Intracranial Atherosclerosis: A Few Good Images?

Lieutenant Daniel Kaffee, U.S. Navy: “I want the truth!”Colonel Nathan Jessup, U.S. Marine Corps: “You can’t handle the truth!”A Few Good Men, Columbia Pictures, 1992

Physicians want the truth when imaging patients suspected of having intracranial stenosis. Currently, four imaging modalities might be considered as pathways to the truth in cases of intracranial atherosclerosis. These modalities are digital subtraction angiography (DSA), CT angiography (CTA), MR angiography (MRA), and transcranial Doppler (TCD). Bash et al have done an unprecedented comparison of three of these modalities in 28 patients with intracranial stenosis. On the basis of their study, it is clear that, compared with CTA and DSA, MRA generally does not get us as close to the truth about intracranial stenosis.

Should we be surprised that MRA does not accurately depict intracranial atherosclerosis? Not really. It is a firmly established fact that MRA tends to overestimate degree of stenosis (1–3). MRA has poor spatial resolution relative to what is now available for CTA and DSA, so we cannot reasonably expect to image stenotic arteries reliably with a lumen of <1 mm. Three-dimensional time-of-flight MRA is susceptible to artifacts secondary to turbulent flow, and some degree of turbulent flow is generally present with stenotic intracranial atherosclerosis. Even normal arteries can be misrepresented on MRA, because curves in normal arteries can cause turbulence that creates an artifactual stenosis. These artifactual stenoses account for the poor positive predictive value of MRA for intracranial atherosclerosis. No contrast or ionizing radiation is needed for an MRA, but how much harm really comes to patients from the use of iodinated contrast material or ionizing radiation? Have we helped a patient by avoiding iodinated contrast material and radiation in exchange for an inaccurate diagnosis? Patient care based on an incorrect diagnosis, no matter how caring and well intentioned, is much more likely to fail than care based on the correct diagnosis.

DSA has traditionally been the criterion standard for imaging intracranial disease. It offers superb spatial resolution and contrast resolution. But DSA is an invasive procedure that carries a small but real (0.7%) risk of permanent neurologic deficit (4). DSA gives physiologic information about flow contribution from the injected artery. This physiologic effect is sometimes a disadvantage, because slow-flow vessels distal to a stenosis may be poorly filled with contrast material and thus poorly visualized. Multiple arteries often need to be injected to show collateral blood flow. For posterior circulation stenosis, both vertebral arteries generally need to be evaluated. For those of us who work in referral centers where complex patients whose symptoms are refractory to medical therapy are referred, angioplasty or bypass surgery is sometimes offered. These patients will all have conventional angiography as part of their preintervention evaluation.

As Bash et al have shown, CTA can give excellent anatomic visualization of intracranial atherosclerosis. The use of CT angiography does not avoid the use of contrast material or ionizing radiation, but these offer trivial risks in most patients relative to the potential risks associated with symptomatic intracranial atherosclerosis. Spatial resolution may occasionally limit our ability to distinguish very severe stenosis from occlusion compared with DSA. Calcium might also occasionally cause overestimation of a stenosis, as was described for CTA of the cervical carotid artery (5).

But perhaps we cannot quite handle the truth yet. There is certainly confusion about the best medical therapy for intracranial atherosclerosis. The latest results of WASID (Warfarin versus Aspirin Symptomatic Intracranial Disease) indicate that coumadin offers no benefit over aspirin (6). Newer antiplatelet agents such as clopidogrel have not yet been subjected to rigorous testing for efficacy in the treatment of intracranial atherosclerotic disease. Patients with ischemic symptoms may get the same antiplatelet therapy regardless of the appearance of intracranial arteries on imaging. Nevertheless, we should strive to keep the art of diagnostic imaging ahead of the art of therapy. Treatments targeted to a patient’s specific disease can be developed only if we can reliably diagnose that disease.

• Copyright © American Society of Neuroradiology