As early as 1948, Ketty & Schmidt44 reported a 13 % decrease in Cerebral Blood Flow (CBF) in normobaric hyperoxia (1 ata, FiO2 = 1) and an 18 % decrease in HBO (2 ata, FiO2 = 1). This decrease has also been observed in anaesthetized dogs under controlled ventilation (Jacobson45, Bergofsky & Bertun46) and in conscious rats in spontaneous ventilation (Hordnes & Tyssebotn47). In all these studies, the drop in CBF was equivalent or slightly greater than the drop in cardiac output. In conscious ewes under controlled ventilation, Matalon15, observed an increase instead of a decrease in CBF in HBO. However, this increase was only observed when signs of oxygen toxicity occurred and at a time when the global cardiac output itself was increasing.
Alterations in blood flow distribution have been studied within the brain itself. Bergo & Tyssebotn48 studied CBF distribution to the various areas of the brain in conscious rats in spontaneous ventilation at 1, 3 and 5 ata FiO2 = 1. At 1 and 3 ata, blood flow only decreased in the areas of the pons, mesencephalon, thalamus and hypothalamus, whereas it decreased in all the areas at 5 ata.
This variation in effects of high pressures of oxygen relative to the application duration of and the pressure itself has also been observed by other authors (Bean49, Lambertsen50, Torbati51). In conscious rats, Torbati51 showed how at pressures of oxygen between 1 and 3.5 ata, the drop in CBF was maintained throughout exposure. At 5 ata, FiO2 = 1, the initial drop in CBF was followed by a secondary increase after 30 minutes. At 7 ata FiO2 = 1, there was no decrease in CBF, on the contrary, an increase in CBF was shortly followed by a convulsive hyperoxic attack.
In human beings, Torbati51 confirmed the data of Ledingham52 by suggesting that the secondary increase in CBF was dependent on an increase in cerebral PCO2. The vasodilating effect of the latter countered the vasoconstricting effect of the hyperoxia. During moderate hyperoxia (1-3.5 ata, FiO2 = 1), PO2 related vasoconstriction limited the increase in tissue PO2. However, the hyperoxia induced an increase in cerebral PCO2 by means of (1) alveolar hypoventilation (due to a decrease in respiratory drive) (2) decrease in carbamino-haemoglobin (Haldane effect) (3) decrease in cerebral blood flow due to hyperoxic vasoconstriction, so that vasodilation appeared. This led to an increase in cerebral PO2 and ultimately toxic manifestations of hyperoxia. The neurological tolerance to hyperoxia is, therefore, related to cerebral vasoregulation which controls PO2 within a range where anti-oxidizing defences of the cells can compensate for the hyperoxia.
To summarize, the variation in CBF in hyperoxia depends on an interrelationship between various regulating systems in which local PO2 and PCO2 seem to play a major role. Thus the state of consciousness, anaesthetic drugs provided, control of ventilation and level of PCO2 are all important factors to be taken into account when analysing alterations in CBF. Also, Torbati & Carey53 have emphasised that brain trauma has distinct effects on the vasoconstrictive response to the hyperoxia. This means that whenever HBO is used on cases involving cerebral pathologies, careful monitoring is recommended of: cerebral functions (continuous EEG), jugular venous pressure and oxygen saturation, jugular arterio-venous gradients for glucose and lactate or of CBF itself.
Was this article helpful?