Biomechanical analysis of screws versus k-wires for lateral humeral condyle fractures

Academic Article


  • Background: Good outcomes have been described for pediatric lateral condyle fractures treated by open reduction and fixation using either screws or Kirschner wires (K-wires). No studies have compared the biomechanical properties of the 2 fixation methods. We hypothesized that screw fixation would be more biomechanically stable than K-wire fixation. Methods: Synthetic humerus models were used for biomechanical testing, following a previously published protocol. A miter saw was used to make an oblique cut to simulate a Milch type II fracture. Fractures were anatomically reduced and fixed with either 2 divergent 0.062-inch K-wires placed bicortically or a 4.0-mm lag screw placed obliquely (perpendicular to the fracture line). Specimens were then embedded in polymethyl methacrylate bone cement for testing. Mechanical testing using displacement control was performed applying compression or distraction from 0 to 1.5mm at a rate of 0.5 mm/s for 10 cycles. The maximum force was calculated based on the maximum force averaged over the 10 cycles. Stiffness was calculated based on the slope of the force-displacement curve of the 10th cycle. A 2-sample t test was used to determine significant differences between the stiffness and maximum force comparing the K-wire and screw groups. A P-value of <0.05 was considered statistically significant. Results: Stiffness and maximum force in tension testing were significantly greater with screw fixation compared with K-wire fixation. Testing in compression revealed statistically significant increased maximum force and a trend towards increased stiffness. Conclusion: Screw fixation in a synthetic bone model of pediatric lateral condyle fractures (Milch type II) provides increased biomechanical stability of the construct as compared with K-wires. Clinical Relevance: If similar effects were seen in vivo, increased biomechanical stability with screw fixation could decrease the occurrence of complications such as loss of reduction and nonunion.
  • Authors

    Published In

    Digital Object Identifier (doi)

    Author List

  • Schlitz RS; Schwertz JM; Eberhardt AW; Gilbert SR
  • Start Page

  • e93
  • End Page

  • e97
  • Volume

  • 35
  • Issue

  • 8