BackgroundSystemic blood flow in patients on extracorporeal assist devices is frequently not or only minimally pulsatile. Loss of pulsatile brain perfusion, however, has been implicated in neurological complications. Furthermore, the adverse effects of absent pulsatility on the cerebral microcirculation are modulated similarly as CO2 vasoreactivity in resistance vessels. During support with an extracorporeal assist device swings in arterial carbon dioxide partial pressures (PaCO2) that determine cerebral oxygen delivery are not uncommon—especially when CO2 is eliminated by the respirator as well as via the gas exchanger of an extracorporeal membrane oxygenation machine. We, therefore, investigated whether non-pulsatile flow affects cerebrovascular CO2 reactivity (CVR) and regional brain oxygenation (rSO2).MethodsIn this prospective, single-centre case-control trial, we studied 32 patients undergoing elective cardiac surgery. Blood flow velocity in the middle cerebral artery (MCAv) as well as rSO2 was determined during step changes of PaCO2 between 30, 40, and 50 mmHg. Measurements were conducted on cardiopulmonary bypass during non-pulsatile and postoperatively under pulsatile blood flow at comparable test conditions. Corresponding changes of CVR and concomitant rSO2 alterations were determined for each flow mode. Each patient served as her own control.ResultsMCAv was generally lower during hypocapnia than during normocapnia and hypercapnia (p < 0.0001). However, the MCAv/PaCO2 slope during non-pulsatile flow was 14.4 cm/s/mmHg [CI 11.8–16.9] and 10.4 cm/s/mmHg [CI 7.9–13.0] after return of pulsatility (p = 0.03). During hypocapnia, non-pulsatile CVR (4.3 ± 1.7%/mmHg) was higher than pulsatile CVR (3.1 ± 1.3%/mmHg, p = 0.01). Independent of the flow mode, we observed a decline in rSO2 during hypocapnia and a corresponding rise during hypercapnia (p < 0.0001). However, the relationship between ΔrSO2 and ΔMCAv was less pronounced during non-pulsatile flow.ConclusionsNon-pulsatile perfusion is associated with enhanced cerebrovascular CVR resulting in greater relative decreases of cerebral blood flow during hypocapnia. Heterogenic microvascular perfusion may account for the attenuated ΔrSO2/ΔMCAv slope. Potential hazards related to this altered regulation of cerebral perfusion still need to be assessed.Trial registrationThe study was retrospectively registered on October 30, 2018, with Clinical Trial.gov (NCT03732651).