Objectives: To biomechanically assess five proximal, humeral, fracture-plate-fixation systems.
Methods: Surgical neck fractures, with and without cortical contact, were created in 25 fresh-frozen cadaveric humeri. Five methods of plate fixation were used for repair: construct A [an eight-hole, low contact dynamic compression (LCDC) plate contoured into a blade shape, supported by one, 70-mm-long, 4.5-mm-diameter cortical screw acting as a truss], construct B (a 10-hole LCDC plate arrangement identical to construct A, but using one, 70-mm-long, 3.5-mm-diameter cortical screw as a truss), construct C [a five-hole dynamic compression (DC) blade plate with one, 6.5-mm-diameter cancellous screw], construct D (a five-hole T-plate supported by three, 6.5-mm-diameter cancellous screws), and construct E (a five-hole cloverleaf plate supported by five, 4-mm-diameter cancellous screws). Plates were posterior to the bicipital groove, 10 mm distal to the greater tuberosity tip, on the lateral aspect of the humeral shaft. Screw fixation was done using standard AO compression plating techniques. Stiffness of constructs was measured in bending and axial compression. Locked plates were not assessed.
Results: For cortical contact [abduction of 20 degrees (P=0.02), flexion of 20 degrees (P=0.02), flexion of 90 degrees (P=0.005)] and no cortical contact [flexion of 90 degrees (P=0.0001)], construct A was significantly stiffer than other constructs. For no cortical contact in abduction of 90 degrees (P=0.05), construct A was significantly stiffer than other constructs.
Conclusions: Construct A was significantly stiffer than other constructs.