The primary reason to treat cerebral arteriovenous malformations (AVMs) is the prophylaxis against intracranial hemorrhage (ICH). However, treatment of AVMs by surgery, embolization and/or radiosurgery carries a significant risk. Newer evidence suggests that distinct natural history sub-groups are present under the general heading of AVM . This project represents a three-pronged plan of research. The accomplishment of these aims could bring to treating physicians and surgeons a rational basis for decreasing morbidity and more cost-effective decision-making in patients with AVMs.
Specific Aim I is a single-center-based study of hemorrhagic risk to (Ia) define factors for spontaneous ICH and to test the hypothesis that high feeding artery pressure and exclusively-deep venous drainage are the strongest predictors for hemorrhage; and to (Ib) extend the ongoing prospective study of new or recurrent hemorrhage after initial presentation. We will test the hypothesis that subsequent AVM hemorrhage is related to initial hemorrhagic presentation of the disease.
This aim will further define hemorrhagic presentation as a risk, factor for subsequent bleeding, and also establishes the critical assumption for (Ia) that recurrent ICH is related to the pathophysiology of ICH at initial presentation.
Specific Aim II is a population-based approach to study spontaneous AVM hemorrhage to broaden the applicability of our findings under Specific Aim Ib. In this New York Islands AVM study, participating institutions include those from Manhattan, Staten Island, Queens and Brooklyn; and Nassau and Suffolk Counties on Long Island (population of aprox. 8,800,000 by 1996 estimates).
For Specific Aim III we propose to develop and test the feasibility of a risk-based allocation algorithm (also called assured allocation) for use in clinical trials using information gained from Specific Aims I and II. This novel alternative to a randomized clinical trial, successfully tested in retrospective studies, has not been previously used in clinical interventional trials. We expect to demonstrate, by development of an statistical model, that a risk-based allocation method is a viable strategy as the basis for constructing a multicenter (possibly multinational) clinical trial.
Showing the most recent 10 out of 96 publications
