Comparative Effectiveness of Surgery, Embolization, and Radiosurgery for Intramedullary Glomus Arteriovenous Malformations: A Systematic Review and Meta-Analysis.
Nathan J Bekelman, Abdul Karim Ghaith, Taha Khalilullah et al.
✦ AI-curated · Sources linked
In 30 seconds
This systematic review and meta-analysis compared surgery, endovascular embolization, and radiosurgery for spinal intramedullary glomus arteriovenous malformations. It included 36 studies comprising 462 patients. Complete nidus obliteration was more frequent after surgery (77%) than after embolization (39%) or radiosurgery (16%) (P < .001). The analysis found no significant differences among modalities in partial obliteration, neurologic symptom improvement, or treatment-related complications. The authors note that treatment comparisons are limited by heterogeneity in lesion classification and that the highest surgical obliteration rates were observed in selected, typically compact lesions.
Key findings
- Complete obliteration rates were 77% after surgery, 39% after embolization, and 16% after radiosurgery (P < .001).
- Partial obliteration (P = .19), symptom improvement (P = .94), and complication rates (P = .35) did not differ significantly among modalities.
- Preoperative embolization did not significantly change complete obliteration rates among surgical patients (P = .55).
- Meta-regression found that complete obliteration rates after embolization increased over time.
Why it matters
These findings provide comparative outcome estimates for treatment options in a rare spinal vascular lesion. They suggest surgery achieves more frequent complete obliteration in selected lesions, while measured symptom improvement and complication rates were not significantly different across modalities. The results also highlight the need to interpret pooled comparisons cautiously because lesion classification varied across studies.
What to keep in mind
The evidence is based on pooled studies with heterogeneous lesion classification, limiting definitive treatment comparisons. The abstract does not report randomised comparisons.