Transcranial color-coded Doppler assessment of cerebral arteriovenous malformation hemodynamics in patients treated surgically or with staged embolization

摘要

Objective The etiology of hemodynamic disturbances following embolization or surgical resection of arteriovenous malformations (AVMs) has not been fully explained. The aim of the study was the assessment of the selected hemodynamic parameters in patients treated for cerebral AVMs using transcranial color-coded Doppler sonography (TCCS). Materials and methods Forty-six adult patients (28 males, 18 females, aged 41 ± 13 years, mean ± SD) diagnosed with AVMs who were consecutively admitted to the Department of Neurosurgery between 2000 and 2012 treated surgically or with staged embolization were enrolled in the study. All patients were examined with TCCS assessing mean flow velocity (Vm), the pulsatility index (PI) and vasomotor reactivity (VMR) in all main intracranial arteries. The examined parameters were assessed in the vessel groups (feeding, ipsilateral and contralateral to the AVM) and they were compared between the examinations, i.e. at admission, within 24 h after the first embolization or surgical resection (I control), and before the second embolization (II control). Results In feeders which were completely obliterated or surgically resected - I control examination showed a nonsignificant V m decrease. The difference between Vm before embolization and II control examination was significant (102.0 ± 47.8 cm/s vs 54.3 ± 19.4 cm/s, p 0.01). A significant increase in PI (0.72 ± 0.18 vs 0.94 ± 0.24, p 0.01) and VMR (1.80 ± 0.59 vs 2.78 ± 0.78, p 0.01) of feeding vessels was observed in I control. No further increase in PI or in VMR was observed. In embolized feeding vessels after partial AVM embolization I control examination showed a significant decrease in Vm (116.1 ± 32.6 cm/s vs 93.4 ± 33.0 cm/s, p 0.01). No further significant decrease in Vm was noted. The pulsatility index increased significantly (I control, 0.54 ± 0.11 vs 0.66 ± 0.15, p 0.01) and then decreased nonsignificantly (II control). No statistically significant differences were found in VMR values between pretreatment, I and II control examinations. Both Vm in the ipsilateral internal carotid artery and the ratio of Vm of the embolized vessel to Vm of the corresponding contralateral vessel were significantly higher in I control examination compared to II control examination (111.8 ± 44.0 cm/s vs 101.3 ± 40.6 cm/s, p 0.01; 1.63 ± 0.61 vs 1.37 ± 0.62, p 0.01; respectively). No statistically significant correlation was observed between the decrease in Vm or the increase in PI in the embolized vessels and the reduction of AVM volume. In the nonembolized feeding vessels after partial AVM embolization II control examination revealed the increase in Vm and a significant decrease in PI (0.71 ± 0.21 vs 0.62 ± 0.16, p 0.01) compared to I examination. No statistically significant changes in the VMR value in the nonembolized feeders between the pretreatment, I and II control examinations were noted. Conclusions The decrease in Vm and the increase in the PI in the embolized feeding vessels after the first complete embolization or surgical resection is observed, whereas the PI returned to normal values before Vm does. The observed decrease in Vm and an increase in the PI in embolized AVM feeders after complete or partial embolization do not correlate with the extent of embolization. In these vessels a relative increase in blood flow velocity is maintained within the first 24 h following embolization as compared to contralateral vessels. The increase in Vm is not related to disturbances in VMR. Blood redistribution to the nonembolized AVM feeders is observed after partial AVM embolization.