Infiltration of sarcomas into the hip joint: comparison of CT, MRI and histologic findings in 67 cases.

We analyzed the incidence, route, and characteristics of hip joint infiltration in pelvic or proximal femoral sarcomas. 67 patients with a sarcoma that originated around the hip joint (50 pelvic and 17 femoral) were included in this study. Preoperative CT and MRI were matched with the histological findings in tumor specimens. Tumor infiltration into the hip joint was suspected on the basis of preoperative imaging in 29 patients due to articular cartilage disruption, diffuse signal changes in the acetabulum or femoral neck, signs of a tumor in the joint, or markedjoint effusion. Of these 29 patients, 15 showed tumor invasion on histological examination. 12 of 31 chondrosarcomas, none of 12 Ewing's sarcomas, and 3 of 24 osteosarcomas infiltrated into the hip joint (p = 0.008). 10 of 26 low-grade sarcomas and 5 of 41 high-grade sarcomas infiltrated into the hip joint (p = 0.02). The joint infiltration rate of the chondrosarcomas was related to their size. Of 10 tumors originating in the acetabulum, 9 penetrated through or around the osseous-ligamentous junction and one through the acetabular cartilage. In 5 proximal femur lesions, all infiltrated the joint through the femoral neck, 3 of them also through the ligamentum teres.

Incidence and mechanisms of infiltration of sarcomas in the shoulder.

The intraarticular infiltration of sarcomas around the shoulder was analyzed. Of 58 sarcomas located around the shoulder, which were treated between 1993 and 1998, 15 osteosarcomas, 12 chondrosarcomas, and one Ewing's sarcoma that abutted the shoulder were selected. Radiologic images of 28 tumors were analyzed and compared with pathologic examinations. In 17 of 28 sarcomas (15 in the proximal humerus and two in the scapula), joint infiltration was suspected radiologically because of the existence of a tumor mass probably inside the joint or disruption of the joint surface and protrusion of the tumor. Seven of 17 tumors with radiologically positive and zero of 11 tumors with radiologically negative findings showed histologically positive findings for tumor inside the joint. Four of 11 osteosarcomas and three of four chondrosarcomas of the proximal humerus showed histologic findings of joint infiltration. Pericapsular extension was seen in two patients, direct articular spread with a pathologic fracture was seen in two patients, direct articular spread was seen in one patient, direct articular spread with pericapsular extension was seen in one patient, and pericapsular extension with a pathologic fracture was seen in one patient. The current study showed that pericapsular or direct articular infiltration into the glenohumeral joint is a phenomenon in sarcoma around the shoulder. Exact assessment of tumor existence inside the joint can be difficult in osteosarcoma and chondrosarcoma, so abnormal radiologic findings should be regarded as proof of existence of tumor, and extraarticular wide resection should be planned.