PReS-FINAL-2191: Imaging of chronic recurrent multifocal osteitis: a french national cohort of 178 cases

Results The average number of lesions per patient detected before diagnosis was 3.5 ± 2.9 [1-26]. The mean number of lesions was 1 ± 0.9 by radiographs, 2.45 ± 1.7 by isotopic bone scan and 3.12 ± 1.33 by MRI. A total of 193 radiographic lesions were detected with the following distribution: tibia (n = 44), the clavicle (n = 34), the femur (n = 23), the fibula (n = 20) and pelvis (n = 19). The lesions of the lower limbs accounted for 52% of the lesions. The lesions of the long bones were most often located in metaphyseal (58/76, 76%) and were lytic in 76/162 (47%) and sclerotic in 60/ 162 (37%). The isotopic bone scan detected 372 lesions localized to the pelvis (n = 64), tibia (n = 51), femur (n = 44), clavicle (n = 40) and vertebrae (n = 29). The lesions were mainly metaphyseal (65/92, 56%). MRI detected 515 lesions distributed as follows: pelvis (n = 100), tibia (n = 93) and femur (n = 73) are the sites most frequently affected. 51 vertebral lesions (10%) were detected in 36 patients. Most of them were localized at the thoracic level. The vast majority of lesions were hypo-T1 and hyper-T2. The description of bone lesions with MRI seems to be more accurate (metaphysoepiphyseal) and more frequently bilateral and symmetric. Imaging has allowed confirming the multifocal pattern in 26/54 patients with clinical monofocal at diagnosis. Of the remaining 28 patients with monofocal lesion, the clinical course and imaging confirmed the multifocal pattern in 16 additional cases: a total of only 12 patients (7%) kept a pure monofocal evolution. In 15 patients, scintigraphy and whole-body MRI were performed at the same time (+/3 months). Analysis of these 15 patients showed a higher sensitivity to detect lesions by MRI (6.7 ± 3.1 vs 3.4 ± 2.4, p = 0.003) and better description of lesions. Jansson score that integrates imaging, was interpretable in 110 patients: the application of this score would have, in this cohort, to avoid 27/110 biopsies.