The prospective natural history of cerebral venous malformations

A 10-year prospective clinical and magnetic resonance imaging study was undertaken to determine the natural history of venous malformations. We assessed the hemorrhage rate and morbidity associated with venous malformations of the brain.

From 1986 to 1996, 80 patients with venous malformations were referred to the University of Pittsburgh multidisciplinary vascular malformation study group for evaluation. Observation was recommended for all patients. Follow-up clinical information was obtained from patients or their referring physicians through questionnaire or phone conversation.

Twenty-two patients presented with neurological signs or symptoms that were thought to be related to the malformations (nine with headaches, four with seizures, three with sensory symptoms, three with motor deficits, two with trigeminal neuralgia, and one with an extrapyramidal disorder). Twenty-three patients presented with headaches that were not considered to be related to the malformations. The retrospective hemorrhage rate (from birth to study entry) was 0.61% (18 bleeds in 2,949 patient-years). Sixteen patients had sustained previous brain hemorrhage in the region of the venous malformations, two of whom had suffered subsequent hemorrhage. During the prospective follow-up period totaling 298 patient-years of clinical observation, two patients suffered hemorrhage (0.68% per year), but only one had a symptomatic bleed (0.34% per year). This patient had not hemorrhaged previously. One of these patients remained asymptomatic, whereas the second developed temporary worsening of facial paresthesias. No patient died as a result of the venous malformations.

The hemorrhage rate of a patient with venous malformations is similar to the rates presented in previous reports for patients with cavernous malformations without previous symptomatic hemorrhage. We think that hemorrhage in a patient with venous malformations may be related to an underlying but not yet documented cavernous malformation. Because of the low risk for new neurological events, we advocate conservative management. The risks associated with surgical intervention greatly exceed the low risk of morbidity related to venous malformation hemorrhage.