Early diffusion-weighted MRI lesions after treatment of unruptured intracranial aneurysms:

J Neurosurg 126:1070–1078, 2017

Diffusion-weighted MRI was used to assess periprocedural lesion load after repair of unruptured intracranial aneurysms (UIA) by microsurgical clipping (MC) and endovascular coiling (EC).

METHODS Patients with UIA were assigned to undergo MC or EC according to interdisciplinary consensus and underwent diffusion-weighted imaging (DWI) 1 day before and 1 day after aneurysm treatment. Newly detected lesions by DWI after treatment were the primary end point of this prospective study. Lesions detected by DWI were categorized as follows: A) 1–3 DWI spots < 10 mm, B) > 3 DWI spots < 10 mm, C) single DWI lesion > 10 mm, or D) DWI lesion related to surgical access.

RESULTS Between 2010 and 2014, 99 cases were included. Sixty-two UIA were treated by MC and 37 by EC. There were no significant differences between groups in age, sex, aneurysm size, occurrence of multiple aneurysms in 1 patient, or presence of lesions detected by DWI before treatment. Aneurysms treated by EC were significantly more often located in the posterior circulation (p < 0.001). Diffusion-weighted MRI detected new lesions in 27 (43.5%) and 20 (54.1%) patients after MC and EC, respectively (not significant). The pattern of lesions detected by DWI varied significantly between groups (p < 0.001). Microembolic lesions (A and B) found on DWI were detected more frequently after EC (A, 14 cases; B, 5 cases) than after MC (A, 5 cases), whereas C and D were rare after EC (C, 1 case) and occurred more often after MC (C, 12 cases and D, 10 cases). No procedure-related unfavorable outcomes were detected.

CONCLUSIONS According to the specific techniques, lesion patterns differ between MC and EC, whereas the frequency of new lesions found on DWI is similar after occlusion of UIA. In general, the lesion load was low in both groups, and lesions were clinically silent. Clinical trial registration no.: NCT01490463 (clinicaltrials.gov)

%d bloggers like this: