Hip Fracture Emergency | Garden Classification | Fix vs Replace
- Garden I/II = undisplaced (fix). Garden III/IV = displaced (replace in elderly)
- Intracapsular location = AVN risk (blood supply)
- Surgery within 36-48 hours is the service standard - it shortens stay and reduces pressure sores and delirium. Note what it does NOT rest on: HIP ATTACK randomised 6-hour against 24-hour surgery and found no mortality or major-complication difference, so avoid saying early surgery 'reduces mortality'
- Hemiarthroplasty vs THR: Depends on cognitive status, mobility, age
- Young patients with displaced fractures: attempt reduction and fixation
- “Blood supply: lateral epiphyseal vessels from MFCA
- “Displaced fractures in elderly: arthroplasty (hemi or THR)
- “Garden I may progress to displaced if not fixed
- “Cannulated screws in inverted triangle configuration
Overview and Anatomy
Who. Neck of femur fractures are common in the elderly with osteoporosis and are a major cause of morbidity and mortality. The mechanism and the management differ between the elderly and the young, so the patient's age runs through every decision on this page.
The intracapsular location. A neck of femur fracture lies within the hip joint capsule. The vessels that feed the femoral head run along the neck inside that capsule, so displacement of the fracture puts the head's blood supply, and therefore the head, at risk of avascular necrosis.
The blood supply. The medial femoral circumflex artery (MFCA), a branch of the profunda femoris, wraps around the femoral neck posteriorly and gives rise to the lateral epiphyseal vessels, which run along the posterior-superior aspect of the femoral neck within the capsule and are the main supply to the femoral head. Displaced fractures disrupt them. Retrograde flow from the ligamentum teres contributes minimally.
The calcar femorale. The calcar is a dense vertical plate of cortical bone arising from the posteromedial femoral shaft beneath the lesser trochanter and radiating upward and laterally into the femoral neck. It is not the medial cortex (the visible medial "Adam's arch") but an internal weight-bearing buttress that reinforces the neck-shaft junction and transmits load from the neck to the shaft. It lies in the path of the primary compressive trabeculae, completing the medial load-transfer column of the proximal femur.
Why the calcar matters. The inferior (calcar) screw or lag screw is placed along this dense buttress, where it gains the best purchase and provides the medial support that resists varus collapse; loss of calcar or posteromedial continuity through comminution removes the buttress and predicts fixation failure. At arthroplasty, a medial calcar that is deficient or resected too low lets a standard stem subside, which is why a calcar-replacing stem is used to restore medial support and leg length.
Trabecular architecture. The neck contains two main trabecular systems. The primary compressive trabeculae run vertically from the medial calcar to the superior femoral head, resist compressive load during weight-bearing and are the strongest system; the primary tensile trabeculae arc from the lateral cortex to the inferior femoral head, resist the tensile forces along the inferior neck and are important for fracture stability. Secondary compressive trabeculae run from the greater trochanter to the head and provide additional support. Ward's triangle, the area of relative weakness with few trabeculae between the systems, is a common site of fracture initiation in osteoporotic bone.
Pathophysiology
Mechanism. Two mechanisms produce the fracture. The most common is a low-energy fall in an elderly osteoporotic patient; the other is high-energy trauma in a younger patient, typically a road traffic accident or a fall from height.
What displacement does to the head. A displaced fracture disrupts the lateral epiphyseal vessels and avascular necrosis follows; the rate, given under Complications, is higher when reduction is delayed beyond 6 hours.
Why the intracapsular location matters for healing. There is no periosteal sleeve at this level, so the fracture lacks the periosteal contribution to healing and relies on endosteal healing. Synovial fluid washes out the fracture haematoma that healing needs. The nonunion rate under Complications follows from both.
Capsular tamponade, the second hit. Displacement tears the retinacular vessels directly, the dominant and irreversible hit in Garden III-IV, but there is a second, potentially modifiable vascular insult. An intracapsular fracture bleeds into a closed, non-distensible capsule and produces a tense haemarthrosis; the rising intracapsular pressure can exceed capillary perfusion pressure and tamponade the already-compromised retinacular vessels, adding a secondary ischaemic hit to the femoral head. This is the rationale offered for early decompression by capsulotomy or joint aspiration.
Where the argument is strongest, and what the evidence says. In undisplaced (Garden I-II) fractures the retinacular vessels are often still intact, so a tense haemarthrosis is the main reversible threat to the head and the theoretical case for decompression is strongest. Routine capsulotomy has not been shown to reduce AVN (the FAITH trial found no benefit), so it is not mandatory. Many surgeons still decompress in delayed presentations (consider it beyond 12 hours) or aspirate or capsulotomise when fixing undisplaced fractures, accepting that the benefit is unproven.
Classification Systems
Neck of femur fractures can be classified by several systems, each providing different prognostic and treatment information. Garden is the everyday language, Pauwels the biomechanical one, AO/OTA the registry one, and Delbet the paediatric one.

Garden (1961) grades the fracture by the degree of displacement and the alignment of the trabecular lines on the AP radiograph. It correlates with AVN risk and guides treatment.
- Fracture
- Incomplete; head tilted into valgus (impacted position)
- Trabecular lines
- Continuous across the fracture
- AVN risk
- 0-10%
- Treatment
- Cannulated screw fixation
- Fracture
- Complete; no displacement; head in neutral position
- Trabecular lines
- Lines of the head align with acetabulum and neck
- AVN risk
- 10-20%
- Treatment
- Cannulated screw fixation
- Fracture
- Complete; partial displacement; head partially rotated but still in some contact with the neck
- Trabecular lines
- Lines of the head misaligned with the neck
- AVN risk
- 20-30%
- Treatment
- Arthroplasty (elderly); urgent reduction and fixation (young)
- Fracture
- Complete; total displacement; head fully dissociated from the neck
- Trabecular lines
- Lines of the head align with the acetabulum but not with the neck
- AVN risk
- 30-100%
- Treatment
- Arthroplasty (elderly); urgent reduction and fixation (young)
The simplification that survives. Garden I-II are undisplaced and are fixed; Garden III-IV are displaced and, in the elderly, are replaced.

Reliability, the number that matters. The four-type system is only moderately reproducible. Thomsen (Int Orthop 1996) had six observers classify 96 fractures and all six agreed on just 14 (15%), kappa 0.39; Cazzato (BMC Musculoskelet Disord 2022) found kappa 0.28-0.73 (mean 0.49) across six surgeons and 150 fractures, with greater experience giving no improvement. Collapse the system to undisplaced versus displaced and reliability rises to kappa 0.68-0.90; stage II versus III is the weak point. The 2018 AO/OTA revision performed worse (mean kappa 0.30), so no radiographic system is clearly superior, and Garden remains the everyday language despite the limitation.
What Garden does not see. Posterior comminution is visible only on the lateral film and predicts loss of reduction even when the AP looks acceptable.
"Incomplete" is how Garden I is classically described and how it is still examined, so learn it that way, but know the challenge to it, because it is the kind of nuance that separates a good viva answer from a recited one.
Cai and colleagues (BMC Surg 2022) found that many fractures called Garden I on plain films are complete when imaged on CT, and questioned whether a truly incomplete femoral-neck fracture exists at all in osteoporotic bone. The mechanism is intuitive: an osteoporotic neck has little structural reserve to arrest a propagating fracture line, and the valgus impaction that makes the fracture look stable on an AP film can conceal a break that runs the whole way across.
The undisplaced fractures are not actually undisplaced. Wang and colleagues measured 120 consecutive "undisplaced" femoral-neck fractures (60 Garden I, 60 Garden II) using 3D CT reconstruction and reported that many had been misclassified as incomplete on the AP radiograph. Impacted Garden I fractures carried a mean femoral-head-centre displacement of 6.22 mm and a mean rotational displacement of 17.8 degrees, with no significant difference from Garden II; only the genuinely incomplete fractures differed (rotation 4.9 degrees). The 3D method was highly reproducible (inter-observer kappa 0.94), so the disagreement is with the radiographic classification, not the measurement.
Which way the error runs. Toward under-treatment: a fracture read as incomplete invites the most conservative option when the underlying injury is complete, rotated and mechanically unstable. Treat an apparent Garden I in an older patient as a fracture that may be complete, and do not let the word "incomplete" argue you out of fixation. Note too that an undisplaced fracture with marked posterior tilt (over about 20 degrees on the lateral) has a high fixation-failure rate and may be better served by arthroplasty.
Observer variation in the radiographic classification of fractures of the neck of the femur using Garden's system
- Six observers classified 96 femoral-neck fractures; all six agreed on only 14 (15%)
- Poor agreement for the full four-type system (kappa 0.39), becoming acceptable when reduced to undisplaced (I-II) versus displaced (III-IV) (kappa 0.68)
- Distinguishing Stage II from Stage III remained problematic
Femoral neck fracture: the reliability of radiologic classifications.
- Six surgeons classified 150 femoral-neck fractures using Garden, 2018 AO/OTA and simplified AO/OTA systems
- Garden had only fair-to-moderate interobserver reliability (kappa 0.28-0.73, mean 0.49); the 2018 AO/OTA was worse (mean 0.30) and the simplified AO/OTA similar to Garden (mean 0.48)
- No classification was superior, and greater surgeon experience did not improve reliability
Clinical Presentation
The older patient. The story is a fall from standing height, at home (bathroom, bedroom) or on the street or stairs, after which the patient cannot weight-bear and has pain in the groin or hip, or referred to the knee. The patient may report hearing or feeling a snap or pop at the moment of injury. The risk factors to ask about:
- Osteoporosis
- Age greater than 65 years
- Previous fragility fracture
- Low BMI
- Sedentary lifestyle
The young patient. A road traffic accident, a fall from height or a sports injury, with severe pain and inability to move the leg. There may be polytrauma with other injuries.
Inspection. The patient lies still and is reluctant to move the affected leg. The classic posture is a leg that is shortened, externally rotated (typically 90 degrees) and adducted. Look for swelling, bruising (which may be minimal acutely) and deformity.
Palpation and neurovascular examination. There is tenderness over the groin or hip, and pain on gently log-rolling the leg through internal and external rotation, which is the specific test. Check the dorsalis pedis and posterior tibial pulses and sciatic nerve function (foot dorsiflexion and plantarflexion, and sensation).
Movement. The patient cannot straight-leg raise or flex the hip, and any attempt at passive movement is painful. Do not stress the fracture: it risks displacement.
The impacted fracture. A Garden I fracture may allow partial weight-bearing, and the patient may have walked after the fall; the deformity is less obvious and the diagnosis can be missed. Keep a high index of suspicion.
The occult fracture. Pain with normal initial radiographs. MRI or CT is needed to confirm, and the hip is treated as fractured until the fracture is excluded.
Differential diagnosis. The painful, non-weight-bearing hip in an older patient has several mimics. The key discriminators are fracture location (intracapsular versus extracapsular) and whether plain films are diagnostic.
- Key Distinguishing Features
- Shortened, externally rotated leg; pain on log-roll; intracapsular = AVN risk
- Investigation / Action
- AP pelvis + cross-table lateral; classify (Garden/Pauwels)
- Key Distinguishing Features
- Extracapsular, lower AVN risk; often more shortening/bruising; cephalomedullary nail or DHS
- Investigation / Action
- AP/lateral hip; distinguishes implant choice
- Key Distinguishing Features
- Below lesser trochanter; high shear/deforming forces; consider atypical (bisphosphonate) fracture
- Investigation / Action
- Full-length femur films; review bisphosphonate history
- Key Distinguishing Features
- Hip pain, unable to weight-bear, NORMAL initial X-ray
- Investigation / Action
- MRI (gold standard) or CT; treat as fracture until excluded
- Key Distinguishing Features
- Prodromal pain weeks-months, minimal trauma, lytic/sclerotic lesion, constitutional symptoms
- Investigation / Action
- Staging CT, myeloma screen, bone scan/PET, biopsy
- Key Distinguishing Features
- Groin pain, may weight-bear partially, normal proximal femur
- Investigation / Action
- AP pelvis; CT if acetabulum suspected
- Key Distinguishing Features
- Lateral tenderness, able to straight-leg raise, no shortening/rotation
- Investigation / Action
- Clinical; X-ray to exclude fracture
Investigations
Radiographs. An AP pelvis, which shows both hips for comparison and the acetabulum, and a cross-table lateral of the hip, which avoids moving the patient excessively. Look for the fracture line, displacement, the Garden stage and disruption of Shenton's line.

CT is for the fracture suspected clinically but not seen on radiographs, for complex fracture patterns, and for pre-operative planning in young patients. It delineates the fracture pattern, posterior comminution and the vertical fracture angle for Pauwels grading better than plain films.
MRI is the gold standard for the occult fracture, with sensitivity of 99% and specificity of 95%. It shows bone marrow oedema and undisplaced fracture lines invisible on radiographs, and is indicated when clinical suspicion is high but the radiograph is normal, or when the patient cannot mobilise after a fall.
Pre-operative bloods. The standard set, each with its reason:
- FBC: baseline haemoglobin (anaemia is common in the elderly and transfusion may be needed)
- U&E: renal function for anaesthesia and contrast studies
- Coagulation: INR if on warfarin, platelet count if on antiplatelet agents
- Group and save or cross-match: for the anticipated blood loss of arthroplasty
- CRP and ESR: baseline, which may be elevated by the fracture and is useful if infection is suspected later
Cardiac assessment. An ECG in every patient, because this is a high-risk population for cardiac disease; troponin if there is chest pain or ECG change, since myocardial injury is common after the fracture; an echocardiogram if there is a significant cardiac history or a murmur.
Bone health, after the acute episode.
- DEXA scan to assess bone mineral density
- Vitamin D, which is often deficient and is corrected post-operatively
- Calcium, phosphate and PTH if a secondary cause of osteoporosis is suspected
- Thyroid function, since hyperthyroidism causes bone loss
Optimisation before theatre. Cardiac clearance by the anaesthetic team, with optimisation of cardiac medication. Anticoagulation is managed actively, reversing warfarin where needed with vitamin K and prothrombin complex concentrate and holding DOACs appropriately; the detail is under Special Populations. The elderly are often dehydrated, so hypovolaemia is corrected. Pain is controlled with a fascia iliaca block, since regional anaesthesia reduces the opioid requirement.
Management
The decision. Displacement and age decide the operation. An undisplaced fracture is fixed at any age; a displaced fracture in the elderly is replaced; a displaced fracture in the young is reduced and fixed to keep the native head.
Undisplaced (Garden I-II). Internal fixation with cannulated screws, typically three in an inverted triangle. The blood supply is preserved, so fixation stabilises the fracture and lets it heal. A Garden I (impacted valgus) fracture can displace if it is not fixed, and most surgeons recommend fixation. Post-operatively the patient is protected in weight-bearing initially and watched for AVN and nonunion.
Displaced in the elderly (Garden III-IV). Arthroplasty, either hemiarthroplasty or total hip replacement. Displaced fractures carry high AVN and nonunion rates, up to 30-50%, and fixation in the elderly has poor outcomes.
Hemiarthroplasty or total hip replacement. A hemiarthroplasty, unipolar or bipolar, suits the frail patient with limited mobility, cognitive impairment or a shorter life expectancy, and carries a lower dislocation risk. A cemented THR is for the patient who is cognitively intact, independently mobile (walks outdoors) and expected to survive more than 4 years, or who has pre-existing hip arthritis; it gives better function but a higher dislocation risk, and NICE recommends it for independently mobile patients. Hold the functional claim against HEALTH, carded below: in independently ambulating patients the functional and quality-of-life gain from THR over hemiarthroplasty at 24 months was, in the trial's own words, clinically unimportant, and instability or dislocation was 4.7% against 2.4%, so THR is a reasonable choice for the active, cognitively intact patient who wants the best possible hip rather than a mandatory one.
MICTHR Indications
Hook:MIC check before THR - Mobile, Intact, Continued survival!
Displaced in the young (under 60-65). Urgent reduction and internal fixation, ideally within 6 hours, to minimise AVN. The native head is preserved at almost any cost: even an AVN risk of 30% is preferable to a THR at age 40-50 and its revision burden. Reduction is closed or open and must be anatomic; the construct is three cannulated screws or a sliding hip screw, with a DHS considered for a vertical (Pauwels III) fracture. Counsel the patient about the high risk of AVN and nonunion and the possibility of arthroplasty later.
After fixation in the young. Follow-up is prolonged, because AVN can appear 2-5 years after fixation. If fixation fails, the salvage is a valgus osteotomy for nonunion or arthroplasty for AVN. A THR in a young patient means uncemented implants and hard-on-hard bearings (ceramic-on-ceramic, highly cross-linked polyethylene), and the patient should anticipate multiple revisions over a lifetime.
Timing. Hip fracture is treated as a surgical emergency, and surgery within 36-48 hours is the service standard in most systems. State the basis for it accurately, because this page's own carded trial complicates the usual claim. HIP ATTACK randomised 2,970 patients to surgery at a median of 6 hours versus 24 hours and found no difference in 90-day mortality (9% vs 10%, HR 0.91, 95% CI 0.72-1.14) and no difference in major complications (22% vs 22%). So there is no trial evidence that accelerating surgery within an already-prompt window saves lives, and no case for emergent overnight operating.
What the target rests on. Observational data in which delay beyond 36 hours tracks with higher mortality, and that association is heavily confounded, because the patients who wait are disproportionately the ones who are too unwell to proceed. The defensible statement is that prompt surgery on a medically optimised patient shortens stay and reduces pressure sores and delirium, that unnecessary delay should be avoided, and that delay to correct a correctable medical problem is not a failure of care. Operating on an unoptimised patient to hit a clock is the wrong trade.
Surgical Techniques
Who. Undisplaced fractures (Garden I-II) at any age, and displaced fractures (Garden III-IV) in physiologically young patients under 60-65.
Set-up. Supine on a fracture table with gentle traction, with the image intensifier positioned for AP and lateral views.
Reduction, if the fracture is displaced. The Leadbetter manoeuvre: gentle longitudinal traction, flexion 15-30 degrees, abduction 20-30 degrees and internal rotation 15 degrees. If closed reduction fails, open reduction through a Watson-Jones or Smith-Petersen approach.
Judging the reduction. Quality of reduction is the factor most clearly linked to outcome, so know how to say it is acceptable:
- Garden alignment index: on the AP the principal compression trabeculae of the head should meet the medial femoral cortex at about 160 degrees; on the lateral they should be roughly colinear, about 180 degrees. An acceptable reduction sits between 155 and 180 degrees on both views. Varus (under about 155 degrees) or marked deviation predicts fixation failure and AVN. Do not confuse the index with the Pauwels fracture-line angle
- Lowell's lines: the cortical outlines of head and neck should form continuous, smooth S and reversed-S curves where a convex meets a concave. A tangency or cusp, two curves meeting at a point rather than flowing, signals malreduction
- Shenton's line restored, with no residual varus or posterior tilt
- Acceptable criteria: less than 20 degrees of angulation and less than 2 mm of displacement
Guidewires. Three wires in an inverted triangle: the inferior wire along the calcar at the inferior neck, the two superior wires divergent, all parallel in the sagittal plane and all within 5 mm of subchondral bone. Check the lateral view to avoid joint penetration.
Screws. Partially threaded cannulated screws of 6.5 or 7.3 mm diameter, with the threads crossing the fracture so that tightening compresses it. The inferior screw provides calcar support and prevents varus collapse.
Why this configuration. Three screws give the best stability, better than two and with minimal advantage from a fourth. The inverted triangle gives superior resistance to rotation, spreading load across the femoral head, with the base of the triangle anterior and the apex posterior on the lateral view; the wider the spread within the head, the better the rotational resistance. Screws that are parallel in the frontal and sagittal planes allow compression, partial threading allows interfragmentary compression, and each screw spans the fracture and engages the far cortex.
Where it is weak. Parallel screws offer no resistance to shear in a vertical (Pauwels III) fracture and have no sliding or compression mechanism of the kind a DHS has, so the construct relies entirely on the quality of the reduction and on bone healing.
Capsulotomy at the time of fixation is optional; the tamponade argument and the FAITH result are set out under Pathophysiology.
Afterwards. Touch weight-bearing initially, advanced as tolerated; the protocol is under Rehabilitation.
Biomechanics and Fixation Principles
The forces on the neck. Compression is the primary load: body weight passes from acetabulum to femoral head at 3-5 times body weight walking and up to 8 times running, concentrated on the superomedial head, the weight-bearing zone. The moment arm of body weight generates tension along the inferomedial cortex, resisted by the thick cortical bone of the calcar. Shear acts parallel to the fracture line and increases with its verticality, so a Pauwels III fracture over 50 degrees carries predominantly shear and a high failure risk. Torsion from the muscle attachments, gluteus medius and iliopsoas, rotates and displaces the fragments.
Absolute stability, or nothing. An intracapsular fracture cannot form periosteal callus, so it must heal by primary bone healing, direct osteonal remodelling, which needs no motion at the fracture site: anatomic reduction, compression and adequate fixation. Relative stability, the controlled micromotion that permits secondary healing with callus, is used for extracapsular (pertrochanteric) fractures and is not acceptable here.
Why displaced fractures fail fixation. The biology is set out under Pathophysiology: torn lateral epiphyseal vessels and AVN, synovial washout of the haematoma, no periosteum. The mechanics add to it:
- Loss of calcar continuity removes medial support and the fracture collapses into varus
- Posterior neck comminution prevents a stable reduction
- A vertical fracture angle (Pauwels III) carries high shear
- A fracture gap of even 2 mm significantly impairs healing
Poor biology and mechanical instability together are why fixation of a displaced fracture in the elderly fails at the rate given under Management, and why it is replaced rather than fixed.
Complications
The first days. The mortality figures are under Rehabilitation and Outcomes. The medical complications of the first days, and what drives each:
- Pressure sores from prolonged pre-operative immobility
- Pneumonia from aspiration, reduced mobility and general anaesthesia
- DVT and PE, a high risk in the elderly immobile patient
- UTI from catheterisation and dehydration
- Delirium in 20-50% of the elderly post-operatively (pain, medication, an unfamiliar environment)
- Myocardial infarction from surgical stress and cardiac demand
Surgical complications.
- Bleeding: arthroplasty loses 300-500 mL and may need transfusion
- Infection: superficial in 2-3%, deep periprosthetic in less than 1%
- Nerve injury: sciatic with the posterior approach (less than 1%), femoral with the anterior approach (rare)
- Vascular injury: the femoral vessels with the anterior approach, rare but catastrophic
Rehabilitation and Outcomes
Out of bed on day one. The goal is out of bed within 24 hours of surgery. There is strong evidence that early mobilisation reduces pneumonia (by 40%), pressure sores (50%), DVT and PE (30%) and delirium (20%).
Weight-bearing after arthroplasty. Weight-bearing as tolerated immediately: the implant is load-bearing, so there is no need for protection, and full weight-bearing is encouraged as pain allows, with a frame or crutches for balance initially.
Weight-bearing after fixation of an undisplaced fracture. Touch weight-bearing for 6 weeks, though some surgeons allow weight-bearing as tolerated if the fixation is stable; partial (50%) weight-bearing at 6 weeks if the radiograph shows healing; full weight-bearing at 12 weeks if united. Full weight-bearing too early risks displacement.
Weight-bearing after fixation of a displaced fracture in the young. Strict non-weight-bearing for 6-12 weeks, because the risk of fixation failure, AVN and nonunion is high, and advance very cautiously on serial radiographs.
Hip precautions after a posterior approach. For 6 weeks, and some surgeons say 12, the patient avoids:
- Hip flexion beyond 90 degrees (no low chairs or toilet seats)
- Adduction across the midline (no crossing the legs)
- Internal rotation (no twisting inward)
The risk is posterior dislocation if the precautions are violated. If the external rotators were repaired well, some surgeons omit the precautions; the evidence is mixed.
Hip precautions after an anterior or anterolateral approach. Avoid hip extension and external rotation. The precautions are less restrictive than after a posterior approach, and the dislocation risk is lower overall.
Guidelines, Registries & Global Practice
Global Burden and Projection of Hip Fracture
- Worldwide hip fractures estimated at 1.26 million in 1990 (338,000 men; 917,000 women)
- Projected to roughly double to 2.6 million by 2025 and 4.5 million by 2050 (age/sex-specific rates held constant)
- Proportional rise greater in men (about 310%) than women (about 240%)
- Asia's share of global hip fractures projected to rise from 26% (1990) to about 45% by 2050
Global Guideline Comparison
The fundamentals are consistent across major bodies; the main areas of genuine variation are the THA-vs-hemiarthroplasty threshold and the surgical-timing target.
- AAOS (USA)
- Within 24-48 h of admission (moderate evidence)
- NICE NG124 / BOAST (UK)
- On day of, or day after, admission (within 36 h)
- AO Foundation / EFORT (Europe)
- Early surgery, generally within 24-48 h
- AAOS (USA)
- Arthroplasty; THA an option for active patients
- NICE NG124 / BOAST (UK)
- THA if independently mobile, not cognitively impaired, medically fit
- AO Foundation / EFORT (Europe)
- Arthroplasty; THA for active, independent patients
- AAOS (USA)
- Cemented favoured (strong evidence)
- NICE NG124 / BOAST (UK)
- Cemented implant recommended
- AO Foundation / EFORT (Europe)
- Cemented favoured to reduce periprosthetic fracture
- AAOS (USA)
- Internal fixation
- NICE NG124 / BOAST (UK)
- Internal fixation; preserve head in younger patients
- AO Foundation / EFORT (Europe)
- Internal fixation; reduction quality emphasised
- AAOS (USA)
- Multimodal; regional/nerve blocks
- NICE NG124 / BOAST (UK)
- Offer nerve block (e.g. fascia iliaca) on presentation
- AO Foundation / EFORT (Europe)
- Multimodal incl. regional blocks
- AAOS (USA)
- Orthogeriatric co-management; treat osteoporosis
- NICE NG124 / BOAST (UK)
- Orthogeriatric/Hip Fracture Programme; FLS, bone protection
- AO Foundation / EFORT (Europe)
- Orthogeriatric pathways; secondary prevention
Registry Evidence (No PMID - Annual Registry Reports)
National Registry Evidence - Joint Replacement and Hip Fracture Registries
- AOANJRR (Australia): cemented stems and dual-mobility constructs reduce revision for fracture arthroplasty vs uncemented in older patients
- NJR (England, Wales, NI): large-volume data support cemented fixation and document dislocation as a leading early revision cause after THA for fracture
- ANZHFR / NHFD (UK): casemix-adjusted benchmarking links time-to-surgery and orthogeriatric input to lower mortality and length of stay
- Registries are observational; used for surveillance, benchmarking and implant performance rather than as randomised evidence
Epidemiology and Registry Snapshot (ANZHFR)
National and regional registries quantify the burden and benchmark outcomes. The ANZHFR (Australia and New Zealand) figures below are illustrative of the orders of magnitude seen across high-income settings:
Incidence and Burden (ANZHFR 2023):
- Approximately 20,000-22,000 hip fractures annually across the binational catchment
- Incidence rate: 150-200 per 100,000 population over age 65
- Female:male ratio approximately 3:1
- Projected to double by 2050 due to aging populations
- Annual economic burden: large by any measure — multi-billion-dollar costs in acute care, rehabilitation and long-term care in high-income health economies worldwide
Mortality (ANZHFR 2023):
- 30-day mortality: 6.2% (national average)
- 1-year mortality: 27.3%
- 5-year mortality: approximately 50%
- Casemix-adjusted variation between centres: top-quartile hospitals achieve 5-6% 30-day mortality, while rural/remote and under-resourced sites report 8-10%
Geographic and Equity Variation (global theme):
- Rural/remote vs urban: rural areas show slightly higher mortality and longer time to surgery (access and transfer logistics)
- Under-served and Indigenous populations: Registry data show hip fractures occur 10-15 years earlier in Indigenous populations, with higher mortality — a disparity mirrored in other under-served groups worldwide
- Remote communities: significant challenges with evacuation, surgical access and rehabilitation
Australian Hip Fracture Registry (ANZHFR)
The ANZHFR is a binational registry (Australia and New Zealand) established in 2015 to monitor and improve hip fracture care quality.
Key Performance Indicators (2023 National Report):
- Surgery within 48 hours: 82.4% nationally (target greater than 85%)
- Best performers: 90-95% (major metropolitan hospitals)
- Challenges: Rural hospitals, weekend admissions, medically complex patients
- Orthogeriatric assessment within 72 hours: 76.8% (target greater than 80%)
- Bone protection prescribed at discharge: 68.5% (target greater than 80%)
- Post-discharge fracture liaison service contact: 45.2% (target greater than 70% - needs improvement)
Registry Data Insights:
- Hospitals with dedicated hip fracture pathways have 30-40% lower mortality
- Orthogeriatric co-management associated with 2-day reduction in length of stay
- Weekend admissions have 15% longer time to surgery (resource constraints)
Special Populations
The problem. The patient with dementia cannot comply with hip precautions, so the dislocation risk with a THR is high, and cannot comply with protected weight-bearing, so fixation of a displaced fracture is unsuitable. Functional outcomes are poor, many never mobilise post-operatively, and mortality is 50% at one year.
The approach. A hemiarthroplasty is preferred over a THR for its lower dislocation risk, simpler surgery and shorter operative time, with a constrained liner or dual mobility considered if the dislocation risk is very high. Analgesia is optimised with a regional block (fascia iliaca) and multimodal analgesia, avoiding opioids where possible because they increase delirium. Delirium is prevented aggressively, with a familiar environment, no restraints, minimal medication, treated pain and encouraged mobilisation, and the geriatrician is involved early. Goals are realistic: the patient may not return to pre-fracture mobility, and the focus is comfort and basic transfers.
MCQ Practice Points
Q: What distinguishes Garden I from Garden II femoral neck fractures?
A: Garden I is an incomplete/impacted fracture with the head tilted into valgus. Garden II is complete but non-displaced with trabecular alignment preserved. Both are undisplaced and treated with screw fixation. Key difference: Garden I has visible trabecular lines crossing fracture; Garden II has complete fracture line but no displacement.
Q: Why do displaced femoral neck fractures have high AVN rates?
A: The medial femoral circumflex artery (MFCA) gives rise to lateral epiphyseal vessels which provide the main blood supply to the femoral head. These vessels run along the posterior-superior femoral neck. Displaced fractures disrupt this supply, causing AVN rates of 20-30% in young patients and even higher in delayed treatment. This is why displaced fractures in elderly warrant arthroplasty rather than fixation.
Q: What are the indications for THR over hemiarthroplasty in displaced NOF fractures?
A: THR is indicated for patients who are: cognitively intact, independently mobile (walking outdoors), and have life expectancy greater than 4 years. THR provides better function and lower reoperation rates. Hemiarthroplasty is preferred for: cognitive impairment, limited mobility, frail patients. NICE guidelines recommend THR for independently mobile patients.
Q: How do you manage a displaced NOF fracture in a 45-year-old?
A: Urgent reduction and internal fixation (ideally within 6 hours) to minimize AVN risk. Use closed or open reduction with 3 cannulated screws (inverted triangle) or sliding hip screw. Despite high complication rates (AVN 20-30%, nonunion 10-20%), preserving the native hip is preferred over arthroplasty in young patients. Counsel patient about potential need for future arthroplasty.
Exam Viva Scenarios
Practise clinical reasoning and management decisions out loud
“An 80-year-old woman with osteoporosis falls at home and has a Garden IV neck of femur fracture. She was previously independently mobile and cognitively intact. How do you manage?”
“A 78-year-old woman presents to the emergency department at 10pm with a Garden III displaced neck of femur fracture following a fall at home. She takes warfarin for atrial fibrillation. Her INR is 3.2 (therapeutic range 2.0-3.0 for AF). She is otherwise stable with no other injuries. The anaesthetist asks whether you want to reverse her warfarin and operate tonight, or wait for the INR to normalize naturally over the next 48 hours. What factors do you consider and how do you proceed?”
“A 42-year-old man is brought to your trauma center following a high-speed motorcycle accident. He has multiple injuries including a closed head injury (GCS 14), liver laceration being managed conservatively, and a Garden IV displaced neck of femur fracture. It is now 8 hours since injury. He is hemodynamically stable. X-rays show the femoral neck fracture is very vertical (Pauwels Type III) with significant posterior comminution. What are your management options and what factors influence your decision?”
“A 68-year-old woman presents with a Garden III femoral neck fracture after a minor fall in the bathroom. On questioning, she mentions she has had worsening hip pain for the past 3 months that was present even before the fall. She attributes it to 'arthritis'. On examination, the fracture site is more tender than expected, and she appears cachectic. X-ray shows a displaced femoral neck fracture with some lytic changes in the intertrochanteric region. How do you approach this case?”
“A 72-year-old man with Parkinson's disease has a fall at home and is brought to the emergency department. X-ray pelvis shows displaced femoral neck fractures bilaterally (Garden IV both sides). He is medically stable but in significant pain. The emergency department physician asks you urgently whether this is possible or if there's been a mix-up with the X-rays. How do you approach this rare presentation and what are your management priorities?”
Garden Classification
- I/II: Undisplaced → SCREWS
- III/IV: Displaced → ARTHROPLASTY (elderly)
Arthroplasty Choice
- THR: Mobile, cognitively intact, good life expectancy
- Hemiarthroplasty: Frail, impaired cognition, limited mobility
Key Points
- Surgery within 36-48 hours
- MFCA supply disrupted → AVN risk
- Young displaced: Attempt fixation urgently
Post-Op Care
- DVT prophylaxis
- Early mobilization
- Bone protection (bisphosphonates)
- Falls assessment
Evidence Base
THR vs Hemiarthroplasty - HEALTH Trial
- 1495 independently ambulating patients, age 50+, displaced femoral neck fracture, 80 centres in 10 countries
- Secondary hip procedure within 24 months: 7.9% THA vs 8.3% hemiarthroplasty (HR 0.95, not significant)
- Hip instability or dislocation: 4.7% THA vs 2.4% hemiarthroplasty
- Mortality similar (14.3% THA vs 13.1% hemiarthroplasty, P=0.48); the trial's own wording is that THA gave a CLINICALLY UNIMPORTANT improvement in function and quality of life over 24 months
- Serious adverse events occurred in 300 (41.8%) after THA and 265 (36.7%) after hemiarthroplasty
Timing of Surgery - HIP ATTACK Trial
- 2970 patients, 69 hospitals, 17 countries; accelerated surgery (median 6 h) vs standard care (median 24 h)
- No difference in 90-day mortality: 9% vs 10% (HR 0.91, 95% CI 0.72 to 1.14)
- No difference in composite of major complications: 22% vs 22% (HR 0.97)
- Accelerated surgery did not significantly lower mortality or major complications
Timing of Surgery - Observational Meta-analysis (Klestil, 2018)
- 28 prospective studies, 31,242 patients aged 60 or older - and note that although randomised trials were eligible, every included study was OBSERVATIONAL
- Surgery within 48 hours: 20% lower risk of death within 12 months (RR 0.80, 95% CI 0.66-0.97)
- Surgery within 24 hours versus after: NO statistically significant mortality difference (RR 0.82, 95% CI 0.67-1.01)
- Adjusted data showed fewer complications with early surgery (8% vs 17%) and rising pressure-ulcer risk with increasing delay
Cemented vs Uncemented Hemiarthroplasty - WHiTE 5 Cost-Utility Analysis
- UK multicentre RCT of cemented vs hydroxyapatite-coated uncemented hemiarthroplasty for displaced intracapsular hip fracture
- Cemented implants were cost-saving (mean difference about minus 961 pounds)
- Cemented implants increased quality-adjusted life years (mean QALY difference 0.010)
- Cemented hemiarthroplasty 95 to 97% probability of being cost-effective
Internal Fixation Method - FAITH Trial
- 1108 patients age 50+ with low-energy hip fracture; sliding hip screw vs multiple cancellous screws
- 24-month reoperation: 20% (SHS) vs 22% (screws) - no significant difference (HR 0.83)
- Avascular necrosis MORE common with sliding hip screw: 9% vs 5% (HR 1.91)
- Subgroups (smokers, displaced or base-of-neck fractures) may do better with a sliding hip screw
