HEALTH Trial
Total Hip Arthroplasty or Hemiarthroplasty for Hip Fracture
HEALTH Investigators · N Engl J Med · 2019
In this international, multicentre randomized trial involving ambulatory patients aged 50 or older with displaced intracapsular hip fractures, total hip arthroplasty (THA) showed no significant difference compared to hemiarthroplasty (HA) in the risk of secondary hip operations at 24 months, while delivering only minor, clinically unimportant functional advantages and a trend toward increased dislocations.
Citation & links
HEALTH Investigators, Bhandari M, Einhorn TA, Guyatt G, Schemitsch EH, Zura RD, et al. Total Hip Arthroplasty or Hemiarthroplasty for Hip Fracture. N Engl J Med. 2019 Dec 5;381(23):2199-2208.
Key findings
- 1.Secondary surgical procedures at 24 months occurred in 7.9% of the THA group versus 8.3% of the HA group, demonstrating no statistically significant difference (HR 0.95, 95% CI 0.64–1.40, p = 0.79).
- 2.Postoperative hip instability or dislocation was twice as frequent after THA compared to HA (4.7% vs. 2.4%; HR 2.00, 99% CI 0.97–4.09).
- 3.Functional outcomes measured by total WOMAC scores slightly favoured THA over HA (mean difference -6.37 points at 24 months, 99% CI -9.18 to -3.56), but this fell short of the established minimal clinically important difference (MCID) of 9 to 22 points.
- 4.No clinically relevant or statistically significant differences were observed across general health-related quality of life metrics, including EQ-5D visual analogue scale, SF-12 physical/mental component scores, and Timed Up and Go (TUG) testing.
- 5.All-cause mortality was virtually identical between groups at 24 months (14.3% in THA vs. 13.1% in HA; p = 0.48).
- 6.Serious adverse events were common overall, occurring in 41.8% of THA patients and 36.7% of HA patients.
Background
Displaced intracapsular femoral neck fractures in active older individuals present a major clinical dilemma: whether to reconstruct the joint with total hip arthroplasty (THA) or hemiarthroplasty (HA). Prior observational data and smaller trials suggested THA yielded better long-term pain relief and lower revision rates compared to HA (which risks acetabular erosion), but carried higher perioperative complexity and instability rates. The HEALTH trial was conducted to resolve clinical equipoise regarding reoperation rates, complications, and patient-reported outcomes within 24 months post-injury.
Methods
The HEALTH study was an international, pragmatic, expertise-based randomized controlled trial conducted across 80 centres in 10 countries. A total of 1,495 previously independently ambulating patients aged 50 years or older with displaced femoral neck fractures were randomized to undergo either THA (n = 749) or HA (n = 746). Surgeries were performed by credentialed orthopaedic surgeons aligned with their procedural expertise. The primary outcome was the incidence of secondary hip procedures within 24 months. Secondary endpoints included all-cause mortality, serious adverse events, hip-specific complications (notably dislocation), and functional/quality-of-life assessments (WOMAC, EQ-5D, SF-12, Timed Up and Go).
Clinical implications
The HEALTH trial challenged the growing routine adoption of THA for unselected older patients with displaced intracapsular hip fractures. Because THA demonstrated no reduction in reoperation rates at 2 years and failed to deliver clinically meaningful functional gains over HA while elevating dislocation risk, hemiarthroplasty remains an excellent, reproducible standard of care. THA should be reserved selectively for younger, cognitively intact, highly active community ambulators or patients with pre-existing symptomatic hip osteoarthritis, rather than being applied broadly to all older independent ambulators.
Critical appraisal
Strengths include the large sample size across 10 countries, high methodological rigor, concealed allocation, and an expertise-based design minimizing surgeon learning curves. Limitations include a follow-up restricted to 24 months, which may fail to capture late acetabular erosion or wear complications commonly seen beyond 5–10 years after hemiarthroplasty. Additionally, 24-month patient-reported functional questionnaires had attrition (around 30–35% missing data), and surgical approaches (anterior vs posterior) and implant fixation methods were not standardized across cohorts.
Teaching note
For exit examinations, know the headline result of HEALTH: at 2 years, THA does not reduce secondary hip procedures compared with HA (approx. 8% in both) and WOMAC differences do not meet the MCID threshold, while THA doubles the dislocation rate (4.7% vs 2.4%). Remember to contrast this with NICE guidelines (CG124), which recommend THA only in patients who were independently mobile outdoors with no more than one stick, have no cognitive impairment, and are medically fit for anesthesia.
Exam-tested points
- ›Primary endpoint: No difference in secondary hip procedures within 24 months between THA (7.9%) and HA (8.3%) [HR 0.95; p = 0.79].
- ›Dislocation rate: Doubled in THA compared to HA (4.7% vs 2.4%, HR 2.00).
- ›Functional outcome: Total WOMAC scores favoured THA by 6.37 points, but this did not achieve the pre-specified minimal clinically important difference (MCID 9–22 points).
- ›Mortality & systemic complications: No significant difference in 24-month all-cause mortality (14.3% THA vs 13.1% HA).
- ›Trial design: Expertise-based randomized controlled trial (1,495 patients, 80 centres, 10 countries).
- ›Inclusion criteria: Age 50 years or older, displaced intracapsular femoral neck fracture, independently mobile prior to injury.
- ›Limitation to cite in viva: 24-month duration is insufficient to detect late acetabular cartilage erosion and groin pain in hemiarthroplasty cohorts.
Test yourself on Trauma
Practise SBA questions covering this subspecialty.