Am Fam Physician. 2000;62(11):2500
Ankle injuries are a common reason for visits to the emergency department. These injuries typically require radiographic evaluation to assess for fractures, which results in a significant expenditure of resources. The standard radiographic evaluation of the ankle consists of anteroposterior (AP), mortise and lateral views. Recent studies have suggested that two views (i.e., AP and lateral or lateral and mortise) are as effective in identifying fractures as the standard three views. However, no study has compared the accuracy of all three combinations in identifying fractures. Brandser and colleagues conducted such a comparison to identify which view or combination of views, if any, was superior in identifying ankle fractures.
A total of 556 consecutive three-view ankle radiographs obtained on 555 patients over a 12-month period was reviewed retrospectively. The radiographs of patients who were found to have at least one ankle fracture were compared with those of patients who had normal studies. Each image was sorted by view and evaluated for fractures independently by two radiologists and two orthopedic surgeons using a four-point confidence rating scale. The radiographs were judged to be possibly abnormal, inconclusive, fairly certain abnormal or definitely abnormal. The accuracy of each view was assessed independently, as well as the accuracy of the two-view and three-view combinations.
Both two-view combinations detected fractures at similar rates, but the three-view combination detected significantly more fractures. Individually, the mortise and AP views appeared to provide more information about fractures of the medial malleolus, while the lateral view was superior in identifying fractures of the posterior malleolus. The study also demonstrated a statistically significant reduction in accuracy when the AP or mortise view was eliminated from the three-view combination.
The authors conclude that reducing the standard radiographic evaluation from three to two views significantly reduces the accuracy of detecting ankle fractures. This difference was apparent for each of the two-view combinations. Therefore, the three-view combination should remain the gold standard for detecting ankle fractures.