Important role of the erythrocyte as a regulator of neurovascular coupling in humans

The erythrocyte has emerged as an integral oxygen sensor and response effector for the regulation of oxygen delivery throughout the vasculature. In vivo rodent models have demonstrated that erythrocyte signaling underscores the coupling of cerebral blood flow to neural activity [i.e. neurovascular coupling (NVC)]; however, this has yet to be examined in humans. In eight healthy males (27±6 years, 23±2 kg/m2) (mean ± standard deviation), NVC was assessed in response to an eyes open visual stimulus using transcranial Doppler ultrasound prior to and following isovolumic haemodilution whereby 20% of blood volume was removed and replaced with an equal volume of 5% human serum albumin. This NVC test assesses the coupling of posterior cerebral neural activity, which is manipulated with the aforementioned visual stimulus, and the corresponding change in posterior cerebral blood flow. Paired t-tests were used to test for statistical significance (α=0.05). Haemodilution resulted in a decreased haemoglobin concentration (14.0 ± 0.9 vs 11.4 ± 0.5 g/dL; P<0.01), hematocrit (42.8 ± 2.5 vs. 35.0 ± 1.5 %; P<0.01) and arterial oxygen content (CaO2; 19.1 ± 1.6 vs. 15.6 ± 0.7 mL/dL; P<0.01). Posterior cerebral artery blood velocity (PCAv) was correspondingly elevated (39 ± 8 to 43 ± 11cm/s; P<0.01) in the face of unaltered mean intra-arterial pressure (99 ± 9 vs. 97 ± 4mmHg; P = 0.67) and partial pressure of end-tidal CO2 (43 ± 2 vs. 42 ± 2 mmHg; P = 0.20). Collectively, these changes in CaO2 and PCAv led to a reduction in estimated cerebral oxygen delivery (PCAv × CaO2, arbitrary units) at rest (751±185 vs. 693±178; P=0.03). The absolute peak change in PCAv (8.5±4.7 vs. 8.5±3.3cm/s; P=0.29) and the absolute average change in PCAv (3.4±2.4 vs. 2.9±2.0cm/s; P=0.20) across the eyes open period were unaltered; however, given the reduced CaO2 following haemodilution, both the peak (177±83 vs. 133±63; P=0.01) and average (65±45 vs. 45±31; P=0.04) increases in estimated cerebral oxygen delivery were reduced during NVC. The attenuated oxygen delivery response is attributable to a reduced peak (24 ± 9 vs. 19 ±9%; P = 0.02) and average (9 ± 5 vs. 6 ± 4%; P = 0.04) percent increases in PCAv as well as smaller peak (-0.48±0.25 vs. -0.36±0.17mmHg/cm/s; P<0.01) and average (-0.20±0.11 vs. -0.13±0.09mmHg/cm/s; P<0.01) reductions in cerebrovascular resistance. These data indicate that reductions in vasomotor tone during NVC in humans are dependent upon erythrocyte-mediated signalling, and that acute experimental reductions in haemoglobin concentration constrain increases in oxygen delivery during neural activation.