Regression of Periventricular Anastomosis After Indirect Revascularization in Adult Ischemic Moyamoya Disease With Misery Perfusion

BACKGROUND AND OBJECTIVES: To investigate whether indirect revascularization is associated with regression of periventricular anastomosis (PA) in adult patients with ischemic moyamoya disease (MMD) with misery perfusion and to identify factors associated with PA regression. METHODS: Adult patients with ischemic MMD with misery perfusion who underwent indirect revascularization between 2011 and 2024 were included. Magnetic resonance angiography and arterial spin labeling were performed preoperatively and at 1 year postoperatively to assess postoperative collateral formation and PA status and to quantify cerebral blood flow. Associations between postoperative collateral formation, PA regression, and baseline factors, including age, arterial spin labeling-cerebral blood flow, and the RNF213 p.R4810K variant, were evaluated. Hemorrhagic events during long-term postoperative follow-up were also assessed. RESULTS: A total of 108 hemispheres in 87 patients were included. Good postoperative collateral formation was observed in 70.4% of hemispheres and was significantly associated with PA regression after indirect revascularization. Complete regression of choroidal PA (CHO-PA) was independently associated with better postoperative collateral formation in the multivariable analysis (odds ratio 4.17; 95% CI: 1.55-12.11, P = .005). The RNF213 p.R4810K variant was significantly associated with CHO-PA regression. During long-term follow-up, recurrent hemorrhage originating from a remnant PA was rare (0.57%/person-year). CONCLUSION: Indirect revascularization in adult-onset ischemic MMD with misery perfusion was associated with favorable collateral formation and regression of PAs, particularly CHO-PA. The RNF213 p.R4810K variant may be associated with these changes. Further studies with longer follow-up are needed to determine whether PA regression is associated with a reduced risk of future hemorrhage.