Abstract
Paramedian pontine infarct (PPI) is usually attributed to basilar artery (BA) atherosclerosis. However, this hypothesis has thus far been supported only by post-mortem studies. The authors show that high-resolution MRI is a promising method that can detect BA plaques in patients with PPI at or near the origin of the penetrating artery, whereas MR angiograms may appear normal.
Paramedian pontine infarcts (PPIs) that reach the pontine basal surface are most often caused by occlusion of the penetrating branch. Branches can be occluded by an atherosclerotic plaque of the parent basilar artery (BA), causing blockage at the orifice of the branch, or by a junctional plaque extended from the BA into the proximal branch.1,2 In large prospective clinical studies, BA branch disease is the most frequent cause of PPI. The diagnosis is considered when there is no large artery stenosis or potential cardiac source of emboli.3,4 However, such a mechanism is only presumed because the plaque cannot be detected by usual technical investigations.
High-resolution MRI (HRMRI) has been useful in the detection of aortic and internal carotid artery plaques in vivo.5,6 We evaluated the yield of HRMRI in the detection of BA atherosclerotic plaques in patients with PPI.
Methods.
Patients.
We studied 24 consecutive patients who were admitted to our stroke unit between March 2003 and May 2004 with acute PPI extending to the basal surface detected by diffusion-weighted and/or T2-weighted MR sequences. All patients underwent a standard protocol of investigations, including brain MRI, MR angiography (MRA), 12-lead EKG, and transesophageal echocardiography, no later than 8 days after stroke event. Vascular risk factors were recorded. HRMRI was performed within 3 months of stroke onset.
MRI protocol.
Images were performed using a 1.5-T GE scanner (Twinspeed, GE Medical Systems, Waukesha, WI) with a standard quadrature radiofrequency head coil. To examine the basilar arterial wall, High-resolution T2- and time-of-flight- (TOF) weighted images were performed. For high-resolution T2 examination, 12 slices were acquired in an axial plane along the short axis of the BA with the following parameters: repetition time/echo time (TR/TE), 3,500/70; field of view (FOV), 12 × 12 cm2; thickness between 2 to 3 mm; 288 × 224 matrix; and number of excitations (NEX), 5. Best voxel size was 0.4 × 0.5 × 2 mm. Three-dimensional TOF images were obtained in an axial plane using TR/TE, 27/6.9; FOV, 24 × 16; thickness, 1.6 mm with 0.8 mm overlapping; matrix, 320 × 256; and NEX, 1. Angiographic data were reconstructed with a maximum intensity projection algorithm and three-dimensional displayed. No smoothing filter was applied. A zip 512 matrix was used to enhance spatial resolution.
Data analysis.
Three readers (I.K., P.L., and P.A.) independently performed a descriptive analysis of MRIs. Images were scored on a consensus. BA luminography was classified into five subtypes on three-dimensional and sources images: 1) “normal” when the lumen was regular; 2) “irregular” for small loss of parallelism; 3) “moderately stenosed” for segmental reduction of lumen with a residual flow; 4) “severely stenosed” if there was a segmental signal loss; and 5) “occluded” when no arterial flow was observed. Presence of BA plaques was assessed using axial TOF source sections and high-resolution T2-weighted images. Based on previous imaging reports, plaques appear as a crescent-shaped arterial wall signal abnormality that thickens the vessel wall.5,6 A plaque was classified as 1) “present” if a typical lesion was identified; 2) “possible” if the basilar wall was irregular without clear ringlike process; and 3) “absent” when the wall was strictly thin. The location of BA lesions was defined according to their level from the pontine infarct. Signal composition of the plaque was classified as homogeneous or heterogeneous. When the signal was heterogeneous, areas of high signal intensity, intermediate, and low intensity, as well as signal void foci, in reference to facial muscles, were recorded.
Results.
We studied 4 women and 20 men (mean age, 67.2 years; range, 40 to 89 years). Seventeen patients had hypertension, 5 had diabetes mellitus, 6 had hypercholesterolemia, and 12 were current or former smokers. Three patients had atrial fibrillation, and four had coronary artery disease.
Basilar luminography analysis.
On three-dimensional and source TOF MRAs, nine patients had normal basilar luminography, eight patients had irregular lumen, three patients had moderate stenosis, three patients displayed severe focal stenosis, and one patient had occlusion of the BA. The site of luminal irregularities or stenosis was always found at the level of pontine infarct.
Basilar arterial wall analysis.
On source TOF images, we did not detect any parietal thickening or crescent signal abnormality that could correspond to an atherosclerotic lesion except in one patient. In this case, focal isointense thickening of the BA wall was possible at the level of irregularities.
On high-resolution T2-weighted sections, atherosclerotic plaque detection was certain in 18 patients and was scored as “possible” in the remaining 6 patients. None had normal BA wall. Plaque was identified in all patients with high-grade stenosis (3/3) and moderate stenosis (3/3) and in seven of eight patients with basilar irregularities and in five of nine patients with normal basilar lumen on TOF sequences (figure). Signal component of large plaques was always heterogeneous, with a thin hypointense external layer, followed by a larger hyperintense band and in some lesions small signal void foci. Thinner plaques appeared more homogeneous but globally hyperintense. Plaque was scored “possible” in four of nine patients with normal BA, in one of eight patients with irregular BA, and in the patient with BA occlusion.
Figure. (A) Diffusion-weighted MRI (DWI) sequence: paramedian pontine infarct reaching the pontine surface (arrow). In the same patient, three-dimensional TOF axial source section displayed regular basilar artery (BA) wall and lumen (B). However, high-resolution T2 short axis slice along BA clearly unmasked a crescent-shaped plaque (white arrow) at the level of anteromedial pontine infarction (black arrow; C).
Discussion.
Our results showed in vivo the presence of atherosclerotic BA plaques at or near the origin of penetrating arteries in patients with PPI. Using HRMRI, a BA plaque was accurately identified in 75% of the patients and was strongly suspected in the other cases. These results are consistent with previous neuropathologic findings.1,2 The authors showed in three neuropathologic cases that a BA atherosclerotic plaque obstructed the origin of perforating arteries, causing PPI. Hence, BA branch disease could be more frequently associated with PPI than previously reported in clinical studies.3,4 Moreover, patients with other potential sources of emboli may also have BA atherosclerosis. For example, in our study HRMRI showed BA branch disease in two of three patients with atrial fibrillation.
In our study, one-half of the patients had either BA irregularities or normal lumen diameter on TOF sequences, whereas HRMRI revealed BA atherosclerotic plaques. In some patients, plaques reduced the luminal diameter up to 50%. This was likely explained by arterial adaptation to plaque development known as Glagov phenomenon.7 The advantage of HRMRI is to directly examine the vessel wall, whereas conventional procedures only show the luminogram. Therefore, this method can identify atherosclerotic plaques before their impact on the luminography. Moreover, axial TOF native slices did not help identify BA wall abnormalities compared with HRMRI. Previous studies on carotid arteries showed a good correlation between HRMRI plaque characteristics and histologic findings.5,6 Consistently, we could observe similar signal patterns but only for large BA lesions. However, pathologic analyses are required to confirm these HRMRI signal profiles.
Our study had several technical limitations. First, we could not establish a direct relationship between plaque and infarct because perforating arteries could not be identified. Nevertheless, BA plaques were located at the level of the pontine infarct in the neighborhood of the perforators, suggesting a causal link. However, technical improvements should be required to delineate small penetrating branches. Second, 25% of patients had a small parietal thickening projecting in the lumen. These morphologic aspects could correspond either to a plaque or to artifactual flow effects. Further prospective studies are needed to differentiate which BA plaques lead to occlusion or may be responsible for new events. HRMRI is the only current imaging method that may characterize intracranial vessel wall lesions. This MRI technique provides a noninvasive tool to study the burden and pathogenesis of intracranial atherosclerosis.
Footnotes
-
Received June 23, 2004. Accepted in final form September 23, 2004.
Disputes & Debates: Rapid online correspondence
REQUIREMENTS
If you are uploading a letter concerning an article:
You must have updated your disclosures within six months: http://submit.neurology.org
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.