Aims To study whether a delayed reduction of supracondylar fractures is safe or not.

Methods Retrospective review of 169 supracondylar fractures type 2 and 3 (Gartland classification) seen during the period of January 2010-June 2014. Randomisation was largely determined by time of arrival to our hospital, as we are a referral centre for Paediatric Orthopaedics. The surgery was carried out by a trained paediatric Orthopaedic surgeon or under his direct supervision. The variables studied were: time interval from the accident to surgery; pre- and postoperative neurological and vascular condition; duration and type of surgery; quality of reduction (only for Gartland 3) and; postoperative complications. For study purposes, patients were divided into 2 groups: those operated <6 hours from the time of fracture (early group) and those operated >6 hours (late group). Gartland 2 and 3 fractures were analysed independently. Pulseless limbs on arrival were excluded as they were treated emergently

Results A total of 98 fractures were operated in the ‘early’ group and, 71 in the ‘late’ group. 2 patients in the late group required open reduction. Both had neurological injury on presentation and during reduction became pulseless. 8 cases needed second reduction under GA due to unacceptable position, either because of unsatisfactory reduction or secondary displacement. 5 belonged to the early group and 3 to the late. The most important factor related to surgery duration was the presence of a neurological deficit. See table1. The incidence of neurological injury related to the time was analysed. See table2.

Conclusion The delay in treatment did not determine the appearance of a neurological deficit. Delayed reduction of supracondylar fractures type 3 can be safely done provided there are no associated neurological injuries at presentation.