Comparison of the Mechanical Behaviors of Locked and Nonlocked Plate/Screw Fixation Applied to Experimentally Induced Rotational Osteotomies in Canine Ilia

Objective To compare the mechanical behaviors of 2 locked (parallel and diverging screws) and 1 nonlocked (NL) version of triple pelvic osteotomy (TPO) plate/screw fixation. Study Design In vitro biomechanical evaluation. Animals Cadaveric canine hemipelves. Methods Comparison 1—NL screws 20° TPO (NL-20) construct versus locked parallel (LP) screws 20° TPO (LP-20) construct (n = 7). Comparison 2—LP-20 construct versus locked diverging (LD) screws 20° TPO (LD-20) construct (n = 6). Condition 1—Nondestructive loading to determine stiffness. Condition 2—Cyclic loading to determine stiffness, screw loosening, and osteotomy gap displacement. Condition 3—Load to failure (yield load, yield displacement, maximum load, load to failure, failure mode). Results Stiffness was not significantly different for NL-20 versus LP-20 constructs (P = .48) or for LP-20 versus LD-20 constructs (P = .83). Screw loosening was significantly more frequent for NL-20 versus LP-20 (P = .01) and for LD-20 versus LP-20 constructs (P = .02). The relative risk for screw loosening with NL-20 constructs versus LP-20 constructs was 1.4 (95% CI: 1.1–1.8). The relative risk for screw loosening with LD-20 versus LP-20 was 1.6 (95% CI: 1.1—2.2). Yield load was significantly greater for LP-20 versus NL-20 and LD-20 constructs (P = .04, P = .03), respectively. Conclusions No TPO constructs tested maintained complete integrity after cyclic loading; however, screw loosening was significantly reduced and yield loads were significantly larger for LP-20 plate/screw constructs.