Acute paediatric ankle trauma: MRI versus plain radiography.

Lohman M, Kivisaari A, Kallio P, Puntila J, Vehmas T, Kivisaari L. Skeletal Radiol 2001 Sep;30(9):504-11

OBJECTIVE: To evaluate the diagnosis of acute physeal ankle fractures on plain radiographs using MRI as the gold standard. METHODS: Sixty consecutive children, 29 with a clinical diagnosis of lateral ligament injury and 31 with physeal ankle fractures, were examined using both radiographs and MRI in the acute period. The imaging data were reviewed by three “masked” radiologists. The fracture diagnosis and Slater-Harris classification of radiographs were compared with findings on MRI. RESULTS: Plain radiography produced five of 28 (18%) false negative and 12 of 92 (13%) false positive fracture diagnoses compared with MRI. Six of the 12 false positive fractures were due to a misclassification of lateral ligament disruption as SH1 fractures, Altogether a difference was found in 21% of cases in either the diagnosis or the classification of the fractures according to Salter-Harris. All bone bruises in the distal tibia and fibula and 64% of bone bruises in the talus were seen in association with lateral ligament injuries. Talar bone bruises in association with fractures occurred on the same side as the malleolar fracture; talar bone bruises in association with lateral ligament disruption were seen in different locations. The errors identified on radiographs by MRI did not affect the management of the injury. CONCLUSIONS: The incidence of false negative ankle fractures in plain radiographs was small and no complex ankle fractures were missed on radiographs. The total extent of complex fractures was, however, not always obvious on radiographs. In an unselected series of relatively mild ankle injuries, we were unable to show a single case where the treatment or prognosis based on plain radiography should have been significantly altered after having done a routine MRI examination. Plain radiography is still the diagnostic cornerstone of paediatric ankle injuries.