Outcomes of 807 Thompson hip hemiarthroplasty procedures and the effect of surgical approach on dislocation rates

The majority of displaced intracapsular fractures in our unit are managed with a Thompson hip hemiarthroplasty. Recent UK guidance from the National Institute for Health and Care Excellence has, however, advised against the continued used of the Thompson implant in patients with hip fracture. The aim of this study was to review the outcomes and complications after Thompson hip hemiarthroplasty, including the impact of modern surgical approaches and cementing, whilst controlling for confounding factors. We reviewed the outcomes following Thompson hip hemiarthroplasty from a series of 807 cases performed between April 2008 and November 2013. Of these, 721 (89.3%) were cemented and 86 (10.7%) uncemented. A total of 575 (71.3%) procedures were performed in female patients. The anterolateral approach was performed in 753 (93.3%) and the posterior approach with enhanced soft tissue repair in 54 (6.7%). Overall, there were 23 dislocations (2.9%). Dislocation following the posterior approach occurred in 13.0% (seven of 54) in comparison to 2.1% (16 of 753) with the anterolateral approach (odds ratio (OR) 8.5 (95% confidence interval (CI) 2.8-26.3), p < 0.001). Patients were discharged home in 459 cases (56.9%), to a care home or other hospital in 273 cases (33.8%). Of the total number of patients, 75 died during their admission (9.3%), and 51.8% (338 of 653) returned home within 30 days. The 30-day mortality was 7.1% (57 cases) and the 1-year mortality was 16.6% (116 of 699). We recommend against the continued use of the posterior approach in hip hemiarthroplasty, as enhanced soft tissue repair did not reduce the dislocation rates to an acceptable level in this series utilising the Thompson implant. Our findings, however, demonstrate satisfactory results for patients treated with the Thompson hip hemiarthroplasty performed through an anterolateral approach. We suggest that the continued use of this implant in a carefully selected patient cohort is justifiable. (C) 2014 Elsevier Ltd. All rights reserved.