BACKGROUND: Displaced paediatric supracondylar humeral fractures pose a significant risk of neurovascular injury and consequently have traditionally been treated as a surgical emergency. Recently, the need for emergency surgery has been questioned. We wished to analyse our experience at a large UK tertiary paediatric trauma centre. METHODS: A retrospective case note review was performed on patients with Gartland Grades 2 and 3 supracondylar fractures observed in a 2-year period from July 2008 to July 2010. We divided children into those treated before 12 h (early surgery) and after 12 h (delayed surgery). Analysis was undertaken using Fisher's exact test. RESULTS: Of the 137 patients, 115 were included in the study; median time-to-surgery was 15:30 (range 2:45-62:50); thirty-nine children were treated before 12 h and 76 patients after. In the early surgery group, three children (7.7%) developed a superficial pin-site infection, four children (10.3%) required open reduction, five children (12.8%) sustained an iatrogenic nerve injury, and two children (5%) required reoperation. In the delayed surgery group, one child (1.3%) had a superficial pin-site infection, four children (5.3%) required open reduction, seven children (9.2%) sustained an iatrogenic nerve injury, and two children (2.6%) reoperation. Bivariate analysis of our data using Fisher's exact t test revealed no statistically significant difference between early and delayed surgery groups with regard to infection rates (p = 0.1), iatrogenic nerve injury (p = 0.53) or need for open reduction (p = 0.44). CONCLUSION: Our results indicate that delayed surgery appears to offer a safe management approach in the treatment of displaced supracondylar fractures, but it is important that cases are carefully evaluated on an individual basis. These results indicate that patient transfer to a specialist paediatric centre, often with consequent surgical delay, is a safe management option and also negates the obligation to carry out these procedures at night.