Rigid metal plates are a promising alternative to wires for reapproximating the sternum after open-heart surgery due to their potential ability to reduce motion at the wound site and thereby reduce the likelihood of post-operative healing complications. Despite initial clinical success, the use of plates has been limited, in part, by insufficient knowledge about their most effective placement. This study compares the ability of five plate configurations to provide stable closure by limiting sternal separation. Commercially available x-shaped and box-shaped plates were used and combinations of parameters (plate type, location, and number of plates) were investigated in vitro. Lateral distraction tests using controlled, uniform loading were conducted on 15 synthetic sterna and the distractions between separated sternum halves were measured at seven locations. Distractions at the xiphoid, a critical region clinically, varied widely from 0.03 +/- 0.53 mm to 4.24 +/- 1.26 mm depending on all three plate parameters. Of the configurations tested, three x-shaped plates and one box-shaped plate resisted sternal separation most effectively. These results provide the first comparison of plate configurations for stabilizing a sternotomy. However, basic mechanical analyses indicate that sternal loading in vivo is non-uniform; future studies will need to accurately quantify in vivo loading to improve in vitro test methods.