Following trauma, should adult wrist radiographic examinations be two or three projections?

Abstract

Literature review and anecdotal evidence appears to suggest a variation in the number of imaging projections used to demonstrate recent injury to the wrist, dependent on which Accident and Emergency (A&E) radiology department a patient attends. Some departments routinely undertake two projections (dorsi-palmar and lateral) whilst others include a dorsi-palmar oblique projection. The aim of this research was to assess if the inclusion of the additional projection altered the sensitivity of bony injury detection. Four readers reported 250 anonymous image sets, initially of two projections, and then at a later date, reported the same two projections along with the additional third projection, however, they did not have access to their previous reports. The readers identified 4.4% more fractures with the additional projection. Additionally, 2.8% of the fractures had not been identified by any reader on two projections, but were identified by all readers with three projections. Statistical analysis using chi-square test identified a statistical difference in the bone injury detection sensitivity (p < 0.001) when including a third projection. Kappa statistics assessed inter-observer correlation which increased from 0.68 on two projections to 0.78 on three projections. The number of equivocal reports reduced to less than half when the third projection was included. There is an increase in the sensitivity of bone injury identification with the additional projection, which is likely to be of significance to patients and A&E departments, with obviously beneficial social and financial implications.