The cerebrovascular response to carbon dioxide in humans

Abstract

Carbon dioxide (CO2) increases cerebral blood flow and arterial blood pressure. Cerebral blood flow increases not only due to the vasodilating effect of CO2 but also because of the increased perfusion pressure after autoregulation is exhausted. Our objective was to measure the responses of both middle cerebral artery velocity (MCAv) and mean arterial blood pressure (MAP) to CO2 in human subjects using Duffin-type isoxic rebreathing tests. Comparisons of isoxic hyperoxic with isoxic hypoxic tests enabled the effect of oxygen tension to be determined. During rebreathing the MCAv response to CO2 was sigmoidal below a discernible threshold CO2 tension, increasing from a hypocapnic minimum to a hypercapnic maximum. In most subjects this threshold corresponded with the CO2 tension at which MAP began to increase. Above this threshold both MCAv and MAP increased linearly with CO2 tension. The sigmoidal MCAv response was centred at a CO2 tension close to normal resting values (overall mean 36 mmHg). While hypoxia increased the hypercapnic maximum percentage increase in MCAv with CO2 (overall means from76.5 to 108%) it did not affect other sigmoid parameters. Hypoxia also did not alter the supra-threshold MCAv and MAP responses to CO2 (overall mean slopes 5.5% mmHg⁻¹ and 2.1 mmHg mmHg⁻¹, respectively), but did reduce the threshold (overall means from 51.5 to 46.8 mmHg). We concluded that in the MCAv response range below the threshold for the increase of MAP with CO2, the MCAv measurement reflects vascular reactivity to CO2 alone at a constant MAP.

Figure 1. Hyperoxic and hypoxic rebreathing tests for subject 13

In the hyperoxic test thresholds T_MAP and T_MCAv are easily discerned. In the hypoxic test, although continues to decrease during hyperventilation, MCAv reaches a minimum.

Figure 2. Hyperoxic test for subject 7

The conversion of MCAv from cm s⁻¹ (open triangles, right axis) to % change (filled triangles, left axis) was based on the mean minimum MCAv during hyperventilation. The sigmoid parameters for the fit to the % MCAv response are indicated, as is the threshold T_MCAv at which MCAv begins to increase linearly with . The upper inset shows the MAP vs. response and the linear fit above the T_MAP threshold. The lower inset shows the MCAv vs. MAP response and the linear fit derived from the supra-thresholds linear MAP and MCAv responses to .

Figure 3

Hyperoxic (filled) and hypoxic (open) rebreathing test MCAv (triangles) and MAP (diamonds) responses for subject 7 showing the sigmoidal and linear fits to the hyperoxic and hypoxic responses (continuous and dotted lines, respectively)

Figure 4

Examples of the MCAv and MAP responses during rebreathing tests illustrating the variety of responses observed