Background: To evaluate clinical and MR features of de novo lesions (DNL) in the familial form of cerebral cavernous malformation (CCM) in 40 patients belonging to 29 unrelated non-Hispanic families.

Methods: Forty patients followed up by serial cerebral MR examinations were included in this retrospective study. First and last available MR examinations were retrospectively analyzed and compared for each patient to diagnose DNL. Gradient-echo (GRE) sequences were performed in only 11 of the 40 patients and were not considered for this study. Incidence of DNL was evaluated in terms of lesions/patient-year. All DNL were characterized by their clinical and MR features (location, size, type). Type of CCM was determined according to the classification of Zabramski (1994). Patient groups with and without DNL were compared for sex, age, number of pre-existing CCMs, and follow-up.

Results: Twenty-three DNL were recorded in 11 patients (27.5%) and the incidence was 0.2 lesions/patient-year (mean follow-up = 3.2 years). All but one DNL were asymptomatic. Twenty DNL were supratentorial and three were infratentorial. Mean diameter was 8 mm (2-35 mm). Six DNL were classified as type 1 (subacute hemorrhage), six as type 2 (hemorrhages and thromboses of varying ages) and 11 as type 3 (chronic hemorrhage with hemosiderin staining). No statistical difference between groups was found in terms of sex, age, or number of pre-existing CCMs. On the other hand, duration of follow-up was significantly longer in the group with DNL.

Conclusion: The occurrence of DNL seems to be a hallmark of the familial form of CCM in non-Hispanic families as well as in Hispanic families. Such DNL are usually asymptomatic and are mainly classified as type 3 (chronic hemorrhage with hemosiderin staining). Within the limits of the retrospective study design and potential selection bias introduced by the varying indications for MR scanning, it does seem that DNL may occur at any time in the lifespan of CCM patients, and occurrence does not seem to depend on age, sex, or the total number of pre-existing lesions.