Clinical Utility of Magnetic Resonance Imaging Radiographs for Suspected Organic Syndromes in Adult Psychiatry

Stephen M. Erhart, M.D.; Alexander S. Young, M.D., M.S.H.S.; Stephen R. Marder, M.D.; and Jim Mintz, Ph.D.

Objective: In psychiatric practice, adult patients are most commonly referred for magnetic resonance imaging (MRI) to screen for suspected organic medical diseases of the central nervous system that can mimic psychiatric syndromes. We identified the most common signs and symptoms prompting MRIs to establish the predictive value of these signs and symptoms for clinically pertinent organic syndromes.

Method: This study was a retrospective chart review of psychiatric patients at the Veterans Affairs Greater Los Angeles Health Care Center (Los Angeles, Calif.) who were referred for MRI of the brain between 1996 and 2002. Patients referred for evaluation of dementia were excluded. The specific indications leading clinicians to obtain MRI were identified and grouped. In order to offset the uncertain significance of many MRI findings, for this study, the predictive value of each indication was calculated based on the percentage of patients in whom clinical management changed in response to MRI findings rather than on the percentage with any abnormal MRI results.

Results: Of 253 patients who had MRIs, 38 (15%) incurred some degree of treatment modification as a result of MRI findings, including 6 patients in whom MRI identified a medical condition that became the focus of treatment. Six indications appeared most likely to prompt clinicians to obtain MRIs. Because pertinent results were associated with each of these indications, statistical evaluation did not reveal significant differences in their predictive values (c² = 4.32, df = 5, p = .505).

Conclusions: Unlike prior studies showing no value to screening radioimaging, this study shows MRI can be a useful screening test among patients suspected of having organic psychiatric disorders and that the common indications for MRI employed at one institution were predictive.

(J Clin Psychiatry 2005;66:968-973)

Received March 23, 2004; accepted Jan. 31, 2005. From the Department of Veterans Affairs Health Care Center (Greater Los Angeles), Los Angeles, Calif.

This project was supported by the Veterans Affairs VISN 22 Mental Illness Research, Education, and Clinical Center (MIRECC).

Drs. Erhart, Young, Marder, and Mintz report no other financial affiliation or other relationship relevant to the subject matter of this article.

Corresponding author and reprints: Stephen M. Erhart, M.D., Department of Veterans Affairs Health Care Center (Greater Los Angeles), 11301 Wilshire Boulevard, MIRECC Building 210, Los Angeles, CA (e-mail: serhart@ucla.edu).