A longitudinal study of patients with venous malformations: documentation of a negligible hemorrhage risk and benign natural history.

Abstract

The natural history of cerebral venous malformations has not been well documented, and the clinical significance of these common lesions remains controversial. The objective of this study was to follow longitudinally the clinical course of patients with cerebral venous malformations to document the natural history of the lesion. Ninety-two patients with radiographically confirmed venous malformations were entered into the study between 1987 and 1996. Annual follow-up was maintained by clinic visits and/or phone interviews. Sixty-three patients (25 men and 38 women) with more than 1 year of follow-up were analyzed. McNemar's test and logistic regression analysis was applied to prevalence of presenting symptoms over time. An average per patient follow-up of 4.2 years yielded 2,721 retrospective and 301 prospective lesion-years for analysis. Average age at diagnosis was 39.1 years (SD, 18.7 years; range, 2 to 73 years). The most frequent lesion locations included the frontal lobe (55.6%, n=35) and the cerebellum (27%, n=17). The most frequent presentations included headache (50.8%, n=32), focal neurologic deficits (39.7%, n=25), and seizure (30.2%, n=19). Prevalence of headache (p=0.048) and seizure (p=0.016) decreased over time without treatment of the lesion. A second cerebrovascular lesion was identified in 12 patients (19%). Two patients had a symptomatic intracerebral hemorrhage attributable to their venous malformation. Risk of hemorrhage was 0.15% per lesion-year (95% CI, 0.06 to 0.38%). This study establishes that the natural history of venous malformations is benign, that the risk of hemorrhage from these lesions is negligible, and that conservative therapy is warranted.