Cochrane in CORR®: Internal Fixation Implants for Intracapsular Hip Fractures in Older Adults

Importance of the Topic

Hip fractures are a major health concern that primarily affect older adults and can result in devastating outcomes, primarily for older adults, including severe pain, loss of mobility and subsequent inability to return to preinjury residence, and loss of independence [^{1, 2}]. Hip fractures are associated with a high 30-day mortality, though operative treatment substantially reduces year-end mortality attributable to the fracture itself [^{8, 14}]. Hip fractures are a substantial challenge to hospitals and local healthcare networks, with each patient often requiring weeks of inpatient admission and subsequent months of multidisciplinary rehabilitation and care. Furthermore, it is likely that we will see an increase in the number of hip fractures over time, owing to an aging population in most countries of the world [³].

Hip fractures can broadly be described as those occurring within the hip joint capsule (intracapsular) or outside of the capsule [¹⁰]. For both intra- and extracapsular fractures where the bone architecture and alignment are minimally disturbed (or nondisplaced), internal fixation with pins, screws, or plates can maintain the alignment of the fracture until it heals. However, subsequent displacement of the fracture after internal fixation often results in a second, more substantial operation, both to remove hardware and to replace part of the hip joint [¹⁵]. The morbidity, mortality, and complication rates associated with revision surgery following internal fixation are substantially higher than they are with primary internal fixation [¹¹].

Thus, it is crucial to understand what type of internal fixation yields the most benefits and least harms to patients who experience this common but severe injury.

Upon Closer Inspection

This Cochrane review included both randomized and quasi-randomized studies comparing internal fixation techniques for managing low-energy intracapsular hip fractures, mostly in older adults [⁹]. Implants were broadly grouped into three categories: smooth pins, screws, or fixed-angle plates. Thirty-eight studies with 8585 participants were included, with 38% of fractures being nondisplaced. Study outcomes were grouped into seven groups: activities of daily living, delirium, functional status, health-related quality of life, mobility, mortality, and unplanned reoperation.

Because it is impossible to blind outcomes assessment in these types of studies, all studies were found to be at a high risk of bias in this domain. However, it is unlikely that this type of bias is of great importance given that patients are unlikely to have a strong preference for one type of implant over another. Furthermore, only six studies were at low risk of both selection bias and bias due to inappropriate allocation concealment. Selection bias is likely to be more of a concern in these studies because inappropriate randomization sequence generation and allocation concealment can lead to surgeons selectively giving patients their preferred treatment (inadvertently or on purpose), thereby undoing the randomization process. Overall, every included study had high risk or unsure risk of bias in at least two risk of bias categories.

When comparing smooth pins versus fixed angle plates, there was no evidence of difference in mobility, mortality, or need for reoperation between any of the three major groups of interventions. Moreover, there were extremely limited data on activities of daily living, delirium, functional status, and health-related quality of life, and as such, no comparisons could be undertaken between these interventions. The FAITH trial found a significant difference in health-related quality of life for sliding hip screws and cancellous screws, particularly for current smokers [¹³], so it would be of value to explore this association further in additional studies.

When comparing screws to fixed angle plates, there was a slight improvement in health-related quality of life and functional status when screws were used, but neither difference was large enough to be clinically important (effect sizes were characterized by whether they reached predefined minimum clinically important difference thresholds). Additionally, the evidence suggesting differences was deemed low-certainty because of imprecision (wide confidence intervals, usually related to small sample sizes) and study limitations (particularly selection bias and lack of blinding outcomes assessors). An assessment of the remaining critical outcomes did not produce any meaningful differences. Lastly, when comparing smooth pins to screws, there were no observed differences in any critical outcome domain, with mostly low or very low certainty of evidence. It was not possible to conduct any subgroup analyses (for example, by age, gender, prefracture mobility, smoking status) because there were not enough studies in each category reporting on the same outcome to conduct meaningful analyses.

Unfortunately, by combining randomized trials with both low and high methodological rigor together, the overall pool of evidence is at higher risk of bias. For example, the FAITH trial provided both the largest cohort of patients to the meta-analysis (1079 of 3057 patients total) and was the trial at lowest risk of all forms of bias among studies included [¹²]. Pooling results with varying levels of risk of bias can lead to misleading conclusions [⁴]. Rather, presenting meta-analyses stratified by risk of bias or incorporating sensitivity analyses can provide a better assessment of true risk or benefit for an intervention [^{5, 6}].

Take-home Messages

The main finding of this Cochrane review—which summarized the best-available evidence from randomized trials—was that no type of implant used for internal fixation of hip fractures was demonstrably superior to any other. Although the Cochrane review concluded that the evidence was of low quality and that further studies are needed, there is one large high-quality study that concluded that there is no important difference between sliding hip screws and cancellous screws, and further trials are unlikely to change this conclusion. The other trials in this meta-analysis are smaller and of higher risk of bias, but their conclusions are generally the same. Therefore, we believe that surgeons can be relatively confident that there are no substantial differences between implant types and can use their preferred implant.

Future randomized controlled trials should include a validated hip fracture core outcome set. For example, Haywood et al. [⁷] recommends a core outcome set for hip fracture trials that includes mortality, pain, activities of daily living, mobility, and health-related quality of life. Individual trials may use such a core outcome set in addition to any trial-specific outcomes to improve ability to pool data in future meta-analyses.