Loss of microvascular blood flow is a critical feature of hypoperfusion across neurovascular diseases, particularly hypertension-associated vascular dementia. However, the early mechanisms underlying capillary flow dysregulation (pre-rarefaction), stalling (functional rarefaction), and eventual microvessel loss (anatomical rarefaction), remain poorly defined and difficult to model. Here, we examine these microvascular pathologies in an Angiotensin II (AngII) infusion-recovery model in which mice receive a pressor dose of AngII (800 ng/kg/min) or vehicle saline for 28 days followed by a 14-day recovery period.

Blood pressure and heart rate were monitored by tail cuff plethysmography and microvascular flow was quantified at two-week intervals through an acute cranial window using two-photon line scan imaging and Radon transform analysis. Flow velocity, temporal variability and stall frequency were assessed in the first four capillary branch orders adjoining precapillary arterioles. Brains were subsequently collected for immunohistochemical analyses of endothelial cells, pericytes and tissue hypoxia.

AngII induced severe hypertension by week two (131±4 mmHg AngII vs. 100±3 mmHg saline; N=8 vs. N=8; p=0.0002) which persisted through week four (136±3 mmHg AngII vs. 111±3 mmHg saline; N=8 vs. N=8; p=0.0002) and normalized by week six (105±4 mmHg AngII vs. 101±3 mmHg saline; N=8 vs. N=8; p=0.8706). Capillary flow velocity slowed considerably by week four (0.79±0.07 mm/s AngII vs. 1.13±0.09 mm/s saline; N=4 vs. N=5; p=0.0125) and remained reduced at week six (0.87±0.06 mm/s AngII vs. 1.18±0.10 mm/s saline; N=6 vs. N=6; p=0.0389) despite normalized blood pressure. Flow variability, measured as temporal coefficient of variation (CV) excluding stalls, was unchanged during infusion but increased at recovery week six (AngII CV=0.28±0.02, saline CV=0.18±0.01; N=6 vs. N=6; p=0.0021). Stalling rose progressively, peaking at recovery week six (AngII 24.2±2.6% vs. saline 9.9±1.9%; N=6 vs. N=6; p<0.0001).

These functional disturbances were accompanied by anatomical rarefaction. Microvessel loss was pronounced in the hippocampus of six-week AngII recovery mice (5.9±0.4% AngII vs. 8.2±0.4% saline CD31+ vessel density; N=6 vs. N=6; p=0.0005) with concurrent pericyte loss (2.1±0.3% AngII vs. 3.8±0.2% AngII PDGFRb+ pericyte density; N=6 vs. N=6; p=0.0002). This coincided with substantial tissue hypoxia (17.0±0.8% AngII vs. 0.9±0.1% saline HypoxyProbe+ tissue area; N=6 vs. N=6; p<0.0001). By contrast, the cortex displayed a trend towards microvessel loss (9.2±0.3% AngII vs. 10.0±0.4% saline CD31+ vessel density; N=6 vs. N=6; p=0.2528), significant pericyte loss (2.6±0.3% AngII vs. 4.5±0.3% saline PDGFRb+ pericyte density; N=6 vs. N=6; p<0.0001) and less pronounced hypoxia than the hippocampus (4.1±0.8% AngII vs. 0.7±0.1% saline HypoxyProbe+ tissue area; N=6 vs. N=6; p=0.0009).

Together, these findings show that four-week AngII infusion initiates early microvascular rarefaction, producing persistent microvessel loss even after a two-week return to normotension. The hippocampus exhibits the greatest anatomical vulnerability, whilst the cortex shows marked flow dysregulation and stalling consistent with early-stage rarefaction. These early disturbances likely precede broader neurological consequences observed in extended AngII models. This preliminary data outlines the early steps linking hypertension to hypoperfusion, regional hypoxia, and cognitive decline, laying the groundwork for future efforts to define the mechanisms driving microvascular rarefaction in vascular dementia.