Background: The aim was to clarify the occurrence of delayed union after surgical treatment of mandibular fracture and investigate whether an association exists between perioperative use of dexamethasone and delayed union. Material and Methods: Thirty-seven patients were included in a prospective randomized study. Of these patients, 19 (51.4%) were randomized to receive a total dose of 30 mg of dexamethasone and 18 (48.6%) served as controls. Patients underwent clinical and radiological investigation immediately, one month, three months and six months postoperatively. Radiographs were evaluated by an experienced, blinded senior oral radiologist. Results: Delayed fracture union was found in 9 patients (24.3%). It was associated significantly with angle fractures (p=0.012). Delayed union occurred more frequently in patients who received dexamethasone (36.8%) than in those who did not (11.1%) (p=0.068). The association of infection with delayed union was significan t ( p=0.027). Moreover, dexamethasone was significantly ( p=0.019) associated with delayed fracture union with concomitant infection. Gender, age group, smoking habit, treatment delay and duration of surgery were not associated with delayed union. Conclusions: Infection was associated with delayed union. Short-term high-dose dexamethasone predisposed to complicated fracture union, especially in patients with angle fractures. The relationship between dexamethasone and delayed bone healing without infection remains unresolved