Girdlestone Procedure (Excision Arthroplasty)

The Girdlestone procedure, or excision arthroplasty, resects the femoral head and neck and puts nothing back. It is now used mainly as a salvage procedure for periprosthetic joint infection (PJI) when other options have failed or are contraindicated.

History. G.R. Girdlestone (1881-1950) described excision arthroplasty in 1943 for tuberculosis of the hip. In the pre-antibiotic era it was the primary treatment for the septic hip. It is now reserved for salvage of the infected arthroplasty, or for the patient who is not a candidate for replacement.

The indication today. The core indication is PJI where DAIR has failed or is contraindicated. The patient must still be medically fit for surgery and able to tolerate prolonged antibiotics. The resection may be definitive, a permanent hip, or interim, clearing the infection before reimplantation.

Failed DAIR. Failure of DAIR (debridement, antibiotics and implant retention) is an indication in its own right. It means persistent positive cultures, a persisting sinus tract, ongoing clinical infection, or CRP and ESR that fail to normalise despite an adequate course.

Fungal infection. Fungi sit among the difficult organisms above, but the fungal systematic review in the evidence below found recurrence in 53.1% after resection arthroplasty against 47.7% after two-stage exchange. In fungal PJI a Girdlestone is not the higher-certainty option.

Beyond PJI. The rare indications are:

• Native hip septic arthritis after failed treatment
• Tuberculosis, historically, and rarely now
• Severe avascular necrosis not suitable for THR
• Certain malignancies, as palliation

When to consider alternatives. Consider another route first in these situations:

No absolute contraindication. The Girdlestone is often the last resort, and there is no absolute contraindication to it as such. The relative ones are a patient unwilling to accept the disability and a patient medically unstable for any surgery; for the severely ill, palliative suppression is sometimes the only option. Discuss functional expectations thoroughly and weigh the patient's overall goals before deciding.

What comes out and what stays. Everything infected is removed, and the structures that will stabilise the hip afterwards are kept wherever they are viable.

The pseudarthrosis. After resection a fibrous pseudarthrosis develops. The proximal femoral stump migrates superiorly until it articulates with the ischium or ilium, the acetabulum is defunctioned, and weight is borne through the ischium. Excessive bone resection allows more proximal migration and gives worse function. Shortening of 4-6 cm is typical.

The abductors. Gluteus medius and minimus insert on the greater trochanter, and this abductor mechanism must be preserved for the pseudarthrosis to be stable. Abductor function determines stability: if it is preserved, the patient may achieve limited ambulation.

View larger

Type of infection. PJI is classified by timing and onset, which helps predict treatment success and guides surgical planning.

DAIR or excision. DAIR succeeds with an acute infection (under 4 weeks), a sensitive organism, a well-fixed prosthesis and a healthy host. The Girdlestone indications are failed DAIR, chronic infection, resistant organisms and poor host status.

Type of Girdlestone. The resection is classified by its intent, which helps determine the approach and how the patient is counselled.

History. Establish the infection, its course and the host:

• Presenting symptoms of pain, swelling and sinus drainage, and whether they are acute or chronic
• Previous surgery, including the number of revisions and the organisms grown
• Antibiotic history: previous treatment and allergies
• Comorbidities such as diabetes and immunosuppression

Examination. Inspect the wound for a sinus tract, drainage and erythema, and assess the joint for warmth, effusion and range of movement. Record sciatic nerve function, the current limb-length discrepancy, and gait and ambulatory status.

Blood tests. CRP is the most sensitive marker, with the ESR alongside it, and serial values monitor the response to treatment. The white cell count is often normal in chronic infection. Take blood cultures if there are systemic symptoms, and albumin, prealbumin and HbA1c to assess nutrition and glycaemic control.

Aspiration. Synovial fluid is sent for white cell count (threshold over 3000/μL), PMN percentage (over 80%) and culture, with alpha-defensin if available. Cultures are held for 14 days, which improves detection of slow-growing organisms such as Propionibacterium acnes, and samples are taken before antibiotics.

Imaging. Radiographs show component loosening, bone loss and periosteal reaction. CT assesses bone defects and the location of cement. MRI is rarely needed, to look for a soft-tissue collection, and nuclear medicine is kept for the case where the diagnosis remains uncertain.

The diagnosis. The MSIS criteria make it: one major criterion, or four of the six minor criteria.

A painful or failing hip arthroplasty is not always infected. Distinguishing PJI from aseptic causes is the pivotal decision before resection — an unnecessary Girdlestone for aseptic loosening is a catastrophic error.

• Resection level is the single biggest surgeon-controlled variable. The femur is resected only as far distal as infection/removal demands - conserving length keeps the femoral stump abutting the ischium/ilium rather than riding high. A high resection (short stump) allows greater proximal migration, worsens the lever arm and produces a longer, more unstable limb-shortening; a more distal, length-preserving resection (consistent with clearing the infection) gives a shorter migration and a more stable articulation. The 4-6 cm shortening is the price; making it worse than necessary is avoidable.
• The abductor lever arm is what powers gait. A pseudarthrosis is stabilised dynamically, not by bone - so an intact abductor complex on a preserved greater trochanter lets the patient generate a functional (if Trendelenburg) gait, whereas a detached/non-viable trochanter or a deficient abductor mass produces a flail hip, a lurching or non-functional gait and aid-dependence or wheelchair use. Preserving the trochanter and its attachment whenever it is viable is therefore a functional imperative, not a cosmetic one.
• Soft-tissue interposition and scar create the fibrous articulation. A broad, well-vascularised soft-tissue envelope between the proximal femur and pelvis (capsule remnants, vastus/glutei, sometimes a formal muscle interposition) matures into a stable fibrous pseudarthrosis and cushions the bone-on-bone contact that causes pain; extensive dead space or a bare femoral stump against pelvis gives instability and pain.
• Host and disease factors set the ceiling. Even with perfect technique, bilateral disease, poor abductors from prior surgery, neuromuscular disease, uncontrolled pain and low functional reserve cap the achievable result - which is why the honest counselling statement is that every patient in these series walked with aids, and many barely walked at all.

• When it is done, and the prerequisites. Conversion is considered once the infection is truly eradicated (normalised inflammatory markers, negative aspirate/intra-operative samples, an interval off antibiotics) in a host now fit and motivated for a major reconstruction. It is the "second stage" that a spacer was meant to facilitate - and the longer the hip has been left as a bare Girdlestone (no spacer), the harder it becomes.
• The soft tissues are the enemy. After months of a shortened, migrated hip the abductors and capsule contract and scar, the femur sits proximal, and restoring length re-tensions a chronically shortened neurovascular envelope - so conversion carries a real risk of sciatic/femoral nerve traction injury when length is regained, and full LLD correction is often not achievable in one sitting. Heterotopic bone and dense scar further obscure planes.
• Bone stock and fixation are compromised. Repeated debridements and the original resection leave deficient femoral and acetabular bone, frequently needing revision/reconstruction implants (long stems, cup-cage/augments, sometimes a proximal femoral replacement) rather than a primary THA.
• Abductor deficiency drives instability - the dominant complication. The scarred, shortened, often-deficient abductor mechanism makes the converted hip prone to dislocation, so constrained or dual-mobility bearings are frequently used; overall complication, dislocation and re-infection rates are markedly higher, and functional outcomes poorer, than a primary THA - though still usually better than a retained Girdlestone in a suitable host.

The sequence. Confirm PJI by the MSIS criteria and identify the organism if possible. Then assess the patient: medical fitness for surgery, bone stock, social support and compliance, and functional expectations. Multidisciplinary discussion with infectious diseases, microbiology and the patient is essential.

Algorithm

View larger

The rung above resection. The table names two-stage revision as the reconstructive route. The INFORM trial in the evidence below found that, in suitable hosts, single-stage revision matched two-stage for medium-term outcome, so resection is reserved for patients unfit for, or failed by, reconstruction.

Spacer or no spacer. If reimplantation is planned, an articulating antibiotic-loaded spacer is preferred. It maintains soft-tissue tension, which a Girdlestone loses, and gives better function while the infection is treated. A true Girdlestone is for definitive cases or when a stable spacer cannot be achieved, and a static spacer is the other definitive option. The spacer adds cost where the Girdlestone is simpler, and both are effective if the debridement is adequate.

Confirm the infection by the MSIS criteria, as set out under Investigations.

Identify the organism. The pathogen determines antibiotic selection and duration. In refractory cases consider resistant organisms, fungi or mycobacteria, and liaise with microbiology and infectious diseases early.

Optimise the patient. Nutrition (albumin over 3 g/dL, total lymphocyte count), glycaemic control (HbA1c under 8%) and smoking cessation, with cardiac and renal comorbidities addressed.

Plan the operation. Review previous incisions and the bone loss on imaging, and consider whether a spacer or a definitive resection is intended. Discuss realistic functional expectations with the patient, including the need for walking aids and the leg-length discrepancy.

Persistent infection. 10-30% may have persistent infection despite adequate surgery. The risk factors are resistant organisms, inadequate debridement, retained cement or foreign material, and immunocompromise.

When infection persists. Work through the causes before choosing between repeat debridement, long-term suppression and accepting a chronic wound:

1. Confirm that the initial debridement was adequate
2. Review antibiotic sensitivities and compliance
3. Consider repeat surgery for retained material
4. Long-term oral antibiotic suppression may control the infection but not eradicate it

Mobilisation. Weight bearing is as tolerated with aids, a walker or crutches. The programme includes hip abduction strengthening, gait training for the leg-length discrepancy and a shoe raise of 3-5 cm, and patients progress to a single stick over months.

Antibiotics. A minimum of 6 weeks of IV antibiotics is followed by oral therapy. The duration depends on the organism, the response and whether reimplantation is planned, and the regimen is set with infectious diseases. Monitor for drug toxicity: renal and ototoxicity with aminoglycosides, hepatic toxicity with rifampicin.

Monitoring. CRP weekly at first, then monthly. Assess the wound, function and pain at each visit, watch for antibiotic side effects, and image if retained material is suspected.

What the patient gains. Pain relief is usually achieved, though only a third of Stoklas's patients were fully pain free, and sitting tolerance is usually preserved. The result is better than ongoing sepsis or repeated failed surgery, and substantially worse than a successful reimplantation.

What it costs. Expect significant disability. Mean shortening was 5.7 cm in Sabelle's refractory-PJI series and 3.8 cm in Stoklas's mixed-indication series, and every patient in Stoklas's series had a positive Trendelenburg sign and used walking aids. The patient should expect a marked limp, permanent use of a stick or frame, and severe restriction of activity. The mean walking distance of 175 m in Sabelle's series carries a standard deviation of 350, meaning many walk almost nothing.

Good and poor results. A good result is pain control, an eradicated infection, independence in activities of daily living, walking with aids, and a patient satisfied with the outcome given the alternatives. A poor one is persistent infection, a chronic wound or sinus, a non-ambulatory patient, intractable pain and a poor quality of life.

Hip PJI affects roughly 1-2% of primary and up to 4-5% of revision total hip replacements worldwide. As arthroplasty volumes rise, the absolute burden of PJI is growing, and resection arthroplasty persists as the salvage of last resort across all health systems.