Pure endoscopic resection of brainstem cavernous malformations: a systematic review of current evidence and future directions

Abstract

Background: Brainstem cavernous malformations (BCMs) are high-risk vascular lesions characterized by a significant tendency for recurrent hemorrhage and devastating neurological morbidity. While open microsurgery is the traditional gold standard, purely endoscopic approaches have emerged as minimally invasive alternatives. This study aims to systematically review the surgical effectiveness, safety, and clinical outcomes of purely endoscopic BCM resection. Methods: Following PRISMA guidelines, a systematic search was conducted across Web of Science, MEDLINE, Scopus, and Embase for all eligible studies published from database inception until October 21, 2025. Data on patient demographics, lesion characteristics, surgical corridors, extent of resection (EOR), and complications were extracted. Risk of bias was assessed using the Joanna Briggs Institute tool. Descriptive statistical pooling and data stratification by surgical approach corridors were performed using R software to calculate pooled clinical frequencies and symptom recovery rates. Results: Twenty-six studies (22 case reports, 4 case series) involving 37 patients were included. The cohort was predominantly female (64%, n = 23 of 36 with documented gender), with a mean age of 41 years. Lesions were primarily located in the pons (76%, n = 28). The endoscopic endonasal transclival approach (EEA-TC) was the most frequent corridor (75.7%, n = 28), followed by various transcranial endoscopic routes. Preoperative diffusion tensor imaging (DTI) was utilized in a subset of cases, consistently demonstrating corticospinal tract displacement. Gross total resection (GTR) was achieved in 89.2% of the cohort (n = 33). Postoperative neurological improvement was documented in 91.9% of patients (n = 34), with no instances of re-bleeding during follow-up. The most significant complication was postoperative cerebrospinal fluid (CSF) leakage, occurring in 16.2% of patients (n = 6), with revision surgery required in 8.1% (n = 3). Other complications included transient neurological worsening and aseptic meningitis. Conclusion: Purely endoscopic resection of BCMs is a technically feasible and effective treatment modality, particularly for ventral and ventromedial pontine lesions. It offers high GTR rates and favorable neurological recovery while avoiding the brain retraction associated with open microsurgery. However, the risk of CSF leakage remains a primary limitation of the endonasal corridor. Future prospective, multi-center registries are necessary to further define patient selection criteria and long-term comparative outcomes against traditional microsurgical standards.