LeVonne Powell-Tillman , MD

30 November 2001

Clinical Decision Making Conference

Question: In patients with suspected CNS vasculitis, how does MRI/MRA compare with cerebral angiography for diagnosis?

Article: Greenan et al Cerebral Vasculitis MR Imaging and Angiographic Correlation Radiology 1992; 182: 65-72

Study design: Retrospective comparison of MR and angiographic findings in seven patients with presumed primary CNS angiitis diagnosed between April 1985 and December 1990.

Inclusion: presumptive diagnosis of primary CNS angiitis (based on abnormal neurologic exam in the absence of other cause and response to immunosuppressive agents); five males and two females ages three to 44 years old.

Exclusion: not explicitly described; however, the study only describes patients with presumed CNS angiitis

Are the results valid?

Was there independent, blinded comparison to a “gold standard”? No. The MR results are compared to angiography, but it is not indicated that independent, blinded analysis of MR and angiographic results was performed.

Is the patient sample appropriate? It represents all patients diagnosed with CNS angiitis in the stated time period.

Did test results influence the decision to perform the gold standard? All patients underwent MR and angiography; three also underwent brain biopsy, but the reasons why were not described.

Are methods described in sufficient detail? Radiographic techniques are described in detail.

What are the results? MR imaging was positive (positive being defined as a number of different findings) in all patients, yielding a sensitivity of 100%. Thirty-six lesions were identified on MR images among the patients. Thirty-two of these lesions were correlated with positive angiographic findings.

In addition, three patients underwent brain biopsy. In two patients, the results were negative. The third was positive for granulomatous angiitis.

Will these results help me in caring for my patients? Tough to say. CNS is a difficult diagnosis; even the gold standard (i.e brain biopsy) has sensitivity in the 50-60% range. Cerebral angiography serves a surrogate gold standard, but findings consistent with CNS angiitis (i.e., segmental narrowing or focal dilatation of vessels) are also found in a number of other disease states. However, MR is a safe, noninvasive , and well-tolerated test with high sensitivity, making it a reasonable initial diagnostic test for the evaluation of CNS vasculitis. A prospective analysis would be helpful.

Article: Wasserman et al Reliability of Normal Findings on MR Imaging for Excluding the Diagnosis of Vasculitis of the Central Nervous System Vasculitis AJR 2001; 177(2): 455-59.