AVMs are a relatively small but important cause of stroke from spontaneous intracranial hemorrhage (ICH) that is potentially treatable. Because the primary reason to intervene in the natural course of the disease is prophylaxis from spontaneous ICH, the decision to treat should be based on hemorrhagic risk. However, insufficient information is available on the mechanisms and natural history of AVM hemorrhage. This interdisciplinary project will bring together a group of clinical neuroscientists, applied mathematicians and engineers in an effort to theoretically model the risk of spontaneous ICH. We will test the primary hypothesis that weighting structural factors (intranidal vessel wall thickness-to-radius ratio in what we will call """"""""structural- weighted"""""""" risk estimate) is more important than weighting hemodynamic factors (intranidal transmural pressure in what we will call """"""""hemodynamic-weighted"""""""" risk estimate) in determining the relative risk of spontaneous ICH. Theoretical predictions of risk (Riskmodel) will be compared to experimentally-derived estimates (Riskexp). (Spec.
Aim 1) We will collect data from 90 consecutive AVM patients (30/yr). A patient-specific computational model of the cerebral circulation will be constructed for all cases based on (a) detailed angiographic and MR data, (b) cerebral arterial pressures measured during superselective angiography and (c) histopathologic examinations of surgically-excised AVM specimens. The two model risk estimates (Riskmodel), """"""""structural- weighted"""""""" and """"""""hemodynamic-weighted,"""""""" will be compared to the risk of spontaneous ICH risk determined by experimental observations (Riskexp). Riskexp is calculated using a previously-described statistical method for determining the relative risk of having initially presented with AVM hemorrhage. (Spec.
Aim II) To partially validate each patient-specific computational model, we will measure changes in cerebral arterial pressure during nitroprusside-induced systemic hypotension and compare this to the patient-specific model's predictions. Use of modeling will allow delineation of pathomechanisms and risk factors for spontaneous ICH, to be used as a means of risk stratification in the design of future clinical trials. Such an approach is important because of the low incidence of AVMs. Overall, interdisciplinary modeling studies of the cerebral circulation can be an important adjunct to experimental studies to increase knowledge of cerebral pathophysiology and to devise treatment strategies by either the screening of proposed theories or the testing of existing ones.
