Elbow arthrography: a reassessment of the technique.

Twenty-six elbow arthrograms were retrospectively reviewed. Each arthrogram consisted of conventional radiography, conventional tomography, standard double-contrast arthrography, and arthrotomography. Each of these four components was independently interpreted in a blinded fashion by six radiologists, each working alone. The four components were evaluated for presence or absence of intraarticular bodies, cartilage loss, or fractures. The results showed that conventional tomography was the most accurate single study. Arthrotomography rarely improved the diagnostic accuracy of noncontrast methods. It was useful in those few patients where detection of purely cartilaginous bodies, precise position of mineralized densities, or status of articular cartilage was desired. It may be possible to reduce the number of elbow arthrograms, thereby reducing time, cost, and radiation dose factors.