The management of unstable ankle fractures is challenging due the difficulty in differentiating between stable and unstable fracture patterns. The aim of our study was to examine our practice and to determine if the operative management of unstable ankle fractures resulted in significantly improved radiographic parameters. Between June 2008 and December 2008, we identified all skeletally mature patients who were diagnosed with an ankle fracture after having radiographs in the radiology department at our institution. We analysed the case notes and radiographs of these patients retrospectively. The fractures were classified according to the Weber and Lauge-Hansen classification. Radiographs were evaluated for shortening of the fibula, widening of the joint space, or malrotation of the fibula. Three measurements were used to ascertain whether the correct fibular length has been restored – the circle sign, the talocrural angle, and the tibiofibular (or Shenton) line. Of 1064 patients who had radiographs, 123 patients sustained a fracture of the ankle. There were 61 females and 62 males, with a median age of 47 years. There were 20 Weber A, 80 Weber B and 12 Weber C ankle fractures. Eleven fractures could not be classified according to the Weber classification. According to the Lauge-Hansen classification there were 44 Supination-External rotation (SER) stage II fractures, 35 Supination-External rotation (SER) stage IV fractures, and 7 Pronation-External rotation (PER) stage III fractures. In the unstable SER stage IV fractures, 30 of the 35 patients had operative treatment and there was no statistically significant difference in the average Talocrural angles in the operative (78.9°) and nonoperative groups (83.4°). None of the patients with an SER stage IV fracture managed nonoperatively had an adequate circle sign compared to 14 of the 30 patients in the operative group who had an adequate circle sign. In the PER stage III fractures 4 of the 7 patients had operative treatment. The average Talocrural angle in the operative group was 79.1° versus 75.3° in the nonoperative group, with all patients in the operative group having an adequate circle sign compared to none in the nonoperative group. The patient numbers in the PER stage III group however were too small to show a statistically significant difference. In 4 patients with unstable fracture patterns, the use of a third tubular plate to bridge a fibula fracture without an inter-fragmentary lag screw led to inadequate restoration of fibular length in all cases. It is important to recognise unstable ankle fracture patterns and, in patients treated operatively, to restore fibular length and rotation. Operative management of unstable SER stage IV and PER stage III ankle fractures can restore normal radiographic parameters. We highlight certain technical pitfalls in restoring fibular length such as the inappropriate use of the semi-tubular plate as a bridging plate