gms | German Medical Science

Objective: Moyamoya angiopathy is a stenoocclusive disease with a progressive spontaneous occlusion of the circle of Willis resulting commonly in ischemic strokes. While conventional cerebral angiography and MRI is a standard in the diagnostic cascade of most treatment centers, functional imaging such as H215O-Positron emission tomography (PET) with acetazolamide challenge is not used routinely. We evaluated the correlation of severity of moyamoya disease on cerebral angiography and MRI with H215O-PET (Baseline and acetazolamide challenge) in the evaluation of surgical indication and planning.

Methods: We analyzed imaging data of adult Moyamoya patients who underwent treatment in our center between 2013-2016. All patients with a complete diagnostic imaging data set (MRI, 6-vessel cerebral angiography, H215O-PET with acetazolamide challenge) before a possible surgical intervention were included. Patients were anonymized and each imaging modality was analyzed separately by a single physician blinded to their identity using a scoring system for severity of disease in each hemisphere and arterial territory. MRI scoring included cortical, subcortical and watershed strokes; stenosis/occlusion of each major artery as well as the presence/absence of collaterals were scored on angiography; and a scoring of the baseline cerebral perfusion and reactivity to acetazolamide was used for H215O-PET. The scores of the three diagnostic imaging were then correlated.

Results: Twenty one adult patients (18 with bilateral Moyamoya disease, 3 with unilateral Moyamoya angiopathy) with 39 affected hemispheres were studied. PET and angiography correlated well in 31/39 (79%) hemispheres, but not in 8/39 (21%). In patients with good collateralization from deep MM collaterals and/or the PCA or ACA into the ACA and/or MCA territory, sufficient acetazolamide reactivity was seen in PET in 5/27 (19%), in patients with partial collateralization in 4/20 (20%) vascular territories. In case of absent collateralization on angiography, acetazolamide reactivity was sufficient in 12/37 (32%) vascular territories. There was no correlation between MRI and reserve capacity in PET.

Conclusion: Our analysis demonstrates that functional perfusion imaging should be obligatory in the treatment of patients with Moyamoya. Absence of functional perfusion imaging such as H215O-PET might lead to wrong treatment decisions in 21% of all hemispheres with stenosis/occlusion present on angiography. Visualizing angiographic collateralization alone might give an incorrect impression on the true perfusion reserves in 89% of all territories. The results show that treatment decision should not be based on the results of angiography and MRI alone.