RPC of the month from the AFIP.

Abstract

FIGURES 1, A and B represent a nosological entity in which the accuracy of roentgenographic diagnosis should approach 100 per cent. The exercise this month emphasizes that the radiograph supplants the gross anatomy in the differential diagnosis of bone lesions. Make the most appropria te choice from the following and then read further for th e differential discussion and correlation : Chondrosarcoma, Fibrosarcoma, Cyst ic Osteomye litis, Giant Cell Tumor , Chondromyxoid Fibroma, Nonosteoge nic Fibroma, or Cortical (subperio steal) Desmoid. The frontal and lateral radiographs of the distal end of the femur (Fig. 1) show a multiloculated, elongated, eccentric, metaphyseal-diaphyseal radiolucency, with a slightly scalloped, well demarcated sclerotic rim. There is no evidence of mineralized matrix or periosteal reaction. Malignant disease would not usually be considered in the absence of rapid bone destruction, break through the cortex, tumor shadow in the soft tissue, or periosteal reaction. In addition, absence of radiologically evident mineralization discourages consideration of such primary neoplasms as chondrosarcoma and osteosarcoma, although a slow-growing fibrosarcoma may mimic the benign features observed here. The absence of malignant characteristics here, however, places fibrosarcoma low on the list of diagnostic possibilities. In the absence of periosteal reaction, an active inflammatory process is quite unlikely. It would also be most unusual for a cystic osteomyelitis not to show evidence of periosteal thickening. Moreover, the cortical appearance of the rind of the lesion is sharp and narrow and differs from the ill-defined and irregular bone sclerosis usually seen with inflammatory lesions. Overall, the characteristics lead to a consideration of benign, slow-growing bone lesions. Among benign bone tumors, we offered giant-cell tumor and chondromyxoid fibroma as possible choices. Giant-cell tumors are well demarcated, radiolucent, and quite frequently multiloculated, just as we have described our present lesion, but they almost never demonstrate the dense margin we note here. More important, giant-cell tumors are oval to round and lie athwart the bone while this lesion is elongate and parallels the axis of the bone. Even more critical is the location. A giant-cell tumor should lie in the metaphyseal- epiphyseal region. In fact, whenever a diaphyseal lesion presents as a giant-cell tumor microscopically, we should consider the possibility of hyperparathyroidism to explain such a location. Furthermore, the giant-cell tumor not only belongs in the metaphyseal region, but its growth carries it into the adjacent epiphysis (if the plate is closed) often down to the very margin of the subchondral bone and frequently with associated bulging of the cortex. Our patient's lesion shows neither of these characteristics.