Left PJI knee infection, with a sinus
Complete excision of all necrotic tissue
Minimal bone loss seen after removal of the tibial component
Betadine placed in the wound at the end of the debridement
Hydrogen peroxide acting as a mechanical debridement
One-stage exchange surgery should be viewed as two entirely separate surgical procedures making up a single operation. Prophylactic antibiotics should not be given preoperatively as per routine and only once sufficient samples have been taken and sent to the microbiologist. A laminar flow theater is preferred, however, not a necessity.
A tourniquet should be applied but not inflated during the initial surgical debridement, making it easier to distinguish between healthy, bleeding tissue and necrotic, infected tissue. This also decreases the potential for complications related to tourniquet use.
The skin is prepared twice with 3M DuraPrep solution which contains iodine povacrylex and isopropyl alcohol . The skin is covered by a 3M Ioban antimicrobial incision drape.
During the initial phase, the surgeon must take great care when planning the skin incision. Previous skin incisions must be clearly marked to ensure they are distinctly visible after skin preparation and draping [31–33]. Skin necrosis may be avoided by ensuring that surgical incisions are generally at an angle of >60° to old incisions and skin bridges of >7 cm are preserved [32–34].
In the presence of multiple scars, the incision most recently made should be used, and the most lateral longitudinal surgical incision through which adequate exposure is possible should be selected. The blood supply to the anterior part of the knee is mostly derived from the medial aspect, and therefore the safest skin incision is the most lateral [32, 33, 35]. Cutaneous blood supply arises from deep to superficial, and therefore thick skin flaps are essential. Preexisting transverse skin incisions should be transected at a perpendicular angle [33, 36].
Fistulae must be incorporated into the skin excision and radically excised down to the joint capsule . A plastic surgeon may need to be on hand if any musculocutaneous flaps are anticipated, but this will often necessitate a two-staged exchange arthroplasty.
Once incised, the first part of the operation involves a knee joint arthrotomy and the aggressive open surgical debridement of the periarticular soft tissue envelope. Careful attention must be paid to the posterior capsular soft tissue which is often neglected. It is essential to remove all prosthetic components without any unnecessary bone loss and inadvertent periprosthetic fractures. It may be easier to remove the cemented prostheses than well-fixed, uncemented implants. It is important to remove all cement within the distal femoral and proximal tibial canals to remove biofilm and necrotic tissue .
Cortical windows may be utilized to gain access to the interface in uncemented implants with good bony ingrowth. High-speed burrs and curved sharp osteotomes may facilitate expeditious extraction. Universal or special extraction instruments may be employed to remove the implants. However, general punches may be used to good effect. The amount of resected material in the debridement of a one-stage exchange must be extensive and may even exceed that removed in a two-stage procedure .
Multiple specimens should be collected for microbiological and histopathological evaluation representing the whole surgical field [37, 38] and labeled appropriately to reflect the anatomical area from which the specimens are taken (minimum of five specimens). This step is imperative to guide appropriate postoperative antibiotic therapy.
The administration of parenteral antibiotics selected empirically preoperatively or based on culture sensitivities should be administered after all specimens have been obtained. The omission of preoperative, prophylactic antibiotics until intraoperative microbiological specimens have been obtained is controversial . Some studies have shown that they have no impact on subsequent isolation of organisms from these specimens. Regardless of preoperative antibiotics, bacterial cultures from intraoperative specimens and preoperative aspirates have been shown to be the same in 97% of cases .
Irrigation of the surgical field with copious amounts of fluid is essential. After debridement, 12 L of warm 0.9% saline via low pressure pulsatile lavage is used. Thereafter, 100 mL of a 50:50 mix of 3% hydrogen peroxide and 100 mL of sterile water solution are added. The excess hydrogen peroxide is subsequently removed from the surgical field by 100 mL of sterile water solution. Next, 200 mL of sterile 10% aqueous povidone-iodine is used as an irrigant.
Antibiotics can be added to irrigation fluids [40, 41]; however, this has not been shown to result in any significant clinical improvements [42, 43]. Care should be taken as systemic absorption and toxicity, bacterial resistance, and additional costs may result from complications of this practice [40, 41].
Once satisfied that the surgical field is macroscopically clear of infected tissue, the wound is packed with povidone-iodine-soaked gauze, and the wound edges are approximated with a continuous one Vicryl.
A sterile antimicrobial drape is used to cover the wound, thus maintaining sterility while in use. Soiled drapes are exchanged for clean sterile ones, and the surgical team don new gowns. Additionally all the equipment, reusable or disposable, are removed from the operating room, and new instruments are brought into the theater.
The wound is then opened, sutures discarded, and the entire surgical field undergoes a further lavage. Similarly to a two-stage exchange, at this point intravenous antibiotics should be given as per recommendations of the microbiologist and infectious disease specialist . Antibiotics may also be added at the time of inserting the definitive implant by adding it to the cement  and mixing it to bone graft (if uncemented)  along the femoral stem or via adjuncts such as calcium sulfate pellets [47, 48].
The tourniquet is inflated before cementing to improve the cement-bone interface . Before closure, the wound is again irrigated with 1 L of 0.9% sodium chloride. Drains may be inserted to prevent postoperative hematoma formation. Expeditious drain removal ideally within the first 24 h will allow high concentrations of antibiotics postoperatively in the surgical field.
18.6 Antibiotic Role
As already discussed, the use of antibiotics and cooperation between the surgeon and their infectious disease and microbiology colleagues are essential in a single-stage exchange.
In uncemented single-stage revisions, Winkler et al. utilizes vancomycin-impregnated allograft and has demonstrated high local levels of vancomycin over 2–8 weeks . This sustained elution prevents local bacterial contamination and biofilm formation, preventing reinfection in 92% of 37 infected hip replacements seen at a minimum of 4 years of follow-up .
Postoperatively, the optimal timing to switch from intravenous antibiotics therapy to oral remains controversial. The decision is influenced by the host, duration of symptoms, bacteria type, and the antimicrobial sensitivity  and typically given for 10–14 days . Streptococci infection may demand longer periods of antibiotic therapy. Oral antibiotics may be commenced after 2 weeks.
The choice of antimicrobial agent may need to be modified, if the resistance profile of the bacteria isolated during the one-stage exchange differs from the previous cultures . Advice should be sought from the microbiologist or infectious disease specialist, an integral member of the multidisciplinary team.
18.7 Outcome Measures
18.7.1 Infection Eradication
The results of the literature comparing one-stage and two-stage exchange arthroplasty are based mainly on individual single center, small retrospective series with short- to medium-term follow-up . In these papers, a standardized definition of the endpoint or treatment failure is often lacking.
Von Foerster et al.  reported the first series of one-stage exchanges in 1991 from the Endo-Klinik in Hamburg . This series involved 104 TKAs that were revised for PJI. A constrained, stemmed, rotating-hinge prosthesis was used in all cases. A favorable outcome was reported in 76 cases (73%), a follow-up of between 5 and 15 years, but persistent infection was seen in 20 knees.
Buechel  reported on a series of 22 TKA PJI managed with a one-stage exchange. The infection was eradicated in 20 knees (90.9%) at a follow-up of 10.2 years. He reported good or excellent knee scores in 18 knees (85.7%). Parkinson  proposed a 6-week postoperative course of sensitive oral antibiotics after eradicating infection in 100% of the 12 patients at a mean follow-up of 2 years.
In a comparison of outcomes between one- and two-stage revisions for infected TKAs, Bauer et al.  conducted a retrospective multicenter study involving 107 patients with no differences in the ability to eliminate infection between either technique. However, Silva  reviewed data from 30 papers, and a one-stage exchange showed a favorable outcome.
Chew et al.  conducted a systematic review of 12 studies describing one-stage exchange arthroplasty in 433 revision surgeries. This meta-analysis highlighted the heterogeneity in reviewing this condition as there were marked differences in these papers with regard to surgical techniques, implant usage, and philosophies in fixation and antibiotic regimens, and often surgical techniques were not described within the original papers.
Early ambulation and a prompt initiation of functional exercises is a significant benefit of one-stage exchange, compared to the two-stage exchange where an articulating or non-articulating spacer may be used. Postoperative rehabilitation focuses on improving range of movement to prevent joint stiffness and fibrosis and recovery of normal gait. Immediate full weight bearing with crutches should be commenced within the first postoperative day.
Buechel  and Singer  reported good to excellent functional outcomes in most patients who underwent one-stage exchange arthroplasty. They published Knee Society Scores (KSS) of 79.5 and 72, respectively, which is better than they had seen in two-stage revisions.
Haddad  used strict selection criteria to determine suitable patients for a one-stage exchange and subsequently compared the outcomes directly with those patients who underwent two-stage revision surgery for chronic PJI of the knee. Infection control was achieved in all of the patients who underwent one-stage exchange surgery, while the infection was eradicated in 93% of the two-stage group. The functional outcomes were also significantly superior in the one-stage group (KSS score, 88 vs. 76; p < 0.001).
Zahar et al.  reported on the Endo-Klinik’s 10-year survival rate for one-stage exchange of infected TKA was 93% (p = 0.007). Correspondingly, the 10-year survival rate for knees revised for a reason other than infection with a one-stage exchange arthroplasty during this period was 91% (mean, 5.2; 95% CI, 86–95%; p < 0.002).
Baker et al.  compared the functional outcomes using Patient Reported Outcome Measures (PROMs) and satisfaction rates in 33 one- and 89 two-stage revision TKA for PJIs. Assessment of outcomes using PROMs was unable to demonstrate any difference between the two approaches. The ultimate satisfaction rates were equivalent, and no difference in general health perception was noted. The conclusion was that eventual functional results could not be used as a determinant in the selection of either one- or two-stage revision.
A meta-analysis undertaken by Kunutsor et al.  comparing both treatment options, demonstrated comparable postoperative knee range of movement and equivalent knee scores, although the numbers in the one-stage exchange cohort are comparably small.
One must also consider the significant psychosocial implications of the infection and the impact of subsequent investigations and medical and surgical treatments. It affects all aspects of the patients’ lives as demonstrated by Moore et al. . They highlighted the emotional distress and psychosocial imposition were greater in those treated with a two-stage revision than a one-stage exchange, as a result of the longer periods of immobility and time in-between surgical procedures. An increased need for psychosocial and rehabilitative support during the duration of treatment and rehabilitation may be necessary.
18.8 Economic Impact
Revision TKA represents a significant economic burden. Individual surgical procedures are expensive as a consequence of the combined costs of preoperative investigations, prostheses and instrumentation, an extended hospital stay, and weighty pharmacological costs . The expense attributed to the revision implants and preoperative investigations has been implicated as a major cost driver to improving patient care . The forecasted cost of revision TKA in the USA is predicted to surpass $2 billion by 2030 .
Direct comparisons of the differences in costs involved between the one- and two-stage philosophies are challenging. The type of hospital facilities, individual patient factors, surgeons, and the infecting organism itself influence these costs.
It may, however, be inferred that in the event of only one major surgical procedure, less patient morbidity, operative time, operating room utilization, hospital and surgeon costs, and duration of antibiotic therapy exist. No overwhelming evidence exists that this is indeed the current situation [30, 62–66].
There may also be added unforeseen costs involved in the treatment of one-stage exchange arthroplasty as a consequence of increased reinfection rates. This amounts to extra costs to supplement additional diagnostic tests and clinical work-up as well as the likelihood of the ancillary expense of reoperation. Contention, therefore, exists whether two-stage revision is perhaps more advisable in PJIs caused by highly virulent organisms. MRSA has been recognized as a source of particular expense as it is onerous and costly to successfully eradicate it [63, 67].
Wolf et al.  endorsed one-stage exchange arthroplasty over two-stage exchange using a Markov expected-utility analysis considering quality-adjusted life years as health endpoints. In doing so, the reported rates of successful infection eradication form the basis when comparing the expense involved in one- and two-stage revision surgeries. A benchmark rate for infection clearance of 93% was deemed appropriate for two-stage revision and 85% for a one-stage exchange.
Therefore, when managing 100 cases of septic TKA with this technique, 200 operations are performed. During a two-stage exchange, 214 operations will ultimately be undertaken (additional 14 surgical procedures as a result of the 7% failure rate). In contrast, the one-stage exchange translates into 15 cases that fail as a result of reinfection, and if revised later to two-stage revisions, this will equate to an additional 30 surgical procedures (total 130 surgical procedures). Subsequently, based upon this hypothetical group of patients, the implementation of a one-stage exchange would ultimately decrease the number of surgical procedures by 84 cases.
The financial cost attributed to each revision procedure is difficult to estimate but quoted as approximately £26,000  but may be as high as £75,000 . Therefore based upon Wolf et al.  theoretical cohort, a potential saving of £2,184,000 may be achieved if these extra 84 operations are not undertaken.