Relative survival following hemi-and total hip arthroplasty for hip fractures in Sweden

Hip fractures are a common problem in the ageing population. Hip arthroplasty is the common treatment option for displaced intracapsular neck of femur fractures. Even though hip replacements are successful in restoring mobility, reducing pain and diminishing loss of health-related quality of life, the potential impact of a hip fracture on life expectancy as well as the postoperative mortality need consideration. The purpose of this study was to describe the mid-term relative survival rate for a cohort of Swedish patients whom underwent total- or hemiarthroplasty surgery following hip fracture. We also explored whether the survival rate is prosthesis-type specific and influenced by comorbidities, sex, socioeconomic and surgical factors.

Using prospectively collected information of the Swedish Hip Arthroplasty Register-linked database we identified 43,891 patients operated between 2005 and 2012. Patient- and surgery-specific data in combination with socio-economic data were available for this analysis. We studied relative survival rate and used multivariable modelling with Cox Proportional Hazards Model in Transformed Time.

Compared to the Swedish general population the baseline excess hazard was very high in the first half year after the operation, thereafter the excess hazard decreased but remained non-negligible through the 8 years' follow-up period. The mortality rate of males was higher compared to women. Higher Elixhauser comorbidity index (ECI) was associated with worsening survival. However, patients who had ECI = 0 had higher mortality than patients with ECI =1 the first 420 days post fracture. Patients with a hemiarthroplasty had a worse survival than patients with a total hip arthroplasty. Of the hospital types considered university hospitals had lower survival rate. Younger patients had a greater loss of expected life span than patients who suffer hip fracture in their more advanced ages.

Swedish hip fracture patients who undergo arthroplasty surgery had a high excess hazard of dying in the first half year following surgery, and this excess hazard never subsided to negligible levels at least up to 8 years after surgery. Interestingly having no prior record of illnesses worsened the initial mortality. Men living alone had the highest long-term excess mortality.