Analysis of the incidence of death, hemorrhage, and neurological deficit in the treatment of intracranial arteriovenous malformations (AVMs): surgery versus other treatments - a systematic review.

To evaluate the incidence of mortality, hemorrhage, and neurological deficits in treating intracranial arteriovenous malformations (AVMs) in patients over 18 through a comparative analysis of surgical approaches and other therapeutic modalities. A systematic review was conducted using MEDLINE, Embase, CENTRAL, and LILACS databases in November 2023. Inclusion criteria included clinical trials, cohorts studies, case-controls studies, and case series comparing patients over 18 undergoing surgery or microsurgery versus other treatments (radiosurgery, isolated embolization, and conservative treatment). Exclusions criteria included studies on non-AVM diseases, non-intracranial AVMs, cavernous malformations, pregnant patients, and those treated with both radiosurgery and microsurgery. Twelve studies included 1,555 patients, with an average age of 42.65 years. Of these, 711 underwent surgery (with or without prior embolization), and 844 underwent other procedures. In the surgery group, the incidence of death was 1.69%, while in the other procedures group, it was 2.96%. The incidence of neurological deficits and hemorrhage in the surgery group was 11.67% and 4.22%, respectively. In other procedures, these incidences were both 9.12%. The average obliteration rate in the surgery group was 97.45%, compared to 49.77% for radiosurgery, 38.46% for isolated embolization, and 0.6% for conservative treatment. This systematic review highlighs that surgery achieves the highest AVM obliteration rates but carries a higher risk neurological déficits. Radiosurgery minimizes these risks, while conservative treatment offers advantages in mortality and hemorrhage reduction. No single modality proves to be universally superior, emphasizing the need for personalized approaches and further randomized trials to clarify comparative safety and efficacy.