RESUMO
OBJECTIVE: To evaluate the proportion of extraskeletal, periosteal, and intramedullary Ewing sarcomas among musculoskeletal Ewing sarcomas. MATERIAL AND METHOD: Our single-center retrospective study included patients with musculoskeletal Ewing sarcoma diagnosed between 2005 and 2019 in our pathology center (cases from our adult bone tumor referral center and adult and pediatric cases referred for review). Recurrences, metastases, and visceral Ewing sarcomas were excluded. Intramedullary Ewing sarcomas were defined by involvement of the medullary cavity. Periosteal cases were defined by involvement of the subperiosteal area without extension to the medullary cavity. Extraskeletal cases were defined by the absence of involvement of the bone tissue and the subperiosteal area. RESULTS: Our series included 126 patients with musculoskeletal Ewing sarcoma, including 118 skeletal Ewing sarcomas (93.7%) and 8 extraskeletal Ewing sarcomas (6.3%). Of the 118 skeletal Ewing sarcomas 112 were intramedullary (88.9%) and 6 were periosteal (4.8%). Extraskeletal Ewing sarcomas were more common in women and in patients older than 40 (p < 0.05). DISCUSSION: The 6.3% proportion of extraskeletal Ewing sarcoma is lower than the median of 30% estimated from the literature. This difference could be explained by an overestimation of extraskeletal Ewing sarcomas of the chest wall (Askin tumors), an underestimation of periosteal cases confused with extraskeletal cases, and the presence of "Ewing-like" soft tissue sarcomas in previous series. Because of its prognostic and therapeutic impact, the distinction of morphologic subtypes requires the cooperation of experienced radiologists and pathologists.