The optimal utilization of magnetic resonance imaging in neuro-ophthalmic d
iagnosis is limited by errors in prescribing and interpreting scans. In a r
eview of case material, we discovered four common prescriptive errors: I) f
ailure to apply a dedicated study, 2) inappropriate use of a dedicated stud
y, 3) omission of intravenous contrast, and 4) omission of specialized sequ
ences. The four common interpretive errors were the following: 1) failure t
o detect the lesion because of misleading clinical information, 2) rejectio
n of a clinical diagnosis because an expected imaging abnormality was absen
t, 3) assumption that a striking imaging abnormality accounted for the clin
ical abnormality, and 4) failure to consider the lack of clinical specifici
ty of imaging abnormalities. Many of these errors could be avoided by impro
ved communication between clinicians and radiologists. (C) 2000 by Elsevier
Science Inc. All rights reserved.