RATIONALE AND OBJECTIVES: The authors quantitate the radiographic features that distinguish the plain radiographic appearance of Ewing's sarcoma (ES) from other solitary lesions of bone. METHODS: A total of 709 cases of focal bone lesions, including 44 ES, were analyzed according to demographic, anatomic, and plain radiographic features. Vector analysis of groups of features was performed to determine those that are most sensitive and specific for the appearance of ES in contrast with other lesions in the database. RESULTS: In our sample, Ewing's sarcoma is most consistently a medullary-based (91%) lytic (89%) lesion with at least a partially permeative appearance (82%), poorly defined edges (82%), no margination (91%), and a soft-tissue mass (61%). When these lesions occur in long bones, they most commonly are found in the diaphysis (75%) and are proximal more often than distal. Vector analysis suggests that any primary bone lesion without radiographically visible matrix and with either a soft-tissue mass, an appearance of permeative destruction alone or in combination with other patterns of bone destruction is suspect for the diagnosis. This small subset of common features appears to have a high sensitivity (89%) and prevalence (47%) of ES among the lesions meeting these criteria. Limiting the age of the patients to progressively younger age groups increases the specificity but lowers sensitivity. The vector analysis-generated differential diagnoses includes osteosarcoma, giant-cell tumor, lymphoma, and chondrosarcoma. CONCLUSIONS: A relatively specific set of radiographic features can be defined, which will assist in the radiographic diagnosis of ES and improve upon current textbook descriptions.
RATIONALE AND OBJECTIVES: The authors quantitate the radiographic features that distinguish the plain radiographic appearance of Ewing's sarcoma (ES) from other solitary lesions of bone. METHODS: A total of 709 cases of focal bone lesions, including 44 ES, were analyzed according to demographic, anatomic, and plain radiographic features. Vector analysis of groups of features was performed to determine those that are most sensitive and specific for the appearance of ES in contrast with other lesions in the database. RESULTS: In our sample, Ewing's sarcoma is most consistently a medullary-based (91%) lytic (89%) lesion with at least a partially permeative appearance (82%), poorly defined edges (82%), no margination (91%), and a soft-tissue mass (61%). When these lesions occur in long bones, they most commonly are found in the diaphysis (75%) and are proximal more often than distal. Vector analysis suggests that any primary bone lesion without radiographically visible matrix and with either a soft-tissue mass, an appearance of permeative destruction alone or in combination with other patterns of bone destruction is suspect for the diagnosis. This small subset of common features appears to have a high sensitivity (89%) and prevalence (47%) of ES among the lesions meeting these criteria. Limiting the age of the patients to progressively younger age groups increases the specificity but lowers sensitivity. The vector analysis-generated differential diagnoses includes osteosarcoma, giant-cell tumor, lymphoma, and chondrosarcoma. CONCLUSIONS: A relatively specific set of radiographic features can be defined, which will assist in the radiographic diagnosis of ES and improve upon current textbook descriptions.
Authors: Hicham Jalal; Zoubida Belhadj; Hind Enneddam; Mohammed Madhar; Tarik Fikry; Omar Essadki; Ahmed Ousehal Journal: J Med Case Rep Date: 2011-09-12