Performing under pressure: hypertension and the regulation of cerebral oxygen delivery.

Abstract

Reductions in arterial oxygen content () elicit increases in cerebral blood flow (CBF) that are adequate to maintain convective delivery of oxygen to the brain in healthy humans (i.e. cerebral oxygen delivery, ) (Hoiland et al. 2016). Deviations from this prototypical response in clinical populations are, however, less well understood. In this issue of The Journal of Physiology is a well-designed study by Fernandes and colleagues that investigated the impact of untreated stage-1 hypertension on the cerebrovascular, femoral vascular and sympathetic nervous activity (SNA) responses to isocapnic hypoxaemia (Fernandes et al. 2018). Several particularly interesting findings, specific to the cerebral circulation, included the following: (1) hypertensive individuals had a markedly lower posterior cerebrovascular reactivity to isocapnic hypoxaemia than normotensive individuals as determined by duplex ultrasound measurements of vertebral artery (VA) flow; (2) this reduction in posterior cerebrovascular reactivity to hypoxaemia was of a considerable magnitude and precipitated a hypoxaemia-induced reduction (−16%) in posterior ; and (3) the changes in posterior CBF and were unrelated to increases in muscle SNA. Internal carotid artery (ICA) flow reactivity was not statistically impaired in hypertensive patients despite an ostensible 36% difference between hypertensive and normotensive individuals. This, coupled with a statistically ascertained impairment in the cerebrovascular conductance response of the ICA during hypoxaemia in hypertensive individuals signifies that the lower reactivity (or vasodilatory capacity) of the VA in hypertensive patients (−80% compared to normotensive) may not be selective per se. A more judicious interpretation is that there is a diffuse impairment of cerebrovascular reactivity to isocapnic hypoxaemia throughout the brain in hypertension, although impairments are clearly exaggerated in the posterior circulation with consequential implications for . This finding is in itself novel, although confirmation in a larger cohort is required. Compromised posterior during hypoxaemia in hypertensive patients is important given that posterior cerebrovascular reactivity to hypoxaemia (in relative terms, % reactivity) is typically, albeit not always, reported to be greater than that of the anterior circulation in healthy individuals – a response characteristic attributed to a preferential/teleological protection of brainstem regions associated with basal homeostatic functions (reviewed in Hoiland et al. 2016). The authors speculated that the impaired posterior cerebrovascular reactivity was due to endothelial dysfunction. This is an interesting point given peripheral arterial endothelial function, indexed via brachial artery flow-mediated dilatation, is impaired in hypertensive patients and coincides with impaired cerebrovascular reactivity to CO2 (Lavi, 2006), thus providing inferential evidence for cerebral manifestation of endothelial dysfunction in hypertensive patients. This latter point is further strengthened by recent findings that more directly reinforce the notion that cerebrovascular CO2 reactivity is indeed indicative of cerebral conduit artery endothelial function in humans (Hoiland et al. 2017). Notably, although hypoxaemia elicited increases in both blood velocity and arterial diameter of the VA in normotensive individuals, the VA did not dilate during hypoxaemia in the hypertensive patients consistent with the manifestation of cerebral endothelial dysfunction (Table 2 in Fernandes et al. 2018). The increase in blood velocity through the VA was also lower – although not abolished – in the hypertensive patients, which coupled with the minimal increase in VA conductance likewise suggests additional functional impairments distal to large cerebral arteries. Whether ATP-sensitive potassium (KATP) channel dysfunction is also related to this decrement in posterior cerebrovascular reactivity, as speculated by the authors, is less clear. Despite evidence in animals that KATP channels possess an important role in hypoxic vasodilatation, to my knowledge no studies have investigated this in healthy humans (reviewed in Hoiland et al. 2016). The reduction in posterior due to exaggerated impairment of posterior cerebrovascular reactivity (Fernandes et al. 2018) holds great relevance for the re-emergent selfish brain hypothesis. This hypothesis encompasses the theory that elevated cerebrovascular resistance and consequent cerebral tissue hypoxia leads to elevations in SNA output that increases blood pressure in order to restore optimal CBF at the expense of ensuing hypertension. Indeed, a recent landmark study in humans provided evidence for an involvement of elevated cerebrovascular resistance and reduced CBF/ in the pathogenesis of hypertension (Warnert et al. 2016). The reduction in VA reactivity observed in the study by Fernandes and colleagues indicates that functional changes in CBF control, in addition to the previously ascribed anatomical variations (Warnert et al. 2016), may be a contributory factor to the sub-optimal perfusion that is thought to underlie reflex SNA activity in hypertension (Warnert et al. 2016). Although blood would be fully oxygenated as it flows through larger cerebral vessels (i.e. VA) in both normotensive and hypertensive individuals, potential reductions in hypoxic vasodilatory signal transduction at the level of microvessels where oxygen levels are lower (due to extraction) may be reflected in reduced flow measured in larger vessels. Indeed, this latter point is consistent with the above speculation that vasodilatory function was impaired in large and small cerebral arteries of hypertensive individuals (Fernandes et al. 2018). It remains to be determined if the increased SNA attributed to central hypoxia in the selfish brain hypothesis is due to independent or intertwined influences of both anatomical and functional cerebral abnormalities/impairments. With respect to this latter point, there was no relationship between cerebrovascular function (ΔVA flow and ) and the increase in SNA elicited with isocapnic hypoxaemia. However, the relationship between resting SNA and VA reactivity as well as posterior during hypoxia may provide more insight into the potential role of functional impairments in the proposed selfish brain hypothesis. While not presented in this manner by the authors, or reflective of causation, an x–y plot of VA reactivity (x-axis) and resting SNA (y-axis) may provide insight into a potential functional contribution to the selfish brain hypothesis. The author has no conflict of interest, financial or otherwise. The author is supported by a Natural Sciences and Engineering Research Council of Canada Post-Graduate Scholarship.