Abstract Background Pediatric arteriovenous malformations (AVMs), vein of Galen malformations (VOGMs), and arteriovenous fistulas (AVFs) are rare high-flow shunts with risks of intracranial hemorrhage, seizures, venous hypertension, and heart failure. Current endovascular management is still challenged by significant gaps in timing, approach, material used, safety, and efficacy, and most evidence is extrapolated from adult series with no focus on resource-limited settings. The objective of this study is to bridge the existing research gaps by evaluating the safety and efficacy of endovascular treatment specifically for pediatric arteriovenous shunts. Methods A mixed cohort study was conducted on patients younger than 18 years with AVMs, VOGMs, or AVFs, consisting of 118 procedures. Clinical outcomes, angiographic occlusion, and predictors of incomplete occlusion were analyzed following standardized embolization protocols. Results Complete angiographic obliteration was achieved in 48.3% of AVMs, 37.5% of AVFs, and 14.3% of VOGMs. Yet clinical improvement was 95%, 100%, and 85.7%, respectively. Major and minor complications occurred in 3.4% and 5% of cases, respectively. Conclusion Although angiographic occlusion was reached in 48.3% of AVM therapeutic procedures (59% of patients), meaningful functional gains occurred despite partial occlusion, highlighting the benefit of targeted, staged embolization. Lesion complexity, characterized by mixed or multiple venous drainage, dural arterial supply, and substantial nidus size, was indicative of incomplete occlusion, thereby facilitating informed patient counseling and the establishment of realistic expectations. This structured, data-driven approach enhances postoperative outcomes, patient satisfaction, and personalized pediatric cerebrovascular care, demonstrating that endovascular therapy can be safely and effectively applied across resource-variable environments with standardized protocols and experienced operators. Graphical abstract
El-shafei et al. (Fri,) studied this question.