COMPARISON OF MAGNETIC RESONANCE IMAGING, COMPUTED TOMOGRAPHY, AND RADIOGRAPHY FOR ASSESSMENT OF NONCARTILAGINOUS CHANGES IN EQUINE METACARPOPHALANGEAL OSTEOARTHRITIS JULIEN OLIVE,MARC-ANDRE ´ D’ANJOU,KATE ALEXANDER,SHEILA LAVERTY,CHRISTINE THEORET We compared the ability of 1.5 T magnetic resonance imaging (MRI), computed tomography (CT), and computed radiography (CR) to evaluate noncartilaginous structures of the equine metacarpophalangeal joint (MCP), and the association of imaging changes with gross cartilage damage in the context of osteoarthritis. Four CR projections, helical single-slice CT, and MRI (T1-weighted gradient recalled echo [GRE], T2 Ã -weighted GRE with fast imaging employing steady-state acquisition [FIESTA], T2- weighted fast spin echo with fat saturation, and spoiled gradient recalled echo with fat saturation [SPGR-FS]) were performed on 20 racehorse cadaver forelimbs. Osteophytosis, synovial effusion, subchondral bone lysis and sclerosis, supracondylar lysis, joint fragments, bone marrow lesions, and collateral desmopathy were assessed with each modality. Interexaminer agreement was inferior to intraexaminer agreement and was generally moderate (i.e., 0.4ojo0.6). Subchondral bone sclerosis scores using CT or MRI were correlated significantly with the reference quantitative CT technique used to assess bone mineral density (Po0.0001). Scores for subchondral lysis and osteophytosis were higher with MRI or CT vs. CR (Po0.0001). Although differences between modalities were noted, osteophytosis, subchondral sclerosis, and lysis as well as synovial effusion were all associated with the degree of cartilage damage and should be further evaluated as potential criteria to be included in a whole-organ scoring system. This study highlights the capacity of MRI to evaluate noncartilaginous changes in the osteoarthritic equine MCP joint. r 2010 Veterinary Radiology & Ultrasound, Vol. 51, No. 3, 2010, pp 267–279. Key words: CT, horse, metacarpophalangeal, MRI, osteoarthritis, radiography. Introduction W HILE MAGNETIC RESONANCE imaging (MRI) is used extensively in humans for assessment of osteoar- thritis, few have assessed it in equine osteoarthritis. 1–5 Ar- ticular imaging scoring systems have been developed to assess the human knee as a whole-organ on MRI, with generation of global scores that can be compared over time and used as an outcome measure. 6,7 However, neither the range of imaging changes in osteoarthritis nor their asso- ciation with cartilage damage have been established in the horse. 8–10 These data are needed for the establishment of an equine joint scoring system. MRI is the modality of choice to assess periarticular soft tissues 11,12 and articular cartilage in human osteoarthritis. 13 On the other hand, the ability of MRI to assess bone changes in the context of osteoarthritis is not well characterized in veterinary medicine. Subchondral bone sclerosis, which re- sults in reduced MR signal intensity, 1 can be quantified, 14 and the subchondral plate can be measured accurately. 5 Moreover, MRI may have an enhanced capacity to detect osteophytes in comparison with radiography. 15–17 However, studies on equine osteoarthritis comparing various imaging modalities for assessment of bone-related parameters, such as osteophytosis and subchondral bone sclerosis, are lacking. The first objective of this study was, therefore, to eval- uate the capability of MRI to assess noncartilaginous features of equine metacarpophalangeal (MCP) joint osteoarthritis by comparing it to conventional computed tomography (CT), computed radiography (CR), and quan- titative CT (QCT) as a reference for subchondral bone density. The second objective was to correlate the imaging findings to the actual cartilage lesions, aiming to develop a total joint osteoarthritis scoring system. Materials and Methods Twenty paired thoracic limbs from mature racehorses, either Thoroughbreds or Standardbreds, were obtained Supported by a grant from the Association des Ve´ te´ rinaires Equins du Que´ bec (AVEQ). Address correspondence and reprint requests to Dr. Marc-Andre´ d’Anjou, at the above address. E-mail: marc-andre.danjou@umontreal.ca Received March 17, 2009; accepted for publication October 27, 2009. doi: 10.1111/j.1740-8261.2009.01653.x From the Department of Veterinary Biomedicine (Olive, The´ oreˆ t) and the Department of Clinical Sciences (d’Anjou, Alexander, Laverty), Faculte´ de me´decine ve´te´ rinaire, Universite´ de Montre´al, 3200 rue Sicotte, PO Box 5000, Saint-Hyacinthe, QC, Canada. 267