantly alter the vasoconstrictor responses evoked in hindlimb muscle by lumbar sympathetic stimulation over a range of frequencies from 0.25 to 10 Hz. This is particularly important because Silva & Schreihofer (2011) found that the reflex increases in splanchnic SNA evoked by various afferent inputs (peripheral chemoreceptor, somatosensory and nasopharyngeal) were greatly augmented by 2 weeks of CIH and yet the evoked increases in ABP were no bigger. Thus, augmented increases in SNA did not translate into greater vasoconstriction. Additional findings made by Lucking et al. (2014) may explain this paradox. Chronic intermittent hypoxia enhanced the elastin content and increased the compliance of the aorta, which should help to buffer rises in ABP caused by peripheral vasoconstriction. Furthermore, vasoconstrictor responses evoked in hindlimb muscle by α-adrenoceptor stimulation were blunted. This could be attributed to desensitization of α-receptors caused by chronic exposure to nerve-released noradrenaline, for they made the novel finding that CIH increased the density of sympathetic nerves on arterial vessels supplying skeletal muscle. However, the question then becomes: does raised vascular SNA contribute to the maintenance of hypertension before 2 weeks? Sharpe et al. (2013) have already shown that by 1 week of CIH there was hypertension in conscious rats but no baseline tachycardia, and hexamethonium normalized ABP. Under general anaesthesia, lumbar SNA responses to a hypertonic stimulus were augmented, whereas the accompanying pressor responses were not. Thus, at 1 week of CIH, sympathetically mediated vasoconstriction apparently does make the major contribution to the hypertension, and vascular SNA is likely to be raised before this. It therefore seems probable that functional and structural changes promoted by raised SNA and occurring very early in CIH lead to the situation described by Lucking et al. (2014), such that by 2 weeks the resting sympathetic vasoconstrictor tone is no longer responsible for the maintenance of hypertension, and sympathetically driven CO plays the greater role. Their novel findings should certainly prompt further studies on these interrelationships. Readers are invited to give their opinion on this article. To submit a comment, go to: http://ep.physoc.org/letters/submit/expphysiol;99/10/1286 None declared. None declared.