Treatment of pilon fractures remains challenging due to the difficulty of fracture reduction and associated soft tissue complications. The aim of this study was to evaluate the pitfalls and strategies of posterior column reduction in the treatment of complex tibial pilon fractures (AO/OTA 43-C3). Thirteen AO/OTA classification 43-C3 type pilon fractures treated between January 2013 and January 2016 were retrospectively analyzed. Nine cases were treated by external fixation within 26 hours (range, 6–56 hours) after injury. The definitive open reduction and internal fixation (ORIF) was performed after the wound was healed without infection and soft tissue swelling had subsided. During the delayed/second-stage operation, the articular surface of the distal tibial plafond was reduced through the posterolateral and anterior approaches. X-ray and CT scans were performed pre- or postoperatively. The reduction quality was evaluated using Burwell–Charnley’s radiographic criteria. The follow-up was performed routinely and all complications were recorded. Ankle function was evaluated using the American Orthopedic Foot and Ankle Society (AOFAS) ankle-hindfoot score. During the delayed/second-stage operation, primary reduction of the posterior column was performed entirely through posterolateral approaches. However, poor posterior column reduction was revealed by fluoroscopy in four cases, three of which were readjusted through the posterolateral and anterior approaches, and the fourth was adjusted directly through the anterior approach. Postoperative CT scan revealed that the step-off of the articular surface was less than 2 mm in 12 cases, and in only one case the step-off was greater than 2 mm but less than 5 mm. The satisfactory rate was 92.3% according to Burwell–Charnley’s reduction criteria. Eleven patients were followed up regularly; superficial infections occurred in two cases but healed after wound care treatment in 3 and 5 weeks, respectively. All eleven fractures were healed within an average of 3.6 months (range, 2.6–5 months). The average range of ankle motion was 19° of dorsiflexion and 28° of plantar-flexion. The mean AOFAS ankle-hindfoot score was 82 (range, 61–92). In our opinion, we suggest that the reduction of the articular surface should be performed through combined posterolateral and anterior approaches in a delayed operation, with flexible fixation of the posterior column. If the posterior column is poorly reduced, the articular surface can easily be manipulated through anterior approaches. According to this strategy, satisfactory outcomes of AO/OTA C3 pilon fractures would be anticipated.