 |
|
|
 |
Detection of Erosions in the Rheumatoid Hand;
A Comparative Study of Multidetector Computerized Tomography versus Magnetic Resonance Scanning
DAVID PERRY, NEAL STEWART, NICK BENTON, ELIZABETH ROBINSON, SUE YEOMAN, JEFF CRABBE, and FIONA McQUEEN
ABSTRACT. Methods. Comparative CT and MR scans of the dominant wrist were obtained from 9 patients with RA and clinical examination was performed to assess disease activity. MR and CT scans were scored for erosions and MR scans for bone edema by 2 radiologists using a validated system. Radiographs of the hands and feet were also scored for erosions using the modified Sharp score. Results. In 117 of 135 (87%) sites there was concordance for erosions between MR and CT scans. At the remaining 18/135 sites (13%), erosions were identified by CT but not MR in 12/135 (9%) and by MR but not CT in 6/135 (4%). Partial volume artefacts on MR images and shifts in slice position were the most common reasons for erosion mismatch between MR and CT. The mean CT bone erosion score was significantly higher than the MR erosion score when individual bony sites were examined (p = 0.024), with the greatest difference being at the metacarpal bases. The total bone erosion score also tended to be higher on CT than MR [median scores of 20 (range 0–66) and 12 (0–51), respectively; p = 0.060]. MR and CT erosion scores correlated strongly with the total Sharp score (r = 0.93, p = 0.0002 and r = 0.94, p = 0.0002, respectively) and with the Disease Activity Score (MR: r = 0.77, p = 0.02; CT: r = 0.71, p = 0.03). Conclusion. Most erosions were detected using both modalities, but erosion scores were higher on CT than MR scans, especially at the metacarpal bases. It is possible that small erosions in some regions are more easily detected by CT because of its ability to clearly delineate cortical bony margins. (J Rheumatol 2005;32:256-67) Key Indexing Terms:
MAGNETIC RESONANCE IMAGING
From the Departments of Rheumatology and Radiology, Auckland District Health Board, and Department of Molecular Medicine, Auckland University, Auckland, New Zealand. Supported by grants from the Health Research Council of New Zealand, the Arthritis Foundation of New Zealand, the Auckland Medical Research Foundation, Lotteries Health, New Zealand, the Auckland Radiology Group, Sanofi-Winthrop, and the Auckland Regional Rheumatology Research Trust. D. Perry, BHB, MBChB, Senior Radiology Registrar; N. Stewart, MBChB, FRANZCR, Consultant Radiologist; N. Benton, BSc, MBBS, FRACP, Consultant Rheumatologist; E. Robinson, MSc, Biostatistician; S. Yeoman, Nurse Metrologist; J. Crabbe, MBChB, FRANZCR, Consultant Radiologist; F.M. McQueen, MBChB, MD, FRACP, Associate Professor in Rheumatology. Address reprint requests to Dr. F. McQueen, Department of Rheumatology, Building 7, Auckland Hospital, Private Bag 92024, Auckland, New Zealand. E-mail: f.mcqueen@auckland.ac.nz Submitted March 12, 2004; revision accepted September 20, 2004. |