Objective To analyze the stability of humerus supracondylar fracture fixation with Kirschner wires comparing intramedullary and lateral (Fi), and two parallel lateral wires (FL) fixation in experimental models, to define which configuration presents greater stability. Methods A total of 72 synthetic humeri were cross-sectioned to simulate the fracture. These bones were divided into two equal groups and the fractures were fixed with parallel Kirschner wires (FL) and with a lateral and intramedullary (Fi) wire. Then, the test specimens were subjected to stress load tests on a universal test machine, measured in Newtons (N). Each group was subdivided into varus load, valgus, extension, flexion, external rotation and internal rotation. An analysis of the data was performed comparing the subgroups of the FL group with their respective subgroups of the Fi group through the two-tailed t test. Results The two-tailed t test showed that in 4 of the 6 evaluated conditions there was no significant statistical difference between the groups ( p > 0.05). We have found a significant difference between the group with extension load with a mean of 19 N (FL group) and of 28.7 N (Fi group) ( p = 0.004), and also between the groups with flexural load with the mean of the forces recorded in the FL group of 17.1 N and of 22.9 N in the Fi group ( p = 0.01). Conclusion Fixation with one intramedullary wire and one lateral wire, considering loads in extension and flexion, presents greater stability when compared to a fixation with two lateral wires, suggesting similar clinical results.

Keywords: biomechanical phenomena; epiphyses/injuries; fracture fixation; humeral fractures.