Why this is important

The Guideline Development Group’s consensus decision to recommend CT over a radionuclide bone scan as an alternative to MRI to detect occult hip fractures reflects current NHS practice but assumes that advances in technology have made the reliability of CT comparable with that of MRI. If modern CT can be shown to have similar reliability and accuracy to MRI, then this has considerable implications because of its widespread availability out of hours and lower cost. It is therefore a high priority to confirm or refute this assumption by direct randomised comparison. The study design would need to retain MRI as the ’gold standard’ for cases of uncertainty and to standardise the criteria, expertise and procedures for radiological assessment. Numbers required would depend on the degree of sensitivity and specificity (the key outcome criteria) set as target requirement for comparability, but need not necessarily be very large. [2011]

Why this is important

No recent randomised controlled trials were identified that fully address this question. The evidence is old and does not reflect current practice. In addition, in most of the studies the patients are sedated before regional anaesthesia is administered, and this is not taken into account when analysing the results. The study design for the proposed research would be best addressed by a randomised controlled trial. This would ideally be a multicentre trial including 3,000 participants in each arm. This is achievable given that there are about 70,000 to 75,000 hip fractures a year in the UK. The study should have 3 arms that look at spinal anaesthesia versus spinal anaesthesia plus sedation versus general anaesthesia; this would separate those with regional anaesthesia from those with regional anaesthesia plus sedation. The study would also need to control for surgery, especially type of fracture, prosthesis and grade of surgeon.

A qualitative research component would also be helpful to study patient preference for type of anaesthesia. [2011]

Why this is important

Between 5% and 15% of people with an intracapsular hip fracture will have an undisplaced fracture. There is variation in the UK in how undisplaced intracapsular hip fractures are recognised, resulting in some people not being offered the most appropriate treatment. Research is needed to help healthcare professionals understand the clinical characteristics of people who have undisplaced hip fracture (on anterior-posterior and lateral X-rays) and how this relates to the effectiveness of different treatment strategies.

The committee also noted a paucity of evidence for 2 of the interventions (total hip replacement and hemiarthroplasty) that could potentially be useful for people with undisplaced intracapsular hip fracture. A randomised controlled trial comparing these interventions would be beneficial. [2017]

Why this is important

The rapid restoration of physical and self-care functions is critical to recovery from hip fracture, particularly where the goal is to return the person to preoperative levels of function and residence. Approaches that are worthy of future development and investigation include progressive resistance training, progressive balance and gait training, supported treadmill gait re-training, dual task training and activities of daily living training. The optimal time point at which these interventions should be started requires clarification.

The ideal study design is a randomised controlled trial. Initial studies may have to focus on proof of concept and be mindful of costs. A phase 3 randomised controlled trial is required to determine clinical effectiveness and cost effectiveness. The ideal sample size will be around 400 to 500 patients, and the primary outcome should be physical function and health-related quality of life. Outcomes should also include falls. A formal sample size calculation will need to be undertaken. Outcomes should be followed over a minimum of 1 year, and compare if possible, either the recovery curve for restoration of function or time to attainment of functional goals. [2011]

Why this is important

Residents of care and nursing homes account for about 30% of all people with hip fracture admitted to hospital. Two-thirds of these come from care homes and the remainder from nursing homes. These people are frailer, more functionally dependent and have a higher prevalence of cognitive impairment than people admitted from their own homes. One-third of those admitted from a care home are discharged to a nursing home and one-fifth are readmitted to hospital within 3 months. There are no clinical trials to define the optimal rehabilitation pathway following hip fracture for these people and therefore represent a discrete cohort where the existing meta-analyses do not apply. As a consequence, many people are denied structured rehabilitation and are discharged back to their care home or nursing home with very little or no rehabilitation input.

Given the patient frailty and comorbidities, rehabilitation may have no effect on clinical outcomes for this group. However, the fact that they already live in a home where they are supported by trained care staff clearly provides an opportunity for a systematic approach to rehabilitation. Early multidisciplinary rehabilitation based in care homes or nursing homes would take advantage of the day-to-day care arrangements already in place and provide additional NHS support to deliver naturalistic rehabilitation, where problems are tackled in the person’s residential setting.

Early supported multidisciplinary rehabilitation could reduce hospital stay, improve early return to function, and affect both readmission rates and the level of NHS-funded nursing care required.

The research would follow a 2-stage design: (1) an initial feasibility study to refine the selection criteria and process for reliable identification and characterisation of those considered most likely to benefit, together with the intervention package and measures for collaboration between the Hip Fracture Programme team, care-home staff and other community-based professionals and (2) a cluster randomised controlled comparison (for example, with 2 or more intervention units and matched control units) set against agreed outcome criteria. The latter should include those specified above, together with measures of the impact on care-home staff activity and cost, as well as qualitative data from patients on relevant quality-of-life variables. [2011]

Why the committee made the recommendation

The committee discussed the clinical evidence on total hip arthroplasty versus hemiarthroplasty. They agreed that although some studies showed greater benefits for total hip arthroplasty, this was not clinically or statistically significant for most outcomes. However, a combination of the clinical evidence and the health economic model developed as part of the guideline indicated that total hip arthroplasty may have some benefits and be more cost effective than hemiarthroplasty beyond 2 years. The committee noted that although recommendation 1.6.2 states that clinicians should offer arthroplasty (either total hip arthroplasty or hemiarthroplasty) to people with a displaced intracapsular hip fracture, hemiarthroplasty tends to be used more often than total hip arthroplasty. The evidence was not strong enough for them to recommend total hip arthroplasty for everyone with a displaced intracapsular fracture.

Based on their clinical knowledge and experience, the committee discussed how the long-term outcomes considered in the health economic model were important but may not be relevant to some people. For example, older people may not live long enough to experience the long-term benefits of total hip arthroplasty, and people who are not very mobile may be less concerned about the potential consequences of having a hemiarthroplasty, such as wear on the acetabulum. The committee agreed that hemiarthroplasty was a less complicated procedure than total hip arthroplasty and could result in lower dislocation rates and less blood loss.

The health economic evidence on the long-term cost effectiveness and potential clinical benefits of total hip arthroplasty led the committee to recommend that clinicians should consider the procedure for those who are most likely to benefit from it beyond 2 years. The list of criteria in the recommendation represents the past (a person’s level of independence before the fracture), present (how they currently present in hospital and if they are fit for the procedure on that day) and future (how much they are likely to benefit beyond 2 years). Including this list gives clinicians more discretion over who to offer total hip arthroplasty to and prevents the procedure being offered to some people who may get the same, or more, benefit from hemiarthroplasty.

The committee discussed how some people with significant cognitive impairments may be at increased risk of dislocations and could be less likely to benefit from total hip arthroplasty. However, they agreed that the evidence for this was too limited to make a specific recommendation for this population. The risk of dislocation can also vary depending on the severity and type of cognitive impairment, or how much support the person has. They agreed that cognitive impairment is one of many important comorbidities that should be considered when making treatment decisions. It is more important for clinicians to think about comorbidities in the context of functionality rather than whether or not a person has them. The committee also agreed that decisions about whether someone is likely to benefit most from total hip arthroplasty or hemiarthroplasty would normally be made as part of a multidisciplinary team.

The committee discussed the potential long-term benefit of total hip arthroplasty in specific groups of people, in particular younger age groups with fewer or less severe comorbidities. As the evidence did not provide much long-term data, and results were not reported for different age categories, it was agreed that further research should be carried out to inform future recommendations. A recommendation for research on long-term effectiveness of total hip replacement was therefore included to highlight the importance of comparing the effectiveness of total hip arthroplasty with hemiarthroplasty in the long term and determining the effect of each type of arthroplasty on different population subgroups.

How the recommendation might affect practice

The recommendation allows clinicians to use their discretion in deciding who is offered total hip arthroplasty. It should prevent people with mild forms of cognitive impairment being excluded from total hip arthroplasty unnecessarily. As more data becomes available on the long-term benefits of total hip arthroplasty in specific subgroups, there may be an increase in the number of people who are considered for total hip arthroplasty.

Return to recommendation

Why the committee made the recommendations

The committee discussed the evidence on people who had been given Thompson, Exeter/Unitrax or Exeter Trauma Stem (ETS) components and agreed that health-related quality of life, mobility, mortality, unplanned return to theatre and adverse-event outcomes were similar across all groups. The committee noted that although there were no cost-effectiveness studies, there was a large amount of variability in femoral component costs across the country for a given type of femoral component and between different types of femoral component.

The Thompson component was cheaper than the ETS or Exeter/Unitrax component, but the committee were aware of future regulatory changes requiring data about implants, meaning that some older designs are unlikely to be used in the future. Without further evidence on other cemented components currently in use, they were unable to recommend one femoral component over another.

To choose the most cost-effective option, the committee agreed it was important for hospitals to consider not only the cost of the component itself, but also the cost of training needs when switching to a new component, alongside any future costs relating to adverse outcomes. There may also be other considerations, in addition to costs. For example, some hospitals may choose to use a femoral component that is suitable for both hemiarthroplasty and total hip arthroplasty to allow consistency and greater efficiency in practice. The committee thought it was important from a training and development perspective that medical teams become familiar with implanting 1 single type of component as standard. They agreed that more research was needed on the effectiveness of different components.

The committee agreed that although the observational evidence was for femoral components not used in the UK, it did emphasise the importance of registry data in exploring longer-term adverse outcomes such as periprosthetic fracture in trauma patients who had undergone hemiarthroplasty. Recording data on hemiarthroplasties for submission to a national registry, such as the National Joint Registry, will help to provide real-world data on the long-term effectiveness and safety of different femoral components in trauma patients.

The committee commented that the 2011 recommendation to use a proven femoral component design (based on Orthopaedic Device Evaluation Panel ratings) came from evidence of people having elective surgery. They queried whether femoral component designs for elective patients who have arthritis were appropriate for trauma patients, given that arthritis often puts people at greater risk of fractures. Therefore, the committee drafted a recommendation for research on femoral component design that would allow data for this fragility fracture population to be captured. Registry data could also be used to evaluate long-term effectiveness in specific subpopulations such as people from different ethnic backgrounds and other groups for which there is currently no evidence.

How the recommendations might affect practice

By recommending 1 femoral component as standard for hemiarthroplasties, surgical teams will become familiar operating with this prosthesis and need less training in different components. Hospitals or trusts will also choose a component that provides the best value for money, but within the context of training requirements, team familiarity and overall costs.

The National Joint Registry already collects data on total hip arthroplasties. Collecting data on hemiarthroplasties in this, or a similar database, may require some extra administrative work. But the real-world data will be valuable in helping future decision makers choose the most clinically and cost-effective femoral component. Having further research on the effectiveness of different femoral components in people from different population groups will also help inform decisions and address health inequalities in this area.

Return to recommendations