Treatment of common hip fractures.

Objectives: To conduct a systematic review and synthesize the evidence for the effects of surgical treatments for subcapital and intertrochanteric/subtrochanteric hip fractures on patient-focused outcomes for elderly patients.

Data sources: MEDLINE, Cochrane databases, Scirus, and ClinicalTrials.gov, and expert consultants. We also manually searched reference lists from relevant systematic reviews.

Review methods: High quality quasi-experimental design studies were used to examine relationships between patient characteristics, type of fracture, and patient outcomes. Randomized controlled trials were used to examine relationships between type of surgical treatment and patient outcomes. Patient mortality was examined with Forest plots. Narrative analysis was used for pain, quality of life (QoL), and functional outcomes due to inconsistently measured and reported outcomes.

Results: Mortality does not appear to differ by device class, or by devices within a class. Nor, on the whole, do pain, functioning, and QoL. Some internal fixation devices may confer earlier return to functioning over others for some patients, but such gains are very short lived. Very limited results suggest that subcapital hip fracture patients with total hip replacements have improved patient outcomes over internal fixation, but it is unclear whether these results would continue to hold if the analyses included the full complement of relevant covariates. Age, gender, prefracture functioning, and cognitive impairment appear to be related to mortality and functional outcomes. Fracture type does not appear to be independently related to patient outcomes. Again, however, the observational literature does not include the full complement of potential covariates and it is uncertain if these results would hold.

Conclusions: Several factors limit our ability to definitively answer the key questions posed in this study using the existing literature. Limited perspectives lead to incomplete sets of independent variables included in analyses. Specific populations are poorly defined and separated for comparative study. Fractures with widely varying biomechanical problems are often lumped together. Outcome variables are inconsistently measured and reported, making it very difficult to aggregate or even compare results. If future high quality trials continue to support the evidence that differences in devices are short term at best, within the first few weeks to few months of recovery, policy implications involve establishing the value of a shorter recovery relative to the cost of the new device. As the literature generally focuses on community dwelling elderly patients, more attention needs to be directed toward understanding implications of surgical treatment choices for the nursing home population.