Far cortical locking can improve healing of fractures stabilized with locking plates

Abstract

Background: Locked bridge plating relies on secondary bone healing, which requires interfragmentary motion for callus formation. This study evaluated healing of fractures stabilized with a locked plating construct and a far cortical locking construct, which is a modified locked plating approach that promotes interfragmentary motion. The study tested whether far cortical locking constructs can improve fracture-healing compared with standard locked plating constructs.

Methods: In an established ovine tibial osteotomy model with a 3-mm gap size, twelve osteotomies were randomly stabilized with locked plating or far cortical locking constructs applied medially. The far cortical locking constructs were designed to provide 84% lower stiffness than the locked plating constructs and permitted nearly parallel gap motion. Fracture-healing was monitored on weekly radiographs. After the animals were killed at week 9, healed tibiae were analyzed by computed tomography, mechanical testing in torsion, and histological examination.

Results: Callus on weekly radiographs was greater in the far cortical locking constructs than in the locked plating constructs. At week 9, the far cortical locking group had a 36% greater callus volume (p = 0.03) and a 44% higher bone mineral content (p = 0.013) than the locked plating group. Callus in the locked plating specimens was asymmetric, having 49% less bone mineral content in the medial callus than in the lateral callus (p = 0.003). In far cortical locking specimens, medial and lateral callus had similar bone mineral content (p = 0.91). The far cortical locking specimens healed to be 54% stronger in torsion (p = 0.023) and sustained 156% greater energy to failure in torsion (p < 0.001) than locked plating specimens. Histologically, three of six locked plating specimens had deficient bridging across the medial cortex, while all remaining cortices had bridged.

Conclusions: Inconsistent and asymmetric callus formation with locked plating constructs is likely due to their high stiffness and asymmetric gap closure. By providing flexible fixation and nearly parallel interfragmentary motion, far cortical locking constructs form more callus and heal to be stronger in torsion than locked plating constructs.

Fig. 1

Fixation constructs. A: Locking plate (LP) with anatomic curvature of the ovine tibia. B: Far cortical locking (FCL) screw for unicortical fixation in the far cortex, enabling elastic flexion of the screw shaft within the motion envelope (Δd) in the near cortex. C: Nearly parallel interfragmentary motion provided by far cortical locking screws under axial loading. D: Mechanically, far cortical locking constructs derive elastic fixation by cantilever bending of far cortical locking screw shafts similar to an external fixator that derives elasticity from fixation pin flexion.

Fig. 2

Progression of periosteal callus formation. A: The far cortical locking (FCL) group formed significantly more callus (mean and standard deviation) than the locked plating (LP) group (p = 0.004). B: Callus projections at the anterior, posterior, and lateral aspects representative for far cortical locking and locked plating specimens at week 9 after surgery.

Fig. 3

Periosteal callus rendering from quantitative computed tomography data at week 9. A: Bone mineral content (BMC, mean and standard deviation in mg hydroxyapatite) in the far cortical locking (FCL) group was 43% greater than in the locked plating (LP) group. B: Far cortical locking constructs yielded symmetric callus formation, evident by equal bone mineral content in the medial half of the callus (near callus) and lateral half of the callus (far callus). Locked plating constructs had an asymmetric bone mineral content distribution, with 49% less bone mineral density in the near callus. C: Transverse slices adjacent to the osteotomy gap as well as callus volume rendered from computed tomography scans illustrate that callus formation was asymmetric in locked plating constructs, but extended from the far to the near cortex aspects in far cortical locking constructs, as reflected by the medial imprint of the plate. D: Parasagittal quantitative computed tomography slice extracted through the near cortex demonstrates that the far cortical locking screw-holes maintained clearly defined and circular boundaries.

Fig. 4

Torsion test results after tibial harvest and implant removal at postoperative week 9 (mean and standard deviation). Compared with the locked plating (LP) group, tibiae in the far cortical locking (FCL) group were not significantly stiffer (A), but they healed to be 54% stronger (B), and they sustained 156% more energy before fracturing (C).

Fig. 5

Typical histological appearance of healing in the far cortical locking (FCL) and locked plating (LP) groups, shown with toluidine blue stain and after image processing for callus differentiation (green = callus, blue = cortex, and pink = fibrous tissue). Three of six locked plating specimens had deficient bridging at the near cortex, while all other near and far cortices had bridged.