The CO2 stimulus duration and steady-state time point used for data extraction alters the cerebrovascular reactivity outcome measure.

Abstract

What is the central question of this study? Cerebrovascular reactivity (CVR) is a common functional test to assess brain health, and impaired CVR has been associated with all-cause cardiovascular mortality: does the duration of the CO2 stimulus and the time point used for data extraction alter the CVR outcome measure? What is the main finding and its importance? This study demonstrated CVR measures calculated from 1 and 2min CO2 stimulus durations were significantly higher than CVR calculated from a 4min CO2 stimulus. CVRs calculated from the first 2min of the CO2 stimulus were significantly higher than CVR values calculated from the final minute if the duration was ≥4min. This study highlights the need for consistent methodological approaches. Cerebrovascular reactivity to carbon dioxide (CVR) is a common functional test to assess brain vascular health, though conflicting age and fitness effects have been reported. Studies have used different CO2 stimulus durations to induce CVR and extracted data from different time points for analysis. Therefore, this study examined whether these differences alter CVR and explain conflicting findings. Eighteen healthy volunteers (24±5years) inhaled CO2 for four stimulus durations (1, 2, 4 and 5min) of 5% CO2 (in air) via the open-circuit Douglas bag method, in a randomized order. CVR data were derived from transcranial Doppler (TCD) measures of middle cerebral artery blood velocity (MCAv), with concurrent ventilatory sensitivity to the CO2 stimulus ( ). Repeated measures ANOVAs compared CVR and measures between stimulus durations and steady-state time points. An effect of stimulus duration was observed (P=0.002, η²=0.140), with 1min (P=0.010) and 2min (P<0.001) differing from 4min, and 2min differing from 5min (P=0.019) durations. sensitivity increased ∼3-fold from 1min to 4 and 5min durations (P<0.001, η²=0.485). CVRs calculated from different steady-state time points within each stimulus duration were different (P<0.001, η²=0.454), specifically for 4min (P=0.001) and 5min (P<0.001), but not 2min stimulus durations (P=0.273). These findings demonstrate that methodological differences alter the CVR measure.