Received: 18 December 2002 Revised: 18 March 2003 Accepted: 31 March 2003 Published online: 3 June 2003 © ISS 2003 Abstract Objective: To determine the usefulness of radiography for interpretation of musculoskeletal (MSK) magnetic resonance imag- ing (MRI) studies. Designs and patients: In a 1-year period, 1,030 MSK MRI studies were performed in 1,002 patients in our institution. For each study, the interpreting radiologist complet- ed a questionnaire regarding the availability and utility of radio- graphs, radiological reports and clin- ical information for the interpreta- tion of the MRI study. Results: Ra- diographs were essential, very im- portant or added information in 61–75% of all MSK MRI cases. Ra- diographs were judged as essential for reading of MRI studies more of- ten for trauma, infection/inflamma- tion and tumors than for degenera- tive and miscellaneous/normal diag- noses (χ 2 =60.95, df=16, P<0.0001). The clinical information was rated as “essential” or “useful” signifi- cantly more often than not (χ 2 =93.07, df=16, P<0.0001). The clinical and MRI diagnoses were the same or partially concordant signifi- cantly more often for tumors than for trauma, infection/inflammation and degenerative conditions, while in the miscellaneous/normal group they were different in 64% of cases. When the diagnoses were different, there were more instances in which radiographs were not available. Conclusions: Radiographs are an important, and sometimes essential, initial complementary study for reading of MSK MRI examinations. It is highly recommended that radio- graphs are available when MSK MRI studies are interpreted. Keywords Radiograph · Magnetic resonance imaging (MRI) · Tumor · Infection · Fracture · Calcification · Chondrocalcinosis Skeletal Radiol (2003) 32:403–411 DOI 10.1007/s00256-003-0648-7 ARTICLE Mihra S. Taljanovic Tim B. Hunter Kimberly A. Fitzpatrick Elizabeth A. Krupinski Thomas L. Pope Musculoskeletal magnetic resonance imaging: importance of radiography Introduction In our experience, it is not uncommon for a referring physician to request a musculoskeletal magnetic reso- nance imaging (MSK MRI) study for a patient even though the patient has not had a radiographic study of the area to be examined by the MRI examination. Ac- cording to the American College of Radiology (ACR) Appropriateness Criteria 2001, this is not an acceptable practice [1]. In our daily practice, radiographs were fre- quently unavailable at the time of the interpretation of MSK MRI studies. Clinical information was frequently insufficient, and we were forced to call the referring phy- sicians for additional information. This situation prompted us to conduct a 1-year study to determine the availability of radiographs and their utility for the interpretation of our department’s MSK MRI studies. We correlated the radiological and clinical diagnoses for different disease categories and assessed the diagnostic agreement between the clinical and radio- logical diagnoses based on the availability of prior radio- logical information (reports and films, films only, reports only, or neither). We also analyzed the utility of the clini- cal information for interpreting the MSK MRI studies. M. S. Taljanovic ( ✉ ) · T. B. Hunter K. A. Fitzpatrick Department of Radiology, The University of Arizona Health Sciences Center, 1501 N. Campbell Avenue, Tucson, AZ 85724, USA e-mail: mihra@radiology.arizona.edu Tel.: +1-520-6267136 Fax: +1-520-6261945 E. A. Krupinski Department of Radiology, University of Arizona, 1609 N. Warren, Building 211, Tucson, AZ 85724, USA T. L. Pope Jr. Department of Radiology, Medical University of South Carolina, 169 Ashley Avenue, P.O. Box 250322, Charleston, SC 29425, USA