Internal Fixation versus Arthroplasty for Displaced Femoral Neck Fractures—A Systematic Review and Meta-analysis
Abstract
Abstract
In the US, there were about 300,000 hip fractures in 1991,1 a number that could triple by 2050.2 About half of hip fractures occur at the femoral neck,3 of which approximately 70% are displaced fractures.4,5 Options for surgical repair of displaced femoral neck fractures include internal fixation (IF), in which the femoral head is preserved and the fracture is reduced and fixed in place using either single or multiple screws or pins, or arthroplasty (AR), in which the femoral head is replaced with a prosthesis. AR subtypes are total hip arthroplasty (THA), in which the acetabulum is also replaced, and hemi-arthroplasty (HA), in which the natural acetabulum is preserved.
Meta-analyses of randomized trials comparing IF versus AR for the treatment of displaced femoral neck fractures have reported no significant mortality difference between IF and AR, reduced risk for revision surgery with AR, and limited data suggesting that AR may be associated with reduced post-operative pain and better functional recovery.6–9 Although the question of whether IF or AR is best in these patients has long been controversial, only 2% of respondents to a recent survey of American Association of Hip and Knee Surgeons reported that IF was their preferred treatment option in patients 65 years of age or older.10 The most important patient factors they cited for not performing IF were metabolic bone disease or osteoporosis, fracture comminution, or pre-fracture hippain, whereas patient age, pre-fracture ambulatory status, neuromuscular disease, and other comorbidities were rated as less important. A survey of members of the Orthopaedic Trauma Association and European AO International-affiliated trauma centers reported that 75–89% of surgeons preferred IF over AR in patients less than 60 years of age compared with 11–25% in patients 60–80 years of age and 4–6% in patients over 80 years of age.11 Approximately half of these surgeons preferred IF in active patients with Garden type III fractures, but they were more likely to prefer AR in frail patients and those with Garden type IV fractures. However, previous meta-analyses have not reported the direct impact of these and other patient characteristics on outcomes after IF versus AR.
Therefore, in the current systematic review and meta-analysis, we aimed to compare IF versus AR for the treatment of displaced femoral neck fractures, limiting our evaluation to randomized trials and highlighting, where available, results for specific relevant patient subgroups.
Patients and Methods
Literature Search
Trials were identified using MEDLINE computer databases for the period January 1, 1991 through April 1, 2008, utilizing an optimally sensitive Cochrane Collaboration search strategy.12 The search strategy combined terms to identify fracture or procedure type (e.g. hip fracture, hip neck fracture, hip fixation, or hip implant) with terms to identify controlled treatment trials (e.g. clinical trial, controlled clinical trial, randomized controlled trial, or multicenter study). In addition, bibliographies of retrieved trials and review articles were reviewed.
Selection Criteria
Trials were eligible if they: included subjects with femoral neck fractures; compared surgical repair of displaced femoral neck fractures by IF with repair by AR; were randomized; assessed relevant post-operative outcomes at least one year after hip fracture surgery (e.g. mortality, complications, ambulation status); included at least 10 subjects; were written in English; and were published on January 1, 1991 or later so as to focus on trials that did not use outdated implants and that were more likely to have been conducted in settings utilizing contemporary perioperative care. Two reviewers independently assessed study eligibility. Differences in eligibility assessments were resolved by discussion.
Outcome Measures
Information on trial characteristics, patient demographics, surgical procedure type, dropouts, and post-operative outcomes was extracted in a standardized fashion and independently reviewed for accuracy. Post-operative outcomes evaluated were mortality, operative revisions, post-operative mobility, postoperative pain, and health-related quality of life (HRQOL).
Assessment of Methodological Quality
We assessed the quality of concealment of randomized treatment allocation according to a scale developed by Schulz, assigning 1 to poorest quality and 3 to best quality.13 Additionally, we assessed whether trial participants and investigators were blinded to the treatment provided, whether trials used an intention-to-treat (ITT) analysis, and the percentage of subjects who dropped out or were lost to follow-up.
