Aims

Patients with femoral neck fractures (FNFs) treated with total hip arthroplasty (THA) have an almost ten-fold increased risk of dislocation compared to patients undergoing elective THA. The surgical approach influences the risk of dislocation. To date, the influence of differing head sizes and dual-mobility components (DMCs) on the risk of dislocation has not been well studied.

Dislocation after total hip arthroplasty in individuals treated for acute hip fracture is up to 10 times more frequent than in elective patients. Whilst approach plays a role, the effect of head sizes in conventional THA and dual mobility cups (DMC) is less studied in fracture cases.

The total dislocation rate at 1-year and 3-year revision rates were recorded in this observational study on 8,031 patients with acute hip fracture, treated with a THA 2005–2014. Swedish Arthroplasty Register data were linked with the National Patient Register. Cox multivariable regression models were fitted to calculate adjusted hazard ratios stratified by approach and head size.

The cumulative risk of dislocation during year 1 was 2.7% (95% CI 2.2–3.2) with lateral approach and 8.3% (7.3–9.3) with posterior approach (KM estimates).

In the posterior approach group DMC was associated with a lower risk of dislocation compared to cTHA=32mm (HR=0.21; 0.07–0.68), whilst a head size <32mm carried a higher risk (HR=1.47; 1.10–1.98). These differences were no longer visible when revision in general was used as outcome.

Neither of the implant designs influenced the dislocation risk when direct lateral approach was used. Male gender and severe comorbidity increased the risk. DMC with lateral approach was associated with a reduced risk of revision in general (HR=0.36; 0.13–0.99). Head size did not influence the revision risk.

When aiming to reduce the risk of any dislocation, lateral approach – regardless of cup/head design – is referable. If, for any reason, posterior approach is used, DMC is associated with the lowest risk of dislocation. This is not reflected in analysing revision in general as outcome. An interpretation could be that there are different thresholds for dislocation prompting revision.