Keywords
arthroplasty, replacement, knee - surgical site infection - treatment - surgical review - antibiotic
Introduction
Before we start the treatment of infection in total knee arthroplasty, ideally, we should have the etiological agent correctly identified and a patient clinically balanced in order to tolerate the surgical interventions that are to come.
It is also desirable that we have good images of the compromised knee so that we can carry out appropriate planning of the surgical treatment to be implemented.
Since, due to biofilm formation, surgical debridement is required for its removal, the main surgical alternatives are joint debridement with implant retention (JDIR), single-stage revision, and two-stage revision. In case these procedures fail, rescue procedures may be required.
Joint Debridement and Implant Retention
Joint Debridement and Implant Retention
Joint debridement is the treatment of infection without removal of the prosthesis, replacing only polyethylene. To optimize its result, it is essential that the infectious process is in the acute phase, when the bacterial biofilm is not yet mature.[1] In addition to time, other criteria should be considered, such as the absence of fistulas, and that the prosthesis is fixed and functional.[2] Joint debridement and implant retention is supported by the II-ICM-2018 (II International Consensus on Muscleskeletal Infection – 2018), with 80% agreement.[3]
Due to the heterogeneity of the studies, JDIR success rate ranges from 16 to 100%, with an overall average of ∼ 50%.[4] Cases of poor results may be related to biofilm formation in a shorter time.[5]
Iza et al.[6] found in a retrospective analysis of cases submitted to JDIR a significant difference between the success rate in patients with acute postoperative infection (93%) and acute hematogenous infection (58%). They also observed a much lower success rate in cases infected with Staphylococcus aureus (33%) compared with other bacteria (82%).[6] The low success rate in infections caused by S.
Aureus has also been reported by several other studies.[7]
[8]
Two scores were developed to predict the risk of JDIR failure.
The KLIC-Score, described for early acute infection, evaluates 5 factors, assigns points to each one, and calculates the chance of failure according to the score obtained ([Table 1]).[8]
Table 1
|
KLIC-SCORE
|
|
Chronic renal failure (K for kidney)
|
2 points
|
|
Liver cirrhosis (L for liver)
|
1.5 point
|
|
TKA indication (I for Index)
|
If fracture or revision, 1.5 point
|
|
Cemented prosthesis (C for cement)
|
2 points
|
|
CRP value (C for CRP)
|
If > 11.5mg/dL, 2.5 points
|
|
JDIR Score x chance of failure
|
|
≤ 2 points
|
4.5%
|
|
2.5 to 3.5 points
|
19.4%
|
|
4 to 5 points
|
55%
|
|
5.5 to 6.5 points
|
71.4%
|
|
≥7 points
|
100%
|
CRIME80, described for late acute hematogenous infection, defines 7 predictors of outcome ([Table 2]).[7] The authors found as the main predictor of success the exchange of modular components (polyethylene), so the subtraction of a point when it is performed. According to the score obtained, they attribute the chance of JDIR failure.
Table 2
|
CRIME80
|
|
Chronic obstructive pulmonary disease (C for COPD)
|
2 points
|
|
CRP value >15mg/dL (C for CRP)
|
1 point
|
|
Rheumatoid Arthritis (R for Rheumatoid)
|
3 points
|
|
Indication of prosthesis (I for Indication)
|
if fracture, 3 points
|
|
Male (M for Male)
|
1 point
|
|
Exchange of modular components (E for Exchange)
|
(-)1 point
|
|
Age > 80 years old
|
2 points
|
|
JDIR Score x chance of failure
|
|
(−) 1 point
|
22%
|
|
0 point
|
28%
|
|
1 to 2 points
|
40%
|
|
3 to 4 points
|
64%
|
|
≥ 5 points
|
79%
|
The literature is also controversial regarding the impact that a failed JDIR causes in a subsequent treatment with removal of implants.
Rajgopal et al.[9] retrospectively analyzed the results of patients submitted to a 2-stage review and found an increased failure rate, worse functional scores, and a higher rate of wound complications in patients with a previous history of JDIR.
Similarly, Lizaur-Utrilla et al.[10] also found better results in functional scores and range of motion (ROM) in patients who were not submitted to JDIR prior to review in 2 stages.
Kim et al.,[11] however, in a retrospective study, found no difference in the results of patients submitted to review as the first treatment option when compared with those after JDIR failure.
Two-stage review
This method is best indicated in chronic infections, in patients with systemic involvement,[12]
[13]
[14] when bacteria identification is not available[12]
[13] or when it is resistant to available antibiotics (fungi, gram negatives and S aureus).
[3]
[13] The technique has few formal contraindications;[3]
[12] it can be applied at any stage of infection (acute or chronic)[12]
[13] with high success rates, being considered the gold standard.[15]
[16]
[17]
[18] However, two-stage treatment is associated with longer hospitalization time, functional recovery and, consequently, higher costs,[19]
[20]
[21] as well as with higher mortality in 1 postoperative year.[19]
[22]
The first stage consists of the removal of the prosthesis. At the surgical access, a skin spindle including the previous scar and fistula(s) should be excised.[13]
[23] At this stage, the implants are removed by the same access of the primary arthroplasty, preserving, as far as possible, the bone stock, followed by broad debridement with the exeresis of all inflammatory and necrotic tissues and of compromised sections of the articular capsule, with the preservation, if possible, of collateral ligaments.[13]
[21]
[23]
[24] Then, the wound, the joint cavity and the medullary canal are irrigated, using pulsatile washing with at least 10 to 12 liters of saline solution, and some antiseptic solution can be used optionally.[21]
[24]
Fragments of debrided material should be sent for culture and histopathology.[25] The crop must consist of three to six samples from different areas of the knee, with cultivation time of at least 14 days.[13]
[24]
[26] The explanted components can be submitted to sonication to break the biofilm and the washing of these components may be sent to culture, which can be useful in cases of infections with negative culture.[26] In situations in which there is positivity for other tissue samples, the positivity of the sonicated fluid should only be taken into account if there are > 5 colony forming units (CFUs).[27]
Histopathological tissue analysis may also be a diagnostic confirmation factor in cases with negative synovial fluid cultures and suspected aseptic loosening, with sensitivity of 75% in freezing cuts, with a threshold of 5 PMN/field.[28] Histopathology by perioperative freezing, histopathological analysis by staining and/or immunohistochemistry are very useful; however, they are examiner-dependent, as defined by the CIIM-2018.[28]
After surgery, the patient is accompanied with serial dosage of inflammatory markers and evaluation of local and systemic clinical improvement.[29]
[30] If there is no improvement, or no reduction in inflammatory markers, a new debridement should be carried out with spacer change.[31] If there is clinical and laboratory improvement, reimplantation is carried out with a prosthesis whose degree of constriction and need for correction of bone failures will be individualized for each case.[3]
[21] In the review surgery, a new debridement and sample collection is performed for culture and histopathological analysis by freezing in order to evaluate the presence of subclinical persistent infection.[3]
[24]
Some controversies regarding the 2-stage review, such as the type of spacer, the moment, and the conversion criteria, as well as the period of antibiotic therapy,[13] still need clarification.
Joint spacers are classified as static, mobile, prefabricated or handcrafted.[18]
[24]
[32]
[33] The static ones are better indicated in cases of extensor apparatus insufficiency, large bone defects, wound healing problems, and ligament instability.[3]
[32] They are associated with complications such as postoperative stiffness and bone loss. This, in particular if there is dislocation of it, more frequent in artisanal spacers and obese patients.[3]
[13]
[32]
[33]
Articulated or mobile spacers can be modeled with antibiotic cement with prefabricated mold,[3]
[32] a new component,[34]
[35] or the removed reprocessed with flash sterilization.[36] The advantages of articulated spacers are preserving ROM, a better of quality of life, and lessening the need for extended approaches at the revision1.[8]
[32]
[37] Although, some studies hasn't shown statistical differences in ROM between static and articulated spacers in long term, and articulated spacers are associated more frequently with joint instability and breakage, specially in prefabricated spacers.[3]
[18]
[33]
[35]
Yu et al.[38] showed in their systematic review that dynamic spacers with metallic femoral component articulated with tibial polyethylene have higher reinfection rates than spacers made entirely of cement. There is controversy in this regard in the literature.[32]
[35]
Although most authors did not find superiority among the types of spacers regarding the cure of infection,[3.32] Romanò et al.,[39] in a systematic review, observed that dynamic spacers have a higher rate of eradication.
The criteria for reimplantation are also grounds for controversy,[16]
[29]
[30] because the ability to define whether the infection is cured still requires further studies.[40]
Some authors recommend the review within 6 to 8 weeks.[13] This measure decreases hospitalization costs in relation to protocols that suggest longer intervals, which can extend to up to 16 weeks, without any difference in reinfection rates.[40] Intervals > 16 weeks are associated with an increased incidence of relapses.[13]
[40]
Several parameters for reimplantation were proposed, such as the criteria for diagnosis of periprosthetic infection of the Musculoskeletal Infection Society (MSIS). These criteria include inflammatory markers (C-reactive protein [CPR], Erythrocyte Sedimentation Rate [ESR], and D-Dimers), cytometry, biochemical markers, and aspirate culture.[41] The MSIS infection criteria have high specificity and low sensitivity for persistent infection, with a high positive predictive value (PPV) and a low negative predictive value (NPV),[30] so several persistent infections are underdiagnosed.[26]
[30]
The leukocyte count of the joint aspirate also has a high NPV at values < 3,000 cells/uL.[16]
[30] The culture of joint aspirate before the review also has high specificity and low sensitivity for persistent infection, besides a great correlation with the germ of possible reinfections.[24]
[30]
[31] It should not be a routine procedure for cases with clinical and laboratory improvement.[31] The II-ICM-2018 could not define a definitive parameter for reimplantation, and even in cases of clinical and laboratory improvement, the persistence of the infection is still suspected.[30]
Perioperative histopathological analysis by freezing has high specificity and PPV, but low sensitivity and NPV.[13]
[16]
[30] Meanwhile, Fu et al.[40] found high sensitivity and specificity in their series of 81 cases.
During the review procedure, local conditions should be reevaluated and at least four culture samples must be collected.[24] In the case of positivity, intravenous antibiotic therapy guided by the results should be initiated.[3]
The period of antibiotic therapy is a matter of controversy. The CIIM-2018 suggests a minimum period of 4 to 6 weeks, but the parenteral and oral percentage should be individualized by the microorganism detected in the culture, as well as by the response to treatment.[42] The CIIM-2018 and other studies show that oral antibiotic therapy extended for at least 3 months after review decreases the rate of reinfection.[42] It is worth remembering that these strategies should be decided together with the infectologist and the general practitioner, emphasizing the multidisciplinary character that should guide the treatment of periprosthetic infections.
Studies also try to demonstrate risk factors for treatment failure. Fu et al.[40] found that perioperative biopsy by positive freezing, atypical germs, and presence of fistula had a high rate of reinfection. Surprisingly, infections with negative culture have reinfection rates similar to those of positive culture, and they are not considered a risk factor for failure.[43]
Single-stage review
Single-stage review is indicated when the etiological agent is known, sensitive to available antibiotic therapy, there is no systemic involvement of the patient, and the patient is not immunocompromised.[19]
[21]
[44]
It is contraindicated when there is soft tissue injury that does not allow primary closure of the surgical wound, in the presence of nonexcisable fistula with the scar of previous access, in the impossibility of rigorous debridement, in the presence of severe bone defect, when the etiological agent is multiresistant, or in the absence of effective antibiotic therapy against the isolated germ.[44]
Aiming to reduce morbidity and the cost of treatment, the single-stage review seeks to achieve the same results in terms of eradication of infection and durability of the two-stage review.[15]
[19]
[20]
[21]
[22]
[24] Some studies have shown that single-stage review in selected patients may have similar or even better results[21]
[45]
[46] than the two-stage review.[3]
The success of this type of review increases considerably with the prior identification of the etiological agent.[22] Previous joint puncture with synovial fluid cultures for an extended period of 14 days is mandatory.[22]
[24]
[47]
Some authors did not find a difference in outcome in single-stage reviews without prior identification of the etiological agent.[14]
[48]
[49] There are reports of revisions at a stage performed "inadvertently", when apparently aseptic revisions were actually septic after intraoperative cultures became positive.[19]
Fungal infections by S. epidermidis and S. aureus have a worse evolution and higher incidence of failures.[3]
[15]
[21] Citak et al.[48] show that Enterococcal infection is 14 to 21 times more likely to be reinfected. Klatte et al.,[47] although they indicate single-stage review as an alternative to fungal infection, showed in their results two failures in four cases. Ji et al.[50] show that single-stage revision may be an alternative in fungal infections. However, two of the seven cases presented reinfection, having been treated only with debridement and antifungals.[50]
The presence of fistula is a controversial contraindication for single-stage revision. While some authors have described a high rate of reinfections in patients with fistula,[21]
[24]
[46] others did not find significant difference in reinfection compared with those of the two-stage review, since the fistula can be excised along with the surgical scar in the joint capsule.[24]
[44]
[46]
The presence of bone defects is also a debatable contraindication to single-stage revision. Zahar et al.[22] define that bone defects should be filled with cement with antibiotics. In 59 patients with a mean follow-up of 10 years, the reinfection rate was of 8.47% (5/59), 7 patients presented with aseptic loosening (11.86%) and there were 25 more patients (42.37%) at high risk of release.[21]
A single-stage review should follow a strict protocol to increase its success rate.[21] The procedure consists of two phases:
First, implants and all cement should be removed, along with broad and aggressive synovectomy, with radical resection of necrotic and devitalized tissues.[15]
[20]
[22] However, how aggressive the debridement should be is a controversial issue. Some authors advocate resection of collateral ligaments, which requires the implantation of constricted prosthesis in rotational hinge,[22.46] while others advocate preservation of the medial collateral ligament (MCL), which allows the use of prosthesis with varus-valgus constriction.[15]
[51]
In this first phase, at least six fragments of different parts of the knee are sent for culture and histopathology.[15]
[46] The wound, the medullary canal and the joint are irrigated with 10 to 15 liters of 0.9% saline solution, preferably with pulsatile washing, and an antiseptic, iodinated or chlorhexidine-based solution associated or not with hydrogen peroxide solution of can be used.[15]
[20]
[21]
[22]
[46] After irrigation, the wound is temporarily sutured and protected with sterile iodated adhesive field. All surgical and instrumental fields used in explanting and debridement surgery are exchanged and, if possible, exchange or hygiene of the surgical environment itself should be performed.[15]
[20]
[21]
[22]
[46]
In a second phase, the patient is prepared with new fields and surgical instruments. The sutures are removed, and the wound is again irrigated with 0.9% saline solution. The prosthesis is implanted using cement with integrated antibiotic, not exceeding 10% of the weight-dose, according to the antibiogram of the infectious agent.[15]
[21]
[22]
[46]
[47] Rods are used to improve the stability of the prosthesis to the bone, and this should have appropriate constriction for the case.[22]
[46] Bone defects should be addressed according to their size, with small flaws being filled with cement and larger flaws with wedges, blocks, or metal cones, avoiding the use of allograft.[15]
[46]
In the 2018 consensus, the recommended period for intravenous antibiotic therapy after single-stage review is 7 to 14 days, followed by oral antibiotic therapy for a total period of 6 to 8 weeks, with a limited level of evidence and 73% agreement,[52] which is corroborated by other authors.[22]
[46] However, several studies show that extended parenteral antibiotic therapy protocols for 6 weeks decrease reinfection rates.[15]
[21]
[42]
It is worth remembering, once again, that the antibiotic therapy strategy, as well as the many adverse reactions related to its use, should be managed in a multidisciplinary manner with the help of the infectologist and of the general practitioner.
Single-stage and two-stage review
Single-stage and two-stage review
Single-stage revisions, in selected cases, have a lower or similar reinfection rate to two-stage revisions. It presents as advantages lower costs, lower mortality rate, shorter hospitalization time and functional recovery. Thus, if there are no contraindications, this option should be considered.[19]
A meta-analysis of 2016 that analyzed 10 single-stage review studies against 108 two-stage review studies, found similar reinfection rates of ∼6.4%.[53]
Thakrar et al.,[14] in a 2019 systematic review, showed comparable results in single- and two-stage reviews in relation to the reinfection index in patients without systemic or immunocompromised involvement. However, they pointed out that most studies are retrospective or observational and lack quality studies such as randomized clinical trials.
It is important to note that, while at the 2013 IIC the agreement rate among panelists on the indications and contraindications of the single-stage review was 78%,[54] at the 2018 IIM, with evaluation of more studies, the agreement was of 93%, with a moderate level of evidence.[44]
Rescue measures
In cases of periprosthetic knee infection refractory to previous treatments, treatment options consist of:
-
arthrodesis
-
transfemoral amputation
-
resection arthroplasty
-
antibiotic suppression
Rescue measures should be indicated early for patients who have many comorbidities.[3] In cases of treatment failure in patients without many comorbidities, it can be treated with another two-stage review attempt. McPherson type C hosts have better results with arthrodesis or amputation.[3]
The functional result of arthrodesis has been shown to be superior to that of amputation. Few amputee patients can adapt to the prosthesis and walk again.[3]
[55] On the other hand, most patients submitted to arthrodesis have preserved walking capacity.[3] In the study by Mozella et al.,[56] 44% of the patients submitted to amputation were able to be protetized, only 27.78% were community ambulators and 56% became wheelchair users.
Patients classified as host type C and with soft tissue involvement requiring coverage procedures have a high rate of recurrence of infection requiring arthrodesis, amputation or antibiotic suppression.[57]
[58]
Resection arthroplasty has the theoretical advantages of limb preservation, of no need for implants or synthesis material, possibility of knee flexion and theoretical gait capacity with immobilizer and compensation of dysmetry.[59]
Antibiotic suppression is indicated as a rescue measure in patients who are unable to undergo new surgical procedures.[55] Antibiotic toxicity, oral availability and infection suppression capacity should be considered in order to indicate this type of treatment.[55]
