Background and Purpose The typical magnetic resonance imaging picture
of arterial dissection, namely, a narrowed eccentric signal void surro
unded by a semilunar signal hyperintensity (corresponding to the mural
hematoma) on T1- and T2-weighted images, has been repeatedly reported
, but the sensitivity of magnetic resonance imaging for the diagnosis
of cervical dissection is poorly known. Another technique, dynamic com
puted tomography, may provide evidence of mural hematoma, but there ha
s been no systematic evaluation of this technique. The aims of this st
udy were to assess both the sensitivity of routine 0.5-T magnetic reso
nance imaging for the detection of a typical picture of cervical arter
y dissection and the value of dynamic computed tomographic scans to pr
ovide evidence of dissecting hematoma. Methods Fifteen consecutive pat
ients with angiographically confirmed extracranial internal carotid (n
=9) or vertebral (n=10) dissections were studied using a standardized
0.5-T spin-echo magnetic resonance imaging protocol with axial slices.
Twelve of these patients had dynamic computed tomographic scans at th
e site of the dissection suggested by angiography. Results A typical m
agnetic resonance imaging picture of cervical artery dissection was ob
served in 12 of 15 (80%) patients and in 13 of 19 (68%) dissected vess
els. The sensitivity of magnetic resonance imaging was higher in inter
nal carotid (78%) than in vertebral (60%) dissections and in stenotic-
type dissections (85%) than in occlusive or aneurysmal-type dissection
s. The dynamic computed tomographic scan showed the mural hematoma in
11 of the 12 (92%) patients and in 12 of 15 (80%) dissected vessels. C
onclusions Routine 0.5-T magnetic resonance imaging with axial slices
is a sensitive technique for the diagnosis of dissection, but in about
20% of patients with cervical artery dissection magnetic resonance im
aging will demonstrate no typical abnormality. Dynamic computed tomogr
aphic scans are a sensitive neuroimaging procedure to confirm the pres
ence of the mural hematoma, but it needs to be directed by prior angio
graphy.