Debridement, Antibiotics and Implant Retention for acute PJI of the hip or knee · advanced
- DAIR is indicated only for acute post-operative PJI (symptoms less than 3 weeks after the index arthroplasty) or acute haematogenous PJI (symptoms less than 3 to 4 weeks) with a well-fixed implant, a susceptible organism on culture or Gram stain, and an intact soft-tissue envelope without a sinus tract.
- The single most important technical step is modular component exchange (liner and femoral head, or tibial insert) — this opens the posterior recesses and gutters that cannot be reached any other way; leaving the original liner in situ is the most common technical error and guarantees incomplete debridement.
- Obtain a minimum of five deep tissue cultures (synovium, capsule, bone-implant interface, liner membrane, fluid) BEFORE any antibiotic is given; each sample is taken with a fresh instrument and sent for aerobic, anaerobic, fungal and mycobacterial culture, plus sonication of the explanted liner when available.
- Success is organism-dependent: streptococcal and coagulase-negative staphylococcal infections achieve 70 to 85 percent implant retention at 2 years, while MRSA and polymicrobial infections drop to 40 to 55 percent. A draining sinus tract reduces success by approximately 30 percent regardless of organism.
- The three strongest independent predictors of failure are MRSA, a sinus tract, and symptom duration greater than 3 weeks; when two or more are present the probability of retention falls below 50 percent and two-stage revision should be considered from the outset.
- Antibiotic duration after DAIR is 6 weeks of intravenous therapy followed by 3 to 6 months of highly bioavailable oral agents (or indefinite suppression in selected cases); biofilm-active agents — rifampicin for staphylococci, a fluoroquinolone for Gram-negatives — are added once the organism is known.
When & Why
What DAIR is. Debridement, Antibiotics and Implant Retention is a single-stage salvage operation for an acute periprosthetic joint infection in which the implant is still well fixed. It buys implant retention by surgically removing the planktonic and early-biofilm burden, exchanging the modular parts that harbour biofilm, and then suppressing residual organisms with targeted biofilm-active antibiotics. Done well, on the right patient, it avoids the morbidity of a two-stage revision; done on the wrong patient it simply delays it and burns the chance of retention. The indication, in one sentence. An acute PJI — either acute post-operative (symptom onset less than 3 weeks after the index arthroplasty) or acute haematogenous (symptoms less than 3 to 4 weeks) — in a patient with a well-fixed implant, a susceptible organism, an intact soft-tissue envelope, and no sinus tract. Absolute indications (all should be present):
- Acute post-operative PJI with symptom onset less than 3 weeks after the index arthroplasty.
- Acute haematogenous PJI with symptom duration less than 3 to 4 weeks.
- Well-fixed implant on preoperative radiographs and confirmed stable intraoperatively.
- Susceptible organism on preoperative aspiration or Gram stain (streptococcus, coagulase-negative staphylococcus, or a known sensitive organism).
- Intact soft-tissue envelope without a sinus tract or severe scarring. Relative indications — early post-operative PJI (less than 6 weeks from the index) with borderline symptom duration but a favourable organism and host; a patient with significant medical comorbidity in whom two-stage revision carries prohibitive risk; or a single early debridement performed elsewhere with a retained implant, now referred for definitive management. Who should NOT have a DAIR — the contraindications that push the decision to a two-stage revision:
- Strength
- Absolute
- Strength
- Absolute
- Strength
- Absolute
- Strength
- Absolute
- Strength
- Absolute
- Strength
- Relative
- Strength
- Relative
- Strength
- Relative
The rule. Acute post-operative PJI less than 3 weeks of symptoms; acute haematogenous PJI less than 3 to 4 weeks. Beyond these windows the biofilm is mature and DAIR success falls sharply. The evidence. Multiple series show success greater than 70 percent when symptoms are less than 3 weeks versus less than 40 percent when greater than 6 weeks. The 3-week cut-off is the most widely accepted threshold for offering DAIR.
The principle. The prosthesis must be mechanically stable at the bone-implant interface. Any radiographic or intraoperative loosening converts the procedure to a one-stage or two-stage revision. The test. Intraoperative assessment by attempting to move the femoral or tibial component with an instrument. If the implant moves or toggles, DAIR is abandoned and the case converts to staged revision.
The fact. A draining sinus tract indicates chronic infection with a mature biofilm and external communication; DAIR success drops to approximately 30 to 40 percent. The management. Any patient presenting with a sinus tract should be counselled for two-stage revision from the first consultation; attempting DAIR exposes the patient to an additional operation with a low chance of success.
The hierarchy. Streptococci and coagulase-negative staphylococci achieve the highest DAIR success (70 to 85 percent). MRSA, enterococci and polymicrobial infections have markedly lower retention rates (40 to 55 percent). The implication. When preoperative aspiration or Gram stain suggests MRSA or a resistant Gram-negative, many surgeons proceed directly to two-stage revision rather than attempting DAIR, especially if other risk factors coexist.
The requirement. The skin and subcutaneous tissues must be healthy enough to achieve primary wound closure over the debrided joint without tension. Previous multiple incisions, radiation, or severe scarring may preclude DAIR. The assessment. If the planned incision would require flap coverage or the soft tissues are severely compromised, two-stage revision with soft-tissue optimisation is the safer plan.
The staging. McPherson host grade (A, B, C) and local extremity grade (1, 2, 3) strongly influence decision-making. Grade C hosts (immunocompromised, dialysis, active malignancy) have DAIR failure rates approaching 60 to 70 percent. The practical rule. In a McPherson C host with additional risk factors, proceed directly to two-stage revision; the incremental morbidity of a failed DAIR followed by later revision outweighs the benefit of attempting retention.
Symptom duration greater than 3 weeks for post-operative PJI, or greater than 4 weeks for haematogenous PJI, is the threshold at which most surgeons abandon DAIR in favour of two-stage revision. The exact cut-off varies with organism virulence and host factors, but the biofilm matures across this window and retention success collapses with every additional week of symptoms.
Consent. Counsel the patient that even with optimal selection and technique DAIR fails in roughly 25 to 45 percent of cases and that a two-stage revision may still be required; that prolonged intravenous and oral antibiotics carry renal, hepatic, gastrointestinal and neurological side effects; that the operated joint may be stiffer; and (for the hip) that dislocation risk after a posterior-approach DAIR is higher than after a primary.
The Operation
The goal is a radical, open debridement of the entire joint through the index approach, exchange of every modular part that can harbour biofilm, copious irrigation, confirmation that the implant is solid, and primary wound closure — all cultures taken before a single dose of antibiotic is given. The exposure is laid out in full below (and in depth on the posterior approach to the hip and medial parapatellar approach to the knee pages).

Operative sequence
- Supine on a radiolucent table for both hip and knee. For the hip, a bump under the ipsilateral sacrum (or lateral decubitus for a posterior approach) gives access; for the knee, a sandbag under the ipsilateral buttock and a footrest let the knee flex to 90 to 120 degrees.
- Regional anaesthesia (spinal or epidural) plus sedation is preferred for muscle relaxation and post-operative pain; general anaesthesia when regional is contraindicated.
- Knee DAIR: thigh tourniquet inflated to 300 mmHg after exsanguination to reduce blood loss and improve visualisation during synovectomy. Hip DAIR: no tourniquet.
- No prophylactic antibiotic is given until every deep culture is in a container — the single most critical point of the procedure. Intraoperative fluoroscopy is available but rarely needed unless implant fixation is in doubt.
- Hip (posterior, Moore-Southern). Lateral decubitus. Use the previous posterior incision, extending proximally along gluteus maximus and distally along the femur as needed. Split gluteus maximus bluntly in its fibres. This gives wide access to the acetabulum and posterior femoral neck for debridement.
- Knee (medial parapatellar). Use the previous midline incision, extending proximally and distally as needed. Raise full-thickness medial and lateral flaps to avoid skin necrosis over a multiply-operated knee.
- Use the same approach as the original arthroplasty — you need a wide window for a radical debridement, not a limited exposure.
- Hip. Identify and protect the sciatic nerve throughout (retracted and guarded). Divide the posterior capsule and tag its flaps for later repair; release and tag the short external rotators for repair at closure. Respect the inferior gluteal nerve to gluteus maximus and the medial femoral circumflex artery. Dislocate the hip — the anterior and inferior recesses can only be debrided once the head is out.
- Knee. Perform a medial parapatellar arthrotomy. Excise the infrapatellar fat pad to visualise the anterior tibia-implant interface. This opens the entire suprapatellar pouch, the medial and lateral gutters, and — once the tibial insert is removed and the knee hyperflexed with posterior retraction — the posterior compartments.
- The posterior recesses of both joints are the most common sites of residual infection after an inadequate DAIR, precisely because they cannot be seen or reached without removing the modular components.
- Inspect the joint for purulence, synovitis and loosening. Obtain the first set of cultures immediately: synovial fluid for cell count, Gram stain, culture and sensitivity; and tissue from the suprapatellar pouch, medial gutter and lateral gutter.
- Take five or more deep tissue cultures (synovium, capsule, bone-implant interface, liner membrane, fluid) using a fresh instrument for each sample, each from a different anatomic site. Send for aerobic, anaerobic, fungal and mycobacterial culture, plus sonication of the explanted liner when available.
- No antibiotic until the fifth sample is in the container. Even a positive Gram stain does not change this — the organism on Gram stain may not be the dominant pathogen.
- Knee. Remove the tibial insert (and the patellar button if present) using the manufacturer-specific extraction tools. With the insert out, the posterior compartments become accessible.
- Hip. Remove the femoral head and the acetabular liner.
- This step is non-negotiable — it is the whole point of the modular exchange and the reason the posterior recesses can be cleared.
- Perform a complete synovectomy with rongeurs, curettes and sharp dissection. Excise all inflamed synovium from the suprapatellar pouch, the medial and lateral gutters, and the posterior capsule (now accessible).
- Curette any membrane off the bone-implant interface and send additional tissue samples from there and from the liner membrane — these are high-yield sites for biofilm organisms.
- For the hip, clear the inferior acetabular recess and the posterior femoral recess specifically — the two most commonly missed areas.
- Irrigate with a minimum of 9 litres of pulsed lavage saline.
- Adjunctive dilute betadine (0.35 percent) or hydrogen peroxide soaks for 3 to 5 minutes each, followed by further saline lavage, are used by many surgeons.
- Attempt to move the femoral and tibial (or acetabular and femoral stem) components with an instrument. Any toggle or movement indicates loosening.
- If the implant is loose, abandon DAIR and convert to a one-stage or two-stage revision depending on bone loss and soft tissues — retention of a loose implant in the face of infection will fail.
- Irrigate the modular taper junctions thoroughly.
- Knee. Insert a new polyethylene tibial insert (and a new patellar button if the original was removed). Use the same thickness, or one size thicker if the joint was lax after debridement. Confirm stability and range of motion.
- Hip. Insert a new head and liner, reduce the hip, and confirm stability and leg length.
- Hip. Repair the posterior capsule and short external rotators to the greater trochanter (this restores the soft-tissue sleeve and reduces post-DAIR dislocation). Close in layers. Apply a compressive dressing.
- Knee. Close the arthrotomy with absorbable suture, then subcutaneous tissue and skin in layers. Apply a compressive dressing and a knee immobiliser. No drain is routinely used.
- Administering antibiotics before all cultures are obtained — the most common and most consequential error in DAIR.
- Taking fewer than five samples, or reusing one instrument for several sites — risks false-negative cultures.
- Failing to record the exact anatomic site of each culture — makes later interpretation difficult when only some samples grow.
- Leaving the original liner in situ — the posterior recesses cannot be accessed or debrided.
- Inadequate irrigation volume — less than 6 litres is associated with higher failure rates in registry data.
- Failure to repair the posterior capsule after a hip DAIR — increases dislocation risk.
- Sciatic nerve injury during retraction — always identify and protect the nerve throughout.
Label each culture tube with the exact anatomic site and the order in which it was taken. Never administer antibiotics until the fifth sample is in the container. If the Gram stain shows organisms, still wait for the full set of cultures — the organism on Gram stain may not be the dominant pathogen.
After the head and liner are out, spend at least 15 minutes on synovectomy alone, with a headlamp and loupes. Clear every recess — the inferior acetabular recess and the posterior femoral recess are the two most commonly missed areas. Send a final set of cultures from these posterior recesses after you think you have finished debriding.
Aftercare & Complications
Immediate post-operative protocol (day 0 to 3). Intravenous antibiotics commence after the final culture is obtained in theatre — the initial regimen is vancomycin plus a third-generation cephalosporin or piperacillin-tazobactam, pending culture results. A drain, if used, is removed at 24 to 48 hours. Mobilise with full weight-bearing as tolerated from day 1 (knee) or day 2 (hip). Review the wound at 48 hours and apply a negative-pressure dressing if drainage persists. Week 1 to 6. Tailor antibiotics to the culture results once available (usually at 48 to 72 hours). Add biofilm-active agents: rifampicin 300 mg twice daily for staphylococci (after confirming susceptibility and once the wound is dry); ciprofloxacin or levofloxacin for Gram-negative organisms. Total intravenous duration is 6 weeks, followed by an oral step-down for 3 to 6 months (or indefinite suppression in selected high-risk cases). Weekly wound review; CRP and ESR at 2, 4 and 6 weeks. Month 3 to 6. Continue oral antibiotics until the CRP normalises and remains normal for 4 weeks off therapy (in patients planned for a finite course). Clinical and radiographic review at 3 and 6 months. If infection recurs, proceed to two-stage revision.
- Intravenous (6 weeks)
- Flucloxacillin or cefazolin
- Oral step-down (3-6 months)
- Trimethoprim-sulfamethoxazole or doxycycline
- Biofilm agent
- Rifampicin
- Intravenous (6 weeks)
- Vancomycin (or daptomycin)
- Oral step-down (3-6 months)
- Linezolid or trimethoprim-sulfamethoxazole
- Biofilm agent
- Rifampicin
- Intravenous (6 weeks)
- Penicillin or ceftriaxone
- Oral step-down (3-6 months)
- Amoxicillin
- Biofilm agent
- None routinely
- Intravenous (6 weeks)
- Ceftriaxone or piperacillin-tazobactam
- Oral step-down (3-6 months)
- Ciprofloxacin or levofloxacin
- Biofilm agent
- Fluoroquinolone
- Intravenous (6 weeks)
- Vancomycin plus ceftriaxone
- Oral step-down (3-6 months)
- Trimethoprim-sulfamethoxazole plus ciprofloxacin
- Biofilm agent
- Rifampicin plus fluoroquinolone
- Recognition
- Recurrent pain, rising CRP, positive cultures, wound drainage
- Prevention
- Strict selection; complete modular exchange; adequate irrigation
- Management
- Proceed to two-stage revision; a second DAIR succeeds in less than 30 percent
- Recognition
- Dehiscence, prolonged drainage, superficial infection
- Prevention
- Meticulous layered closure; negative-pressure therapy in high-risk wounds
- Management
- Superficial: local care and antibiotics. Deep: return to theatre, consider staged revision
- Recognition
- Acute pain, shortening, deformity
- Prevention
- Meticulous posterior capsule and rotator repair; 36 mm or larger head
- Management
- Closed reduction, abduction brace; revision if recurrent
- Recognition
- New pain, progressive radiolucencies
- Prevention
- Avoid iatrogenic interface damage during debridement
- Management
- Revision arthroplasty for loosening
- Recognition
- Rising creatinine or LFTs, diarrhoea, peripheral neuropathy
- Prevention
- Monitor renal and hepatic function; rational duration
- Management
- Adjust or switch agent; treat C. difficile; review risk-benefit of suppression
Intraoperative triggers — implant loosening; severe bone loss precluding stable reimplantation of modular components; inability to achieve primary soft-tissue closure; or an unexpected finding of a chronic-appearing membrane or sinus tract not appreciated preoperatively. Post-operative triggers — persistent CRP elevation beyond 6 weeks despite appropriate antibiotics; recurrent wound drainage after 3 weeks; or positive cultures from more than two sites growing a resistant organism not covered by the current regimen.
Surveillance. CRP and ESR at 2, 4 and 6 weeks, then at 3, 6 and 12 months, then annually to 5 years — late recurrence is possible. Any rise in markers or return of symptoms prompts repeat aspiration and consideration of two-stage revision. Weight-bearing is as tolerated from day 1 to 2, with no routine bracing.
Viva & Exam Focus
DAIRDAIR — indications for implant retention
TECHTECH — the four non-negotiable operative steps
Clinical Decision Scenarios
Practise clinical reasoning and management decisions out loud
“A 68-year-old man presents 18 days after primary total knee arthroplasty with increasing pain, swelling and warmth. Aspiration shows 45,000 white cells with 92 percent neutrophils and Gram-positive cocci in clusters on stain. He has no sinus tract and the implant appears well-fixed on radiographs. How do you proceed?”
“You are performing a DAIR on a 72-year-old woman with acute haematogenous PJI of the hip 11 days after symptom onset. Intraoperatively you find a well-fixed stem and cup but discover a small sinus tract communicating with the greater trochanter that was not appreciated on preoperative examination. What do you do?”
“A 65-year-old diabetic man undergoes DAIR for acute post-operative knee PJI. Cultures grow methicillin-sensitive Staphylococcus aureus. At 8 weeks the CRP is normal and the wound is healed. How long do you continue antibiotics and what is your surveillance plan?”
Indications — the four pillars
- Symptom duration less than 3 weeks (post-operative) or less than 4 weeks (haematogenous)
- Well-fixed implant on imaging and confirmed stable intraoperatively
- No sinus tract and an intact soft-tissue envelope allowing primary closure
- A susceptible organism (streptococcus, coagulase-negative staphylococcus preferred; MRSA acceptable if no other risk factors)
Contraindications — when to choose two-stage
- Symptom duration greater than 3 to 4 weeks — biofilm maturation drops success below 50 percent
- Sinus tract present — success falls to 35 to 40 percent; two-stage is standard
- Implant loosening — any toggle on intraoperative testing mandates staged revision
- McPherson C host with a grade 3 extremity — failure rate exceeds 70 percent
- Previous failed DAIR — a second attempt succeeds in less than 30 percent
Critical operative steps — non-negotiable
- A minimum of five deep cultures with fresh instruments from different sites BEFORE any antibiotics
- Modular exchange is mandatory (liner and head, or insert) to access the posterior recesses
- Radical synovectomy of the entire capsule including the posterior compartments
- A minimum of 9 litres pulsed lavage; adjunctive dilute betadine or hydrogen peroxide soaks
- Confirm implant stability before reimplantation; abandon if loose
Organism-specific success rates
- Streptococcus: 75 to 85 percent implant retention at 2 years
- Coagulase-negative staphylococcus: 70 to 80 percent
- MSSA: 65 to 75 percent with a rifampicin-based regimen
- MRSA: 40 to 55 percent — consider two-stage if additional risk factors
- Polymicrobial or resistant Gram-negative: 35 to 50 percent
Antibiotic protocol
- Initial IV: vancomycin plus a third-generation cephalosporin or piperacillin-tazobactam pending cultures
- 6 weeks of intravenous therapy once the organism is known
- Biofilm agents: rifampicin for staphylococci; a fluoroquinolone for Gram-negatives
- Oral step-down for 3 to 6 months (or indefinite suppression in high-risk hosts)
- Monitor renal and hepatic function; watch for C. difficile and neuropathy
Failure predictors and management
- MRSA, sinus tract and symptom duration greater than 3 weeks — each reduces success by 25 to 30 percent
- A failed DAIR proceeds to two-stage revision; a second DAIR is rarely successful
- Persistent CRP elevation beyond 6 weeks or recurrent drainage: early conversion to staged revision
- Aseptic loosening after a successful DAIR occurs in 5 to 10 percent at 5 to 10 years
Post-operative surveillance
- CRP and ESR at 2, 4 and 6 weeks, then 3, 6 and 12 months, then annually to 5 years
- Any rise in markers or return of symptoms prompts repeat aspiration
- Wound review at 48 hours, 2 weeks, 6 weeks and 3 months
- Weight-bearing as tolerated from day 1 to 2; no routine bracing
Background & Evidence
Epidemiology and success rates. Contemporary series report overall DAIR success — implant retention without further surgery for infection — of 55 to 75 percent at 2 years when strict selection criteria are applied, with success highly dependent on timing, organism and the presence of a sinus tract. A systematic review of 39 studies (928 knees, 677 hips) found pooled success of 67 percent when DAIR was performed within 3 weeks of symptom onset versus 42 percent when performed later; staphylococcal infections (especially MRSA) and sinus tracts were independent predictors of failure on multivariate analysis. Biofilm-relevant anatomy. Periprosthetic biofilm forms preferentially at the bone-implant interface, the modular taper junctions and the polyethylene surface. The posterior recesses of both hip and knee are the most common sites of residual infection after an inadequate DAIR because they cannot be visualised or debrided without removing the modular components — the anatomical basis for the mandatory modular exchange. Classification systems that drive decision-making. Two staging systems stratify risk and frame the DAIR-versus-revision decision:
- Description
- Intraoperative positive culture
- DAIR role
- Not DAIR — treat with antibiotics alone
- Description
- Early post-operative (less than 1 month)
- DAIR role
- DAIR appropriate
- Description
- Acute haematogenous (with a well-fixed implant)
- DAIR role
- DAIR appropriate if short symptom duration
- Description
- Chronic
- DAIR role
- Not DAIR — two-stage revision
McPherson staging combines a host grade (A uncompromised, B compromised, C significantly compromised — for example immunocompromised, dialysis, active malignancy) with a local extremity grade (1 uncompromised, 2 compromised, 3 severe — multiple previous incisions, radiation, severe scarring). A McPherson C host with a grade 3 extremity has a DAIR failure rate exceeding 70 percent, which is why most surgeons proceed directly to two-stage revision in that setting.
References
Outcome of debridement and retention for prosthetic joint infection
- Single-centre series of 99 DAIR procedures with a minimum 2-year follow-up
- Implant retention achieved in 60 percent overall; success 78 percent for streptococcal versus 41 percent for staphylococcal PJI
- Symptom duration greater than 8 days and the presence of a sinus tract were the strongest predictors of failure
DAIR for periprosthetic joint infection: a single-centre experience
- Review of 104 DAIR cases (74 knees, 30 hips) with a mean 4-year follow-up
- Success rate 55 percent at final follow-up; MRSA and polymicrobial infection had the lowest retention rates
- Modular exchange was performed in only 42 percent of cases — the authors now mandate liner exchange in all DAIR procedures
Risk factors for failure of DAIR in periprosthetic joint infection
- Registry-based study of 345 DAIR procedures from the Danish Knee Arthroplasty Register
- Failure rate 38 percent at 2 years; independent predictors were MRSA, sinus tract and symptom duration greater than 21 days
- Each additional week of symptoms increased failure risk by 12 percent on adjusted analysis
Systematic review of debridement and implant retention for PJI
- Meta-analysis of 39 studies comprising 1,605 DAIR procedures
- Pooled success 67 percent (95 percent CI 62 to 72 percent) when performed early; success fell to 42 percent after 6 weeks
- Staphylococcal aetiology and sinus tract were associated with a 25 to 30 percent absolute reduction in implant retention