Relationship of pulsatility and resistance indices to cerebral arteriovenous malformation angioarchitectural features and hemorrhage

- Pulsatility (PI) and resistance (RI) indices can be measured using quantitative magnetic resonance angiography.
- PI and RI are lower in cerebral arteriovenous malformation (AVM) arterial feeders compared to normal vessels.
- We found no significant correlation between PI, RI, and AVM angioarchitectural characteristics thought to be associated with increased hemorrhage risk.

The role that hemodynamics plays in the pathophysiology of cerebral arteriovenous malformation (AVM) hemorrhage remains unclear. Here, we examine the relationship of pulsatility and resistance indices to AVM angioarchitectural features and hemorrhage. Records of patients with cerebral AVMs evaluated at our institution between 2007-2014 and with flows obtained before treatment using quantitative magnetic resonance angiography (QMRA) were retrospectively reviewed. Flow volume rate and flow velocity were measured in primary arterial feeders and compared to their contralateral counterparts. Pulsatility index (PI) = [(systolic flow velocity − diastolic flow velocity)/mean flow velocity] and resistance index (RI) = [(systolic flow velocity − diastolic flow velocity)/systolic flow velocity] were calculated for each feeder and compared to the normal contralateral vessel. Relationships between PI, RI and AVM clinical and angioarchitectural features were assessed using linear regression. Seventy-two patients with a total of 101 feeder arteries were included. PI and RI were significantly lower in AVM arterial feeders compared to normal vessels, thereby resulting in significantly higher flow volume rates and flow velocities in feeder vessels. There was no significant association of PI and RI with hemorrhagic presentation, exclusive deep venous drainage, venous stenosis, single draining vein, or deep location. In conclusion, PI and RI can be measured using QMRA and are lower in AVM arterial feeders compared to normal vessels. Although we found no significant correlation between PI, RI, and AVM angioarchitectural characteristics thought to be associated with increased hemorrhage risk, future studies with larger sample sizes may better elucidate this relationship.