J Neurol Surg A Cent Eur Neurosurg 2019; 80(03): 187-197
DOI: 10.1055/s-0039-1678710
Original Article
Georg Thieme Verlag KG Stuttgart · New York

Correlation of Appearance of MRI Perinidal T2 Hyperintensity Signal and Eventual Nidus Obliteration Following Photon Radiosurgery of Brain AVMs: Combined Results of LINAC and Gamma Knife Centers

Osama Abdelaziz
1  Department of Neurosurgery, Alexandria University, Alexandria, Egypt
,
Ahmed Shereen
1  Department of Neurosurgery, Alexandria University, Alexandria, Egypt
,
Takuro Inoue
2  Department of Neurosurgery, Koto Kinen Byoin, Higashiomi, Shiga, Japan
,
Hisao Hirai
2  Department of Neurosurgery, Koto Kinen Byoin, Higashiomi, Shiga, Japan
,
Ayako Shima
2  Department of Neurosurgery, Koto Kinen Byoin, Higashiomi, Shiga, Japan
› Author Affiliations
Further Information

Publication History

09 January 2018

31 July 2018

Publication Date:
20 March 2019 (eFirst)

Abstract

Background A wide variety of radiologic changes occur within and adjacent to the nidus of arteriovenous malformations (AVMs) after stereotactic radiosurgery (SRS). Our objective was to study the magnetic resonance imaging(MRI)-defined changes following photon radiosurgery of AVMs and specifically to correlate the appearance of a perinidal T2 hyperintensity signal with the eventual angiographic obliteration of an AVM nidus in response to SRS treatment.

Material and Methods This retrospective study was conducted on 62 patients with brain AVMs who received photon SRS treatments between 2004 and 2017, using either a technique based on a linear accelerator at the Alexandria LINAC Radiosurgery Center in Egypt (21 patients/AVMs) or a technique based on a gamma unit at the Koto Memorial Gamma Knife Center in Japan (41 patients/AVMs). All patients included in the study had serial clinical and radiologic follow-ups for ≥ 2 years after SRS treatments.

Results In the combined study series of 62 patients/AVMs treated with photon SRS, the follow-up MRIs revealed that 50 AVMs (80.6%) showed nonvisualized nidus and 12 AVMs (19.4%) showed decreased nidus size. Radiation-induced changes, defined as appearance of perinidal T2 hyperintensities in post-SRS MRIs, occurred in 34 patients (54.8%). Of the 35 patients with available follow-up angiographic studies, 30 AVMs (85.7%) demonstrated complete nidus obliteration at a mean of 36 months (range: 8–66 months) after SRS. Of the 30 AVMs with both MRI evidence of a nonvisualized nidus and angiographic verification of complete nidus obliteration, 20 AVMs (66.7%) were associated with prior MRI evidence of the appearance of a perinidal T2 hyperintensity signal at an average of 12 months (range: 6–45 months) after SRS. Of the five AVMs with both MRI evidence of decreased nidus size and angiographic verification of partial nidus obliteration, four AVMs (80%) showed perinidal T2 hyperintensity signal on post-SRS follow-up MRIs. Lower Spetzler-Martin grade (p = 0.013), smaller AVM volume (p = 0.017), and appearance of post-SRS perinidal T2 hyperintensity signal (p = 0.007) were the statistically significant independent predictors of AVM obliteration. The appearance of perinidal T2 hyperintensity signal in the post-SRS MRIs had a sensitivity of 66.7%, a specificity of 20%, and an overall accuracy of 60% in predicting the eventual obliteration of the AVM nidus.

Conclusions The present study may help improve our current understanding of the mechanisms behind the radiation-induced tissue changes following AVM SRS. Because the SRS-induced hemodynamic changes within the AVM nidus initiate the cascade of the subsequent formation of perinidal vasogenic brain edema, the appearance of perinidal high T2 signal in the follow-up MRIs after SRS would be a valuable indicator of the AVM response to SRS. The development of perinidal hyperintensity was the strongest predictive factor of AVM obliteration (p = 0.007), with relatively high sensitivity (66.7%) and accuracy (60%) and fairly low specificity (20%), as a prognostic sign of eventual complete angiographic obliteration of the AVM nidus following SRS.