Clinical and biomedical researchers interested in cerebral hemodynamics and the application of novel MRI techniques in the clinic.Sickle Cell Disease (SCD) is caused by a mutation of the hemoglobin gene, the main oxygen-carrying protein in the blood. At low oxygen tension, sickle hemoglobin polymerizes resulting in vascular injury as a consequence of compromised flow and altered arterial wall shear stress (WSS). 4D Flow MRI is a non-invasive technique allowing blood flow velocity measurements and estimation of WSS. We investigated dynamic changes in velocity, WSS and vessel diameter in the anterior circulation of the Circle of Willis (CoW) in response to a vasodilator (acetazolamide [ACZ]).¹ We hypothesized that velocity, WSS and vessel diameter will increase in the CoW after administration of ACZ. 15 SCD patients (age=34±12 years) with HbSS or HbSß⁰-thalassaemia genotypes, and 4 age- and ethnicity-matched healthy controls (aged=34±16 years) were scanned with retrospectively gated 4D Flow MRI before and after an intravenous administration of 16mg/kg ACZ on a 3.0 Tesla system (Ingenia, Philips Healthcare, Best, the Netherlands). Scan parameters: spatial resolution: 0.5x0.5x0.5 mm³; TR/TE/FA=8.6/4.1/20°; VENC=100cm/s in all directions. To keep acquisition time short (7 minutes), the temporal resolution was kept low (4 heart phases). Magnitude images with the highest signal intensity were used for segmentation of the full CoW and the left and right internal carotid arteries, middle cerebral arteries and anterior cerebral arteries (L/RICA, L/RMCA and L/RACA) in commercial software (Mimics, Materialise, Leuven). The segmentation process was repeated for 18 scans (randomized for pre and post ACZ administration and controls and patients) by two observers. Venous blood was obtained prior to MRI for hematocrit values to estimate viscosity2. WSS was subsequently calculated in the entire CoW with subject-specific viscosity using an algorithm as previously described³. The vessel diameters in the CoW were calculated by tracking the inward normal on the points where WSS was calculated upon exiting the opposite side of the vessel. The ICA, MCA and ACA masks were used to measure locally averaged velocity, WSS and diameter values. Wilcoxon rank-sum tests were performed to compare time-averaged velocity, WSS and diameter in the MCA, ACA and ICA segments between controls and patients, and before and after ACZ. P<0.05 was considered significant. Inter-observer variability was expressed as the intra-class correlation coefficients (ICC) and coefficient of variation (CV) for velocity, WSS and diameters in the ICA, MCA and ACA.Hematocrit values were 27±3% in patients and 44±1% in controls. Figure 1 shows a representative example of 3D velocity, WSS and diameter maps before and after administration of ACZ. It can be seen that velocity increased in the vessels of interest, that WSS increased in the left ACA and that the diameter increased in the right ACA. In table 1, the values for velocity, WSS and diameter are given for the MCAs, ICAs and ACAs for controls and patients. Except for the left ACA, velocity increased significantly in controls in all vessel segments after ACZ. In patients, velocity increased significantly in all vessel segments. WSS increased significantly in the right MCA and both ICAs in controls, whereas for the patients, WSS increased significantly in all vessels except for the left ACA. The vessel diameter did not change after administration of ACZ. Velocity was significantly higher in patients compared to controls in the left MCA and right ICA. WSS was significantly lower for the patients compared to the controls in the ICA after ACZ. In table 2, the results of the inter-observer analysis is given for velocity, WSS and diameter. Excellent agreement was found for the velocity and WSS values but was poor for the ACAs. As hypothesized, both the control and patient cohort responded to ACZ administration with a significant increase in velocity and WSS values in most vessel segments. At the same time, vessel diameters did not increase. Thus, ACZ may dilate the microvasculature rather than the larger vessels of the CoW. Interestingly, the change in velocity after administration of ACZ was larger in controls than in patients, who had a higher baseline velocity. This could indicate that cerebral blood flow (CBF) is elevated in patients compared to controls. The finding that vessel diameters did not increase may not hold for the ACAs, where a low inter-observer ICC was found. More advanced segmentation methods are needed to accurately delineate smaller CoW vessels. In conclusion, we found that velocity and WSS increased in the CoW after administration of the vasodilator ACZ, as measured with 4D flow MRI,. This indicates that cerebral blood flow increases in controls and patients with sickle cell disease after ACZ.