Right frontal cortical-subcortical AVM
This 31 year old male presented with sudden onset of headaches followed by left hemiparesis.
Emergency CT examination revealed a right fronto-parietal intracerebral hematoma. At conventional, catheter based cerebral angiography an AVM was found with an associated false aneurysm in the nidus [36].
Two month later, after clinical recovery, a follow-up MRI-MRA examination was performed. The AVM was unchanged, however the previously identified false aneurysm was not visualized, suggesting thrombosis. This was confirmed by subsequent repeat conventional angiography performed during the first embolization session of the lesion.
Right frontal cortical-subcortical AVM, 0.5 T
Examination 1
Fig.1 Transverse non-enhanced CT images. Abnormal slightly hyperdense serpiginous structures are seen in proximity to the right lateral ventricle (above). Moreover, a rounded hyperdense intracerebral lesion, surrounded by a hypodense halo is also detected in the right centrum semiovale, suggesting a subacute intraparenchymal hematoma with perifocal edema.
Fig.2 Transverse contrast-enhanced CT images. The previously seen periventricular serpiginous structures exhibit marked enhancement and are consistent with an arteriovenous malformation.
Examination 2
Fig.3 DSA images after selective injection of the right internal carotid artery. The lateral views (above and middle) clearly demonstrate the arteriovenous malformation fed by the pericallosal branches of the anterior cerebral artery. A typical pseudoaneurysm is also detected (arrow) in conjunction with the nidus, presumably corresponding to the site of rupture of the nidus. Same observations on the A-P view (below).
Examination 3 (3 month follow-up)
Fig.4 Sagittal T1-weighted spin-echo images. The nidus of the AVM is well demonstrated on the medial (interhemispheric) surface of the left hemisphere in the posterior frontal region. A semioval high signal intensity structure is detected at the upper margin of the signal void nidus (arrow).
Fig.5 Transverse proton density weighted fast spin-echo images. The signal void components of the nidus are particularly well seen on these images. An abnormal, high signal intensity area is detected in the adjacent cerebral white matter. A rounded, inhomogeneous structure with a well defined and slightly hypointense rim is noted above the nidus (below right). The topography of the latter coincides with that of the pseudoaneurysm, detected by the previous cerebral angiography.
Fig.6 Transverse T2-weighted fast spin-echo images. Same observations as on Fig.5.
Fig.7 Sagittal survey 2D PC MR angiograms with different velocity encoding values. Above left: 20 cm/s, above right: 40 cm/s, below left: 60 cm/s, below right: 80 cm/s. On the basis of this pilot study, this appears to be a relatively low flow velocity lesion, therefore the 40 cm/s Venc value was selected for the subsequent 3D PC MRA acquisition.
Fig.8 Sagittal averaged modulus (left) and magnitude of complex differences (right) type source images from a Gadolinium-enhanced 3D PC MRA acquisition. The cortical topography of the lesion is well appreciated on the anatomical images (left). The pseudoaneurysm is also identified (arrow), however, no flow is detected within it on the basis of the corresponding flow image (below right). This suggests spontaneous intraaneurysmal thrombosis, but could be theoretically related to an inappropriate Venc value selection as well.
Fig.9 Sagittal targeted MIP reconstruction from the Gadolinium-enhanced 3D PC MRA acquisition data set (Venc: 40 cm/s). The feeders (red arrows) of the AVM arise from the anterior cerebral artery. The draining veins (blue arrows) lead towards the vein of Galen, as well as the superior sagittal sinus. The pseudoaneurysm is not seen.
Fig.10 Transverse averaged modulus (above) and magnitude of complex differences (below) type source images from a second Gadolinium-enhanced 3D PC MRA acquisition. Verification of the previous findings. No intravascular signal is seen within the pseudoaneurysm (arrow).
Fig.11 Transverse collapsed (left) and targeted (right) MIP reconstructions from the second Gadolinium-enhanced 3D PC MRA acquisition. This view allows optimal visualization of the inferior draining vein of the AVM, leading towards the vein of Galen.
Examination 4 (2 days after the MRI-MRA examination)
Fig.12 DSA images (lateral views) after selective injection of the right internal carotid artery. Unchanged appearance of the AVM except for the absence of visualization of the pseudoaneurysm, confirming the hypothesis of spontaneous thrombosis based on the prior MRI-MRA examination.
Examination 5 (post-embolization follow-up)
Fig.13 DSA images (lateral views) after selective injection of the right internal carotid artery. Most of the nidus has disappeared, compared to the pre-embolization images (Fig.12).
Examination 6 (2 month post-embolization follow-up)
Fig.14 Sagittal T1-weighted spin-echo images. The nidus of the AVM appears to be smaller than on the previous MRI examination.
Fig.15 Transverse proton density weighted fast spin-echo images. The residual abnormal vascular structures of the nidus are well depicted on these images. Note that the associated high signal intensity area in the adjacent white matter has also disappeared, confirming the beneficial effect of embolization, even if partial only.
Fig.16 Transverse T2-weighted fast spin-echo images. Same observations as on Fig.15.
Fig.17 Sagittal averaged modulus (left) and magnitude of complex differences (right) type source images from a Gadolinium-enhanced 3D PC MRA acquisition. The reduction of the size of the nidus is well demonstrated (compare to Fig.8).
Fig.18 Sagittal targeted MIP reconstruction from the Gadolinium-enhanced 3D PC MRA acquisition data set (Venc: 40 cm/s). The nidus is reduced in size and the draining veins appear to be less congested as well.
Fig.19 Sagittal targeted MIP reconstruction from the pre- and postembolization Gadolinium-enhanced 3D PC MRA acquisition data sets, for comparison. Same observations as on Fig.18.
The ESNR CD-Rom Series