Late screw-related complications in locking plating of proximal humerus fractures: A systematic review

•One third of the treated fractures with locking plates had complications, with 11% of the cases having screw-related complications.•Screw perforation was found in 612 cases, accounting for the 80% of the screw-related complications.•Onset was provided for only 49% of the perforated, 12.5% of the cut-out, 35.9% of the loose and 13.3% of the retracted screws.•For the late screw perforation, the percentage of known information ranged from 1% for comorbidities to 16% for the status of the bone. Locking plating is a common surgical treatment of proximal humeral fractures with satisfactory clinical results. Implant-related complications, especially screw-related, have been reported, however, the lack of information regarding their onset, used surgical technique, complexity of the fracture, bone quality etc., prevents from understanding the causes for them. The aim of this systematic review is to identify the potential risk factors for late screw complications by gathering information about the patient characteristics, comorbidities, fracture types, surgical approaches and implant types. A PubMed search was performed using humerus, fractures, bone and locking as keywords in clinical papers written in English. All abstracts and manuscripts on distal or humerus shaft fractures, and those on proximal humerus fractures without any or with only iatrogenic complications were excluded. One hundred studies met the inclusion criteria, resulting in 33% of the reported cases having at least one complication, with 11% of all complications being screw-related. Most of the latter were secondary screw perforations and screw cut-outs, being predominantly linked to poor bone quality, while screw loosening and retraction were found less frequently as a result of locking mechanism failure. Overall, the amount of information for complications was limited and screw perforation was the most frequent screw-related complication, mostly reported in female patients older than 50 years, following four-part or AO/OTA type C fractures and detected four weeks postoperatively. The sparse information in the literature could be an indicator that the late screw complications might have been under-reported and under-described, making the understanding of the screw-related complications even more challenging.