Hemiarthroplasty vs Total Hip Replacement for Displaced Hip Fractures: HEALTH, WHiTE 5 and Who Gets What

A displaced intracapsular hip fracture in an older patient is replaced, not fixed. The argument at the trauma meeting is which replacement: hemiarthroplasty vs total hip replacement, and whether the hemiarthroplasty should be cemented. Two trials settle most of it. HEALTH showed that in patients who walked independently, a total hip replacement gave a small functional gain, no fewer reoperations and about twice the dislocation rate. WHiTE 5 showed that a cemented hemiarthroplasty did better than a modern uncemented one. The answer in one sentence: a cemented hemiarthroplasty for most, and a total hip replacement considered for the independent patient expected to benefit beyond 2 years. The fracture itself is covered on the neck of femur fracture topic page.

What did the HEALTH trial find?

The HEALTH investigators randomised 1,495 patients aged 50 or over with a displaced femoral neck fracture, at 80 centres in 10 countries, to total hip arthroplasty or hemiarthroplasty. Every patient had been able to walk without the help of another person before the fracture. The primary endpoint was a secondary hip procedure within 24 months.

How to read it. The primary result is null: the honest summary is that reoperation was the same, not that either operation was superior. The WOMAC scores favoured total hip arthroplasty, but the authors concluded that it gave "a clinically unimportant improvement" in function and quality of life over 24 months. The dislocation difference is about twofold, but its 99% confidence interval crosses 1.

The 2022 Cochrane review of arthroplasty for hip fracture (Lewis and colleagues) pooled 17 studies with 3,232 participants comparing the two. Dislocation was higher after total hip arthroplasty (RR 1.96, 95% CI 1.17–3.27; 12 studies), the 12-month mortality estimate was RR 1.00 (95% CI 0.83–1.22), consistent with both benefit and harm, and the estimate for unplanned return to theatre (RR 0.63, 95% CI 0.37–1.07) was low-certainty and included both benefit and no effect. The authors' conclusion matches HEALTH: any benefit of total hip arthroplasty "is likely to be small and not clinically appreciable".

What HEALTH cannot tell you. It enrolled only people who walked without another person's help, so it says nothing about the frail, dependent or institutionalised patient who makes up much of a hip-fracture list. And 24 months is short. Acetabular erosion after hemiarthroplasty is a late problem, so the follow-up may underestimate the difference that matters to a patient who lives for another decade.

What does NICE now recommend?

NICE CG124 recommends replacement arthroplasty, total hip replacement or hemiarthroplasty, for a displaced intracapsular fracture (recommendation 1.6.2). The choice between them was rewritten in January 2023, and the new wording is softer than the version most candidates learnt.

The current text is on the NICE CG124 recommendations page (opens in a new tab); the 2011 wording is preserved in the Royal College of Physicians summary (opens in a new tab). The committee's reasoning, published alongside it, explains each change.

• Why "consider". Some studies showed greater benefit from total hip arthroplasty, but for most outcomes the difference was not clinically or statistically significant. The guideline's own health economic model suggested total hip arthroplasty may have some benefit and be more cost-effective beyond 2 years; the evidence was not strong enough to recommend it for everyone.
• Why the 2-year horizon. Older people may not live long enough to see the long-term benefit, and people who are not very mobile may be less troubled by the consequences of a hemiarthroplasty, such as wear on the acetabulum. The committee also noted that hemiarthroplasty is less complicated and could mean less dislocation and less blood loss.
• Why cognition was dropped. The committee recognised that some people with significant cognitive impairment may be at greater risk of dislocation, but judged the evidence too limited for a specific recommendation. It asks clinicians to weigh comorbidities by their effect on function, and expects the change to stop people with mild cognitive impairment being excluded from total hip replacement unnecessarily. The decision is normally made by a multidisciplinary team.

Much exam teaching still quotes the 2011 criteria and a life expectancy of more than about 4 years. Quote the current wording, and name the year.

Who gets what?

The fracture decides that the head is replaced. The patient decides whether the acetabulum is replaced as well.

The structured version of this decision, with the arguments for each side, is on the total hip arthroplasty indications page.

Cemented or uncemented: what WHiTE 5 showed

WHiTE 5 was a UK multicentre trial of 1,225 patients aged 60 or over with an intracapsular hip fracture, randomised to a cemented hemiarthroplasty (610) or a modern hydroxyapatite-coated uncemented hemiarthroplasty (615). The primary outcome was EQ-5D utility at 4 months.

• Quality of life: EQ-5D 0.371 cemented against 0.315 uncemented (adjusted difference 0.055, 95% CI 0.009–0.101; P = 0.02). The difference at 1 month was similar; at 12 months it was smaller.
• Periprosthetic fracture: 0.5% against 2.1% (odds ratio for uncemented 4.37, 95% CI 1.19–24.00).
• Mortality at 12 months: 23.9% against 27.8% (OR 0.80, 95% CI 0.62–1.05), not a significant difference.
• Other complications: similar in the two groups.

Two cautions belong with the headline. Follow-up data were available for only 71.6% of patients at 4 months. And the trial's own range for the minimal clinically important EQ-5D difference was 0.050 to 0.075, so 0.055 sits at its bottom edge; the authors' word is "modestly". The within-trial economic evaluation found cemented implants cost-saving (mean −£961, 95% CI −£2,292 to £370) with a small QALY gain (0.010), and a 95–97% probability of being cost-effective at the thresholds tested.

The wider evidence points the same way. The 2022 Cochrane review (17 studies, 3,644 participants) found moderate-certainty evidence that cement improved quality of life and reduced 12-month mortality (RR 0.86, 95% CI 0.78–0.96), with fewer intraoperative (RR 0.20) and postoperative (RR 0.29) periprosthetic fractures. It also found a higher risk of pulmonary embolus with cement (RR 3.56, 95% CI 1.26–10.11). The Norwegian Hip Fracture Register (7,539 uncemented and 22,639 cemented bipolar hemiarthroplasties) found 1.5 times the risk of reoperation for any reason after uncemented fixation, and 5.1 times the risk of reoperation for periprosthetic fracture.

NICE's position is direct: use cemented implants in people having arthroplasty (1.6.4), and hospitals should aim to use a single type of cemented femoral component for hemiarthroplasty as standard (1.6.5, 2023).

What about bone cement implantation syndrome?

Cement is not free. The joint safety guideline from the AAGBI, BOA and British Geriatrics Society (2015) reports adverse cardiovascular events in about 20% of hip fracture operations using a cemented prosthesis: grade 1 (saturation under 94% or a fall in systolic pressure over 20%) in about 20%, grade 2 in about 3%, and grade 3, needing cardiopulmonary resuscitation, in about 1%. The risk is higher with increasing age, male sex, significant cardiopulmonary disease and diuretic use.

The guideline's answer is a three-stage process, not avoiding cement:

1. Identify the high-risk patient before the list.
2. Prepare the team, with a multidisciplinary discussion where appropriate, a pre-list briefing and the WHO checklist time-out.
3. Assign roles. The surgeon tells the anaesthetist before cementing, washes and dries the canal (pressurised lavage is recommended), inserts cement retrograde from a gun over a plug with a distal suction catheter, and avoids vigorous pressurisation in high-risk patients. The anaesthetist keeps the patient well filled, monitors blood pressure with an arterial line or non-invasive cuff on the "stat" setting, aims to keep systolic pressure within 20% of the pre-induction value, and has vasopressors ready.

The grading, mechanism and differential diagnosis of the syndrome are on the PMMA bone cement page. The hemiarthroplasty operative guide reserves an uncemented stem for patients with significant cardiopulmonary disease or pulmonary hypertension, a previous episode of the syndrome, or a canal where cementing is impractical.

Unipolar or bipolar hemiarthroplasty?

The evidence does not separate them. The 2022 Cochrane review pooled 13 studies with 1,499 participants and found little or no difference in early or 12-month mortality, with low-certainty or very low-certainty evidence on the other outcomes. Dislocation was uncommon, about 1.6%, with no evidence of a difference between designs. The authors concluded that there is "no evidence to suggest a bipolar HA is superior to a unipolar prosthesis". A bipolar head is a reasonable choice; it is not one you have to defend with outcome data.

Dislocation, approach and dual mobility

Dislocation is the main cost of choosing a total hip replacement for a fracture. The usual explanation is the soft tissues: a fractured hip lacks the capsular thickening and contracture of a chronically arthritic one, so a total hip replacement put in for a fracture is less stable than one put in for osteoarthritis. The THA dislocation page sets out the risk factors and the bearing choices.

Which approach for a hemiarthroplasty? NICE still says to consider an anterolateral approach in preference to a posterior one (1.6.8), a recommendation last reviewed in 2011. The trial evidence since is weak in both directions. Parker randomised 216 patients having a cemented hemiarthroplasty to a lateral or posterior approach and found no significant difference in mortality, residual pain or regain of walking at one year; surgeons found the lateral approach subjectively easier. The 2025 Cochrane review of approaches (27 studies, 3,369 participants) concluded there is insufficient evidence to say whether the anterior, lateral or posterior approach is better. The signals it did find were low-certainty: earlier independent walking with the anterior approach than the posterior (RR 1.64), and, in a single study, less Trendelenburg gait with the anterior approach than the lateral.

If you use the posterior approach, repair it. The hemiarthroplasty operative guide quotes enhanced repair of the capsule and short external rotators as reducing dislocation from 5–10% to 1–2%, and the dislocation topic gives the posterior approach two to three times the dislocation rate of the anterior and anterolateral approaches in total hip replacement.

Dual mobility. For the patient who needs a total hip replacement but is at high risk of dislocation, a dual-mobility cup is a common choice, but the evidence is observational. A 2020 meta-analysis of 23 studies (7,189 patients, mean age 77.8) found lower dislocation with dual mobility than with a conventional total hip replacement (OR 0.26) or a hemiarthroplasty (OR 0.27). An international registry meta-analysis in 2022 (15,024 dual-mobility and 97,200 conventional total hip replacements for fracture) found no reduction in all-cause revision (HR 0.96, 95% CI 0.86–1.06; cumulative revision at 5 years 4.7% vs 4.3%), with fewer revisions for dislocation but more for infection; in most of the registries the dual-mobility patients were older and had more comorbidities. Use it selectively. The concept is covered on the dual mobility page and in our post on dual mobility hips.

The late cost of a hemiarthroplasty

A hemiarthroplasty puts a metal head against native acetabular cartilage. Over time the head can erode the acetabulum and cause groin pain, and the remedy is conversion to a total hip replacement. NICE's committee noted that people who are not very mobile may be less concerned about wear on the acetabulum. The patient who will be walking for many years is the one most likely to meet it.

The technique of the salvage operation is in the conversion of a failed hemiarthroplasty guide.

In the exam

Examiners want the decision justified from the patient, the trials quoted with their limits, and NICE quoted in its current form.