We report the largest series of periprosthetic fractures in the literature, describing the changing epidemiology and predictors of outcome.

A retrospective search of prospectively compiled trauma and elective electronic databases identified 630 periprosthetic fractures presenting to the study centre between 1995 and 2010. Patient demographics, comorbidities, socioeconomic status, mechanism of injury, fracture type, classification, method of fixation, and outcome were recorded using the patients’ notes. The General Register Office for Scotland was used to obtain the mortality status of the patients.

There were 276 total hip replacements (THR), 123 total knee replacements (TKR), 117 hemiarthroplasty, and 114 “other” implants. The incidence of periprosthetic fractures increased significantly during the study period for all implants: THR (p<0.001), TKR (p<0.001), hemiarthroplasty (p=0.002), and other (p=0.003). The majority of fractures were fixed by open reduction and internal fixation (72%). This failed in 14% of THR, 15% of TKR, 21% of hemiarthroplasties, and 18% of “other” implants. Isolated independent predictors of failure of fixation, after multivariate regression analysis, were increasing age, deprivation, a past medical history of asthma or chronic obstructive airways disease, osteoporosis, and steroid use (p<0.05). Failure of fixation was associated with a significantly increased one year mortality rate (OR 12.5, p=0.003).

Periprosthetic fractures involving THR and TKR are becoming more prevalent. Patient demographics can be used to calculate the risk of failure of fixation, and those with an increased risk may benefit from revision of their implant, and avert the associate morbidity of failure of fixation.