Importance of locking plate positioning in proximal humeral fractures as predicted by computer simulations.

Multifragmented proximal humeral fractures frequently require operative fixation. The locking plates commonly used are often placed relative to the greater tuberosity, however no quantitative data exists regarding the effect of positional changes. The aim of the study was to establish the effects from variations in proximal-distal PHILOS humeral plate positioning on predicted fixation failure risk. Twenty-one left-sided low-density virtual humeri models were created with a simulation framework from CT data of elderly donors and osteotomized to mimic an unstable three-part malreduced AO/OTA 11-B3.2 fracture with medial comminution. A PHILOS plate with either four or six proximal screws was used for fixation. Both configurations were modelled with plate repositioning 2 and 4 mm distally and proximally to its baseline position. Applying a validated computational model, three physiological loading situations were simulated and fixation failure predicted using average strain around the proximal screws-an outcome established as a surrogate for cycles to failure. Varying the craniocaudal plate position affected the peri-implant strain for both four and six-screw configurations. Even though significant changes were seen only in the latter, all tests suggested that more proximal plate positioning results in decreased peri-screw strains whereas distalizing creates increases in strain. These results suggest that even a small distal PHILOS plate malpositioning may reduce fixation stability. Plate distalization increases the probability of being unable to insert all screws within the humeral head, which dramatically increases the forces acting on the remaining screws. Proximal plate shifting may be beneficial, especially for constructs employing calcar screws.