Hemorrhage during pregnancy in the latency interval after stereotactic radiosurgery for arteriovenous malformations

Object. The presentation for patients with arteriovenous malformations (AVMs) is often intracranial hemorrhage; for women, this frequently occurs during the prime childbearing years. Although previous studies have addressed the risk for AVM hemorrhage during pregnancy, such studies have not assessed the risk for hemorrhage among women who become pregnant during the latency interval between stereotactic radiosurgery (SRS) and documented obliteration of the lesion. The authors sought to evaluate the risk for hemorrhage in patients who become pregnant during the latency interval after SRS. Methods. This single-institution retrospective analysis reviewed the authors' experience with Gamma Knife SRS during 1987-2012: During this time, 253 women of childbearing age (median age 30 years, range 15-40 years) underwent SRS for intracranial AVM. The median target volume was 3.9 cm(3) (range 0.1-27.1 cm(3)), and the median marginal dose was 20 Gy (range 14-38 Gy). For all patients, the date of AVM obliteration was recorded and the latency interval was calculated. Information about subsequent pregnancies and/or bleeding events during the latency interval was retrieved from the medical records and supplemented by telephone contact. Results. AVM obliteration was confirmed by MRI or angiography at a median follow-up time of 39.3 months (range 10-174 months). There were 828.7 patient-years of follow-up within the latency interval between SRS and the date of confirmed AVM obliteration. Among nonpregnant women, 20 hemorrhages occurred before AVM obliteration, yielding an annual hemorrhage rate of 2.5% for nonpregnant women during the latency interval. Among women who became pregnant during the latency interval, 2 hemorrhages occurred over the course of 18 pregnancies, yielding an annual hemorrhage rate of 11.1% for women who become pregnant during the latency interval. For the 2 pregnant patients who experienced hemorrhage, the bleeding occurred during the first trimester of pregnancy. Conclusions. The authors present the first series of data for women with intracranial AVMs who became pregnant during the latency interval after SRS. Hemorrhage during the latency interval occurred at an annual rate of 25% for nonpregnant women and 11.1% for pregnant women. The data suggest that pregnancy might be a risk factor for AVM hemorrhage during the interval between SRS and AVM obliteration. However, this suggestion is not statistically significant because only 18 patients in the study population became pregnant during the latency interval. To mitigate any increased risk for hemorrhage, patients should consider deferring pregnancy until treatment conclusion and AVM obliteration.