Outcomes of infected revision knee arthroplasty managed by two-stage revision in a tertiary referral centre

The Knee 22 (2015) 56–62

Contents lists available at ScienceDirect

The Knee

Outcomes of infected revision knee arthroplasty managed by two-stage revision in a tertiary referral centre John Stammers ⁎, Steven Kahane, Vijai Ranawat, Jonathan Miles, Rob Pollock, Richard W.J. Carrington, Timothy Briggs, John A. Skinner Royal National Orthopaedic Hospital, Brockley Hill, Stanmore, Middlesex, London HA7 4LP, United Kingdom

a r t i c l e

i n f o

Article history: Received 27 April 2013 Received in revised form 14 October 2014 Accepted 28 October 2014 Keywords: Periprosthetic joint infection Knee revision Two-stage revision Knee arthroplasty

a b s t r a c t Background: A two-stage revision remains the gold standard to eradicate deep infection in total knee arthroplasty. Higher failure rates are associated with a number of factors including poly-microbial infections, multiresistant organisms and previous operations. The aims are to investigate [1] the overall success rate of a two-stage revision for infections in TKA, [2] the outcome of repeat two-stage revisions in recurrent infections and [3] the factors affecting the outcomes of such cases. Methods: We present the outcomes of a consecutive, retrospective case series of 51 periprosthetic joint infections managed with a two-stage revision knee arthroplasty over a three year period. Results: Forty-six (90%) of 51 were referred from other hospitals. Infection was successfully eradicated in 24 (65%) of 37 patients undergoing an initial two-stage procedure. Following a failed two-stage revision, a repeat two-stage revision was performed in 19 patients eradicating infection in 8 (42%). A third two-stage was performed in five of these patients eradicating infection in three with an average follow-up of 43 months. Multidrug resistance was present in 69%, and 47% of the patients were infected with multiple organisms. All unsuccessful outcomes involved at least one multidrug-resistant organism compared to 43% in the successful cohort (P = 0.0002). Serological markers prior to a second-stage procedure were not significantly different between successful and unsuccessful outcome groups. Conclusion: Single or multiple two-stage revisions can eradicate infection despite previous failed attempts. In this series, failure is associated with multidrug resistance, previous failed attempts to eradicate infection and a less favourable host response. Level of evidence: IV © 2014 Elsevier B.V. All rights reserved.

1. Introduction Total knee arthroplasty surgery is a commonly performed surgical procedure with almost 85,000 cases being in England and Wales in 2012. Based on joint registry data, 6009 revisions were performed in 2012 of which 22% were performed for infection. This included 255 single stage revisions and 706 two-stage revisions [1]. Two-stage revision arthroplasty remains the gold standard to eradicate deep infection in total knee arthroplasty (TKA) with quoted success rates of 80–100% [2]. Higher failure rates are associated with number of factors including medical co-morbidities, immunosuppressive states, poly-microbial infections, multiresistant organisms and previous operations [3]. [4,5] Rates of infection after total knee arthroplasty are low, but with increasing life expectancy and the consequential increase in the number of procedures performed, this burden is likely to increase further [6,7]. ⁎ Corresponding author at: Royal National Orthopaedic Hospital, Stanmore, Middlesex, London HA7 4LP. Tel.: +44 7793534545. E-mail address: [email protected] (J. Stammers).

http://dx.doi.org/10.1016/j.knee.2014.10.005 0968-0160/© 2014 Elsevier B.V. All rights reserved.

The treatment options include irrigation and debridement with exchange of polyethylene liner, [8–10] single stage revision, [11,12], two-stage revision [3,4,9,13–15], resection arthroplasty [16], arthrodesis [17], long-term suppressive antibiotics [18] and above knee amputation (AKA). [19] Repeat two-stage revision surgery can be undertaken where initial attempts to eradicate infection have failed. There is little evidence in the literature specific to deep chronic infection, with poor outcomes in small series [20–23]. A number of recent papers presenting series at tertiary units, where all patients are included regardless of prior treatment, particularly in the presence of multidrug resistance or after failed irrigation and debridement with prosthesis retention, are showing significantly worse results than previous papers. [24–27] The aims of our study were to investigate [1] the overall success rate of a two-stage revision for infections in TKA [2], the outcome of repeat two-stage revisions in recurrent and persistent infections and [3] the factors associated with worse outcomes in such cases. We hypothesised that cases involving multiple organisms or those that are infected with multidrug-resistant organisms would lead to a

J. Stammers et al. / The Knee 22 (2015) 56–62

worse outcome. We also hypothesised that inflammatory markers prior to a second-stage procedure, age and gender would have no direct correlation with outcome.

1.1. Patients and methods Between January 2006 and December 2008, 430 consecutive total knee arthroplasty revisions were performed for all causes at our unit. We reviewed all of these cases with the aim of identifying those that were performed for infection. Our criteria for infection were defined by at least one of these factors [28]: (1) a sinus communicating with the prosthesis, (2) pus within the joint and (3) documented pathogens from at least two samples of tissue or fluid culture following aspiration, washout or first-stage revision. We used the Tsukayama system for classifying infection [29]. Where there was doubt, repeat tissue and fluid samples were taken having stopped all antibiotics for a period of at least two weeks. We identified 58 revisions for infection. We excluded seven cases. This left us with 51 cases of infected knee arthroplasty that we had treated with a two-stage revision. Four patients had the primary TKA performed for bone tumour and three patients had primary TKA performed for infected internal fixation following open fractures. These cases were excluded from the study. Irrigation and debridement procedures were not included as a revision procedure in this study. This resulted in a total of 51 patients who underwent a two-stage revision surgery (27 male, 24 female) with a mean age of 72 years (range 23–89 years). Case notes of all the patients were reviewed and data recorded on the date and the institution performing the primary arthroplasty, previous operations for infection, surgical treatment at our institution, microbiological results, ESR and CRP prior to the second-stage revision, follow-up and outcome. Surgery was performed by multiple surgeons, but principles of infected revision surgery were constant. The first-stage revision surgery involved debriding all unhealthy tissue including sinuses, synovium and ligaments, tailored to each individual case. The implants were removed and the femoral and tibial bone surfaces were resected to healthy bone. Intra-medullary canals were cleared of all cement and membranes. Multiple tissue samples were taken again at this stage for culture and sensitivity prior to commencing antibiotics. The first-stage joint reconstruction was performed using an articulating cement spacer in the presence of minimal bone loss. When significant bone loss was encountered, a prosthesis (Stanmore Modular Individualised Lower Extremity System, SMILES, Stanmore Implants Worldwide, Middlesex, UK) wrapped in gentamicin and vancomycin cement was loosely implanted to maintain function and the soft tissue envelope between stages. Local flap coverage was performed in house in conjunction with plastic surgeons where required, or negative pressure dressings were applied for staged reconstructions at our regional plastic surgical unit. Although not a definite contraindication to two-stage revision if free tissue transfer was required for large soft tissue defects, arthrodesis was favoured. Additional antibiotics were added to the cement in cases where microbiology results were known pre-operatively. Broad spectrum intravenous (IV) antibiotics were continued via a long line, guided by enrichment cultures and sensitivity results. The second-stage procedure was performed after at least six weeks of IV antibiotics in correlation with clinical and serological assessments. In cases where infection was not eradicated the antibiotics were either prolonged or the patient was taken back to theatre for a repeat firststage procedure on an individualised basis. Bone loss was calculated using the Anderson Orthopaedic Research Institute (AORI) method [30] based on radiographs before a secondstage procedure (see Table 1). Results were independently assessed by two authors and the inter-observer variability calculated using Spearman's correlation coefficient.

57

Table 1 Anderson Orthopaedic Research Institute grading of bone loss. Type

Severity of bone deficiency encountered

1

Minor femoral or tibial defects with intact metaphyseal bone, not compromising the stability of a revision component. Damaged metaphyseal bone. Loss of cancellous metaphyseal femoral bone requiring reconstruction to provide stability to the revision component. A: Defects in one femoral or one tibial condyle. B: Defects in both femoral or both tibial condyles. Deficient metaphyseal segment compromising a major portion of either femoral condyles or tibial plateau.

2

3

A successful outcome was defined as a functioning prosthesis with healed wounds, no sinuses or other clinical evidence of sepsis together with no radiographic evidence of infection. Unsuccessful outcomes included patients on long-term antibiotic suppression, those awaiting a further procedure due to ongoing infection or above knee amputation (AKA). Microbiological and serological investigations were added where there was clinical concern. A maximum likelihood of contingency tables was performed in JMP (SAS Inst. Inc. Cary, NC, USA) for microbiology data analysis and SPSS version 17 for all others (SPSS Inc., Chicago, IL, USA) 2. Results A total of 51 patients who had revision TKA were managed with two-stage revision performed for infection over the study period of three years. Forty-six patients were referred from other hospitals, and five patients had primary TKA performed at our institution. There were six Tsukayama type III and 45 type IV infections. 2.1. Overall success rate Overall, deep infection was successfully eradicated in 35 (69%) of 51 patients. Of the remaining 16 patients, seven are currently being treated by long-term suppressive antibiotics due to patient choice or lack of fitness for surgery, seven have undergone AKA and two are awaiting further surgery for ongoing infection. The results of our series are illustrated by Fig. 1. 2.2. Prior attempts to eradicate infection Where patients were referred without any surgical intervention to eradicate infection and a two-stage procedure was subsequently performed at our institution, 24 (75%) of 32 were infection free. In the 14 cases where the referring hospital had attempted a single or two-stage procedure that failed prior to referral, only five were infection free (36%) after repeat two-stage revision at our institution. 2.3. Outcome of repeat two-stage revision surgery A subgroup of 19 patients had a repeat two-stage procedure. Five of these were from a failed initial two-stage at our institution and 14 were referred from other institutions having undergone two-stage revision prior to referral. In this series of 19 patients undergoing a repeat two-stage procedure, 42% (8/19) were infection free. Where infection was not eradicated, four patients had AKA, one patient is on long-term antibiotics, one is awaiting a further procedure and five patients went on to a third two-stage procedure, eradicating infection in three. Following an unsuccessful third two-stage procedure, the remaining two patients had an AKA. 2.4. Factors associated with outcomes From our initial hypothesis, we found that all members of the unsuccessful outcome group had at least one multidrug-resistant organism (P = 0.0002). Multiple organisms did have an effect, but this did not reach statistical significance (P = 0.056). There was no significant difference in gender (success 20 M/15, failure 7 M/9 F, P = 0.55); however, the unsuccessful group were significantly younger (mean failure 65, mean success 74, P = 0.012) 2.5. Microbiology Following the first-stage revision, an organism was cultured from intra-operative fluid or tissue in 88% of cases (45/51). No organisms were grown in six patients, one in the unsuccessful cohort and five in the successful cohort. Despite being culture negative, frank pus was present at the time of the first-stage procedure. Table 2 summarises the organisms that were cultured.

58

J. Stammers et al. / The Knee 22 (2015) 56–62

Fig. 1. Chart illustrating the origin, operations performed and outcome of the 51 infected arthroplasties.

The most commonly encountered organism was coagulase-negative staphylococcus (CNS) found in 27 (60%) patients. Multiple organisms were present in 67% (10/15) of the unsuccessful cohort compared to 37% (11/30) of those successfully treated (chi-squared = 3.658; P = 0.056). All members of the unsuccessful outcome group had at least one multidrug-resistant organism compared to 43% (16/30) in the successful cohort (chi-squared = 14.3; P = 0.0002).

2.6. Bone loss Table 3 outlines the bone loss prior to the second-stage revision procedure. There was an inter-observer correlation of 92%. Larger femoral and tibial bone loss was observed in unsuccessful cases compared to successful cases. Table 4 demonstrates the prostheses used in second-stage procedures.

2.7. Previous procedures prior to infection Eighteen patients had previous knee surgeries prior to infection in their knee arthroplasty with no effect on outcome.

2.8. Serological markers CRP and ESR were recorded prior to the second-stage and in those that had a repeat two-stage procedure. Box and whisker plots are shown in Graph 1a and b. Table 5 shows the sensitivity and specificity eradicating infection using a CRP b10 mmol/L and ESR b30 mm/hr prior to second stage and repeat second stage.

3. Discussion The overall success at eradicating infection in our series was 69%, lower than the 80–100% success often reported in the literature. The referred patients represent a series unique to tertiary units. In addition to in-house infections, it includes those from local units, patients highlighted by their local units as difficult to manage or supra-regional referrals, following failed attempts to eradicate infection. Patient choice results in an increased number of referrals to tertiary units with chronic prosthetic infections, reluctant to consider suppressive antibiotics, regular wound dressing, arthrodesis or amputation. Our results add to the current data in the literature about the chance of success treating deep chronic infections in this complex sub-series of all infected revisions. In agreement with a recent systematic review, it is difficult to make direct comparisons due to incomplete methodology, including selection criteria for treatment protocols and outcome parameters [2]. Reviewing the literature revealed 15 comparable series. [3,5,9,10,13, 15,24,25,27,31–36] (Table 6). All but one [9] were retrospective cohorts. The majority of prospective studies regarding two-stage revision have strict inclusion and exclusion criteria and therefore not applicable to our heterogeneous series of which the only exclusion criterion was Table 3 Bone loss subdivided by outcome. Femoral bone loss

Table 2 Cultured organisms. Organism

Total

Group B Streptococcus Group D Streptococcus Group G Streptococcus Streptococcus viridans Non-haemolytic streptococcus Streptococcus acidominimus Staphylococcus aureus MRSA Coagulase-negative staphylococcus Diphtheroids Enterococcus Pseudomonas Proteus

1 1 1 2 1 1 5 1 27 2 3 2 1

Successful Unsuccessful

Tibial bone loss

F1

F2a

F2b

F3

T1

T2a

T2b

T3

66% 12.5%

0% 19%

6% 6%

28% 62.5%

72% 37.5%

12.5% 12.5%

3% 12.5%

12.5% 37.5%

Table 4 Prosthesis used at latest follow-up. Prosthesis used

Number

Custom made SMILES Distal femoral replacement Proximal tibial replacement Total femoral replacement Off the shelf prostheses

26 5 1 1 11

J. Stammers et al. / The Knee 22 (2015) 56–62

59

Graph 1. (a) Box and whisker predicting successful eradication of infection based on a CRP b10. (b) Box and whisker predicting successful eradication of infection based on an ESR b30.

primary arthroplasty for previously infected fractures and for bone tumours. As in our series, only the data from Tsukayama type III and IV PJI were included. Two-stage revision combining these series Table 5 Sensitivity and specificity of CRP of b10 and ESR b30 predicting success of two-stage revision.

CRP ESR CRP pre-repeat two stage ESR pre-repeat two stage

Sensitivity

Specificity

73% 30% 56% 33%

47% 50% 43% 43%

eradicated infection in 806 knees of 1138 (70.8%, range 57–91%). Some studies excluded patients who did not proceed to the secondstage revision. In our series, these were included as they represent failure to achieve an infection free, functioning prosthesis. After a failed two-stage revision, chances of eradicating infection fall. We report the outcomes of one of the largest series of repeat two-stage revisions. Of the nineteen patients who remained infected after an initial two-stage revision, 8 (42%) were infection free with a repeat two-stage, and eleven (58%) patients overall had successful eradication of infection with multiple two-stages. Multiple-stage revisions are a significant undertaking, both surgically and for the patient, with worse outcomes including a high risk of AKA (25%).

60

J. Stammers et al. / The Knee 22 (2015) 56–62

Table 6 Review of comparable papers. Paper Hart and Jones (2006) Mittal et al. (2007) Mont et al. (2000) Rasul et al. (1991) Haddad et al. (2000) Hirakawa et al. (1998) Mortazavi et al. (2011) Zmistowski et al. (2011)

Sherrell et al. (2011) Kurd et al. (2010) Westrich et al. (2010) Goldman et al. (1996) Segawa et al. (1999) Mortazavi et al. (2010) Hoad-Reddick et al. (2005) This series

Success/overall 42/48 (88%) 28/37 (76%) 61/69 (88%) 15/19 (78%) 44/48 (92%) 41/55 (75%) 84/117 (72%) 129/228 (57%)

55/83 (66%) 75/102 (74%) 68/75 (91%) 58/64 (91%) 24/29 (83%) 72/91 (79%) 34/38 (89%) 35/51 (69%)

There is limited published data on repeat two-stage revision [15,23, 21,20,22] with success reported in 58 of 103 (56%, range 4%–100%) repeat two-stage revision procedures based on primary data or subseries analysis. The small sample sizes, discrepancies in the definition of a successful outcome and individual institution or surgeon indication for repeat two-stage are of limited value for comparisons to this study. Polymicrobial infections were associated with unsuccessful outcomes, but the difference was not statistically significant (P = 0.056). Published series share reduced success treating polymicrobial infections between 0% (0/8) [25] and 77% [36] compared to single organisms in 69% (65/94) and 96% respectively. Sixty-seven percent of cases in our series grew at least one multidrug-resistant organism, higher than most of the published studies. Multidrug resistance in our series is significantly (P = 0.0002) associated with an unsuccessful outcome in agreement with several studies [4, 5,31]. Conversely, Westrich et al. [32] found that the incidence of multidrug-resistant infections were rising, but there was no significant difference in outcome based on 75 knees. Our study failed to find any such association between previous knee surgery and outcome of two-stage revision. However, a high proportion of these patients were seen and often treated at other trusts, and therefore, we may have underestimated what treatment had been carried out prior to referral as our data collection often came from a sole referral letter. A number of studies [3,14,15,27] have suggested previous knee surgeries carry a higher failure rate from a two-stage revision, particularly following failed irrigation and debridement. Using a proposed cutoff of a CRP less than 10 and an ESR less than 30 did not reliably predict success. The median CRP was in fact higher in the successful than the unsuccessful cohort with wide variability and crossover of the 50th centiles. The sensitivity and specificity of CRP and ESR prior to repeat two-stage is again unreliable, more so than the initial two-stage revision. This suggests that the timing of reimplantation should not be guided by a single test and that serological markers should be correlated with clinical findings; however, the absence of falling inflammatory markers may suggest antibiotic resistance or residual infected tissue after the first-stage. A number of authors [37–39] have similarly investigated the role of serological testing between stages and were also unable to identify any patterns to reliably diagnose persistent infection or success following the second stage. It was hypothesized that younger patients would be associated with increased chances of eradicating infection. The significantly lower rate of success in young patients could be the association with medical comorbidities that have an earlier age of arthroplasty and increased risk of wound infection, such as rheumatoid arthritis or obesity. Institutions

Series characteristics

Mean FU months

23% polymicrobial Multiresistant subseries

48.5 (26–85) 51 (24–111) 63 (36–114) 42 (24–120) 48 (20–112) 42.4 (19–96) 45.6 (24–113)

Deep infection subseries 56% MDR 27% MDR 34% culture −ve Subseries 42% MDR 4% polymicrobial Previous irrigation and debridement 30% MDR 15% culture −ve 45% MDR 8% polymicrobial 11% polymicrobial Subseries 27% MDR 69% MDR, 47% multiple organisms

36 (12–85)

39 (12–104) 34.5 (24–90) 52.4 (24–108) 90 (24–204) 48 (3.6–168) 65 (25–159) 54 (24–114) 43 (24–59)

maybe more likely to refer younger patients with infected knee arthroplasty, particularly where prior attempts to eradicate infection have failed, than elderly patients due to patient choice and expectations. Expedient treatment is essential to limit the time where the implant and tissues are exposed to infection. Any increased time increases the risk of multidrug resistance and multiple organisms which are associated with worse outcomes. [4,5,9,20] The presence of histology confirmed osteomyelitis following the first-stage and late referral following previous failed treatment suggests prolonged exposure to infection but due to incomplete records we were unable to analyse if early or late referrals affected success of two-stage revision. Table 7 summarises the series of patients in whom infection was not eradicated. Two of these patients had undergone total knee replacement after previous septic arthritis. In these cases, which are known to have a higher infection rate, we favour staged primary arthroplasty resecting soft tissue and bone with insertion of an antibiotic laden cement spacer. After a period of antibiotics, the primary prosthesis is implanted. Three patients were referred from other tertiary revision units for second opinions as they were reluctant to consider amputation or arthrodesis offered to them. Two patients did not have the planned second-stage procedure but underwent AKA. One had a periprosthetic fracture around the cement spacer with non-healing wounds due to lymphedema, and one had MRSA bacteraemia and persisting infection despite recurrent debridement and antibiotic regimes. This study has some limitations. It is retrospective in design, and due to multiple surgeons and microbiologist variability, there may well be some additional bias and variability. This paper does not aim to demonstrate a specific method to manage these patients due to the accepted patient heterogeneity and individualised treatment regimens but illustrates the challenges and multiple variables. Standardising and stratifying patient factors, pre-referral attempts to eradicate infection and extent of infection at point of referral may enable comparisons with other published series. Prospective studies controlling some of these variables may optimise treatment strategies. The follow-up of 43 months is relatively short and potentially patients could be re-infected after this time. However, Mont et al. [9] reported that most recurrences occur in the first post-operative year, and Haleem et al. [40] found that infection with the same organism commonly occurred in the first year, whereas infection after one year tended to be due to a new organism. Goldman et al. [33] reported reinfection occurring on average 1.9 years post re-implantation. The diagnosis of prosthetic joint infection can be challenging. Newer techniques are being introduced including leukocyte esterase reagent strips, synovial fluid inflammatory markers and polymerase chain

J. Stammers et al. / The Knee 22 (2015) 56–62

61

Table 7 Summary of failures. Summary of operations

Comorbidities

Previous surgery

Microbiology

Multidrug resistance

Bone loss

Outcome

Primary elsewhere, ×2 two-stage our institution Primary elsewhere, two-stage elsewhere, repeat two-stage our institution Local primary, two-stage our institution Primary elsewhere, two-stage elsewhere, ×2 two-stage our institution Primary elsewhere, two-stage elsewhere, ×2 two-stage our institution Primary elsewhere, two-stage elsewhere, two-stage our institution Primary elsewhere, two-stage elsewhere, two-stage our institution Local primary, two-stage our institution Primary elsewhere, two-stage our institution

HTN, IHD

No

Non-haemolytic strep, CNS

Yes

F3/T3

AKA

Septic arthritis, IHD

No

Group B Streptococcus

Yes

F3/T3

AKA

HTN, IHD

No

CNS, MRSA

Yes

F1/T1

AKA

RA on steroids

No

Enterococcus, CNS

Yes

F3/T2B

AKA

HTN

No

CNS, proteus

Yes

F3/T3

AKA

RA, Ankle abscess ipsilateral leg, MRSA No bacteraemia

MRSA, pseudomonas, acinetobacter

Yes

F3/T2B

AKA

Lymphoedema, PVD, high BMI

No

Citrobacter, enterobacter, serratia

Yes

T3/T3

AKA

RA, on methotrexate and steroids

No

CNS

Yes

F2A/T1

Septic arthritis post arthroscopy, alcoholic

No

Yes

F3/T2a

Primary elsewhere, two-stage our institution

Nil

Yes

F1/T1

Long-term antibiotics

Primary elsewhere, two-stage our institution Primary elsewhere, two-stage our institution

High BMI, wheelchair bound, PVD

Yes-fall dislocated patella, exploration and lateral release No

Citrobacter, enterobacter, serratia, staph haemolyticus CNS

Long-term antibiotics Long-term antibiotics

No growth

N/A

F2A/T1

Long-term antibiotics Long-term antibiotics

Primary elsewhere, two-stage our institution Primary elsewhere, ×2 two-stage our institution Primary elsewhere, two-stage elsewhere, two-stage our institution Primary elsewhere, two-stage our institution

Type II diabetes, fractured neck of femur prior to planned repeat twostage HTN, depression

Yes, revision for instability

Pseudomonas

Yes

F2B/T3

No

CNS

Yes

F2A/T1

Anaemia, stroke, high BMI

No

Proteus

Yes

F3/T3

Nil

No

Group G streptococcus, CNS Yes

F3/T1

Long-term antibiotics Long-term antibiotics Fail-on list

Chemotherapy

No

CNS, E coli

F3/T2A

Fail-on list

Yes

Legend: HTN—hypertension, IHD—ischaemic heart disease, RA—rheumatoid arthritis, MRSA—methicillin-resistant staphylococcus aureus, CNS—coagulase-negative staphylococcus aureus, AKA—above knee amputation, BMI—body mass index, PVD—peripheral vascular disease.

reaction (PCR) identification [41–43]. These tests were not routinely used during the study period. It may be that infection is being underdiagnosed. Whilst the presence of infection is a binary answer, there is a spectrum of disease. This series is made from those at one extreme where diagnosis is rarely questioned, often due to the presence of pus. These tests equally could be applied to testing that infection is eradicated prior to the second stage knowing that inflammatory markers are inconsistent with diagnosis. PCR may reduce the number of culture-negative results and improve antibiotic decision making. Mont et al. [44] suggested that obtaining repeat microbiology samples prior to second-stage re-implantation after stopping antibiotics is of value; however, other studies indicate that there are there are many sampling errors, high false-negatives and poor sensitivity. Single stage revision has published success between 83–100% [45, 11,46,47]. There is limited evidence comparing single to two-stage revision. Most institutions reserve single stage revision for infections with minimal bone loss, healthy soft tissues, non-immunocompromised patients, single organism infections with known, non-resistant sensitivities or no previous attempts to eradicate infection. [48] The majority of the patients in this series would not be eligible. Arthrodesis is a viable alternative, particularly where the soft tissues are compromised particularly an irreparable extensor mechanism. Overall outcome scores comparing arthrodesis to arthroplasty are similar but physical outcome scores are lower [49]. Suppressive long-term antibiotics or amputation

remain last resort options. Some patients report improved quality of life following amputation following many years and multiple failed attempts to eradicate infection. Particularly following a failed twostage revision, this series includes patients because they are not happy to accept regular wound dressings, long-term antibiotics or amputation and would accept a 50% chance of limb salvage. 4. Conclusion Eradicating chronic infection in total knee arthroplasty presents a challenge to revision units. The best opportunity to eradicate infection is in the early stages. In this series, failure is associated with multidrug-resistant organisms and a less favourable host response. Despite extensive bone and soft tissue debridement with access to endoprostheses for second-stage reconstruction, higher rates of failure are found in this series than in the literature. Repeat two-stage revision can offer the patient a chance of eradicating infection but our data suggests that success falls with each previous failed attempt. Series from other tertiary units with a high proportion of multidrugresistant organisms or where previous attempts to eradicate infection have failed have similar rates of success. Due to tertiary referral patterns within the NHS, our institution is often the last resort for many challenging cases, particularly where patients refuse to accept arthrodesis, amputation or long-term antibiotics.

62

J. Stammers et al. / The Knee 22 (2015) 56–62

Acknowledgement The authors thank Greg Jordan for his statistical support and Mr. Sabur Malek for classifying the bone loss. References [1] National Joint Registry for England, Wales. 10th Annual Report. The National Joint Registry; 2013. [2] Jamsen E, Stogiannidis I, Malmivaara A, Pajamaki J, Puolakka T, Konttinen YT. Outcome of prosthesis exchange for infected knee arthroplasty: the effect of treatment approach. Acta Orthop 2009;80:67–77. [3] Hart WJ, Jones RS. Two-stage revision of infected total knee replacements using articulating cement spacers and short-term antibiotic therapy. J Bone Joint Surg Br 2006;88:1011–5. [4] Kilgus DJ, Howe DJ, Strang A. Results of periprosthetic hip and knee infections caused by resistant bacteria. Clin Orthop Relat Res 2002;116–24. [5] Mittal Y, Fehring TK, Hanssen A, Marculescu C, Odum SM, Osmon D. Two-stage reimplantation for periprosthetic knee infection involving resistant organisms. J Bone Joint Surg Am 2007;89:1227–31. [6] Dixon T, Shaw M, Ebrahim S, Dieppe P. Trends in hip and knee joint replacement: socioeconomic inequalities and projections of need. Ann Rheum Dis 2004;63: 825–30. [7] Kurtz S, Ong K, Lau E, Mowat F, Halpern M. Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint Surg Am 2007;89:780–5. [8] Burger RR, Basch T, Hopson CN. Implant salvage in infected total knee arthroplasty. Clin Orthop Relat Res 1991;105–12. [9] Mont MA, Waldman BJ, Hungerford DS. Evaluation of preoperative cultures before second-stage reimplantation of a total knee prosthesis complicated by infection. A comparison-group study. J Bone Joint Surg Am 2000;82-A:1552–7. [10] Rasul Jr AT, Tsukayama D, Gustilo RB. Effect of time of onset and depth of infection on the outcome of total knee arthroplasty infections. Clin Orthop Relat Res 1991; 98–104. [11] Goksan SB, Freeman MA. One-stage reimplantation for infected total knee arthroplasty. J Bone Joint Surg Br 1992;74:78–82. [12] Silva M, Tharani R, Schmalzried TP. Results of direct exchange or debridement of the infected total knee arthroplasty. Clin Orthop Relat Res 2002;125–31. [13] Haddad FS, Masri BA, Campbell D, McGraw RW, Beauchamp CP, Duncan CP. The PROSTALAC functional spacer in two-stage revision for infected knee replacements. Prosthesis of antibiotic-loaded acrylic cement. J Bone Joint Surg Br 2000;82:807–12. [14] Wilson MG, Kelley K, Thornhill TS. Infection as a complication of total kneereplacement arthroplasty. Risk factors and treatment in sixty-seven cases. J Bone Joint Surg Am 1990;72:878–83. [15] Hirakawa K, Stulberg BN, Wilde AH, Bauer TW, Secic M. Results of 2-stage reimplantation for infected total knee arthroplasty. J Arthroplasty 1998;13:22–8. [16] Falahee MH, Matthews LS, Kaufer H. Resection arthroplasty as a salvage procedure for a knee with infection after a total arthroplasty. J Bone Joint Surg Am 1987;69: 1013–21. [17] Knutson K, Lindstrand A, Lidgren L. Arthrodesis for failed knee arthroplasty. A report of 20 cases. J Bone Joint Surg Br 1985;67:47–52. [18] Segreti J, Nelson JA, Trenholme GM. Prolonged suppressive antibiotic therapy for infected orthopedic prostheses. Clin Infect Dis 1998;27:711–3. [19] Sierra RJ, Trousdale RT, Pagnano MW. Above-the-knee amputation after a total knee replacement: prevalence, etiology, and functional outcome. J Bone Joint Surg Am 2003;85-A:1000–4. [20] Azzam K, McHale K, Austin M, Purtill JJ, Parvizi J. Outcome of a second two-stage reimplantation for periprosthetic knee infection. Clin Orthop Relat Res 2009;467: 1706–14. [21] Hanssen AD, Trousdale RT, Osmon DR. Patient outcome with reinfection following reimplantation for the infected total knee arthroplasty. Clin Orthop Relat Res 1995;55–67. [22] Maheshwari AV, Gioe TJ, Kalore NV, Cheng EY. Reinfection after prior staged reimplantation for septic total knee arthroplasty: is salvage still possible? J Arthroplasty 2010;25:92–7. [23] Backe Jr HA, Wolff DA, Windsor RE. Total knee replacement infection after 2-stage reimplantation: results of subsequent 2-stage reimplantation. Clin Orthop Relat Res 1996;125–31. [24] Mortazavi SM, Vegari D, Ho A, Zmistowski B, Parvizi J. Two-stage exchange arthroplasty for infected total knee arthroplasty: predictors of failure. Clin Orthop Relat Res 2011;469:3049–54.

[25] Zmistowski B, Fedorka CJ, Sheehan E, Deirmengian G, Austin MS, Parvizi J. Prosthetic joint infection caused by gram-negative organisms. J Arthroplasty 2011;26:104–8. [26] Fehring TK, Odum SM, Berend KR, Jiranek WA, Parvizi J, Bozic KJ, et al. Failure of irrigation and debridement for early postoperative periprosthetic infection. Clin Orthop Relat Res 2013;471:250–7. [27] Sherrell JC, Fehring TK, Odum S, Hansen E, Zmistowski B, Dennos A, et al. The Chitranjan Ranawat Award: fate of two-stage reimplantation after failed irrigation and debridement for periprosthetic knee infection. Clin Orthop Relat Res 2011; 469:18–25. [28] Parvizi J, Gehrke T, Chen AF. Proceedings of the International Consensus on Periprosthetic Joint Infection. Bone Joint J 2013;95-B:1450–2. [29] Tsukayama DT, Goldberg VM, Kyle R. Diagnosis and management of infection after total knee arthroplasty. J Bone Joint Surg Am 2003;85-A(Suppl. 1):S75–80. [30] Engh GA, Ammeen DJ. Bone loss with revision total knee arthroplasty: defect classification and alternatives for reconstruction. Instr Course Lect 1999;48:167–75. [31] Kurd MF, Ghanem E, Steinbrecher J, Parvizi J. Two-stage exchange knee arthroplasty: does resistance of the infecting organism influence the outcome? Clin Orthop Relat Res 2010;468:2060–6. [32] Westrich GH, Walcott-Sapp S, Bornstein LJ, Bostrom MP, Windsor RE, Brause BD. Modern treatment of infected total knee arthroplasty with a 2-stage reimplantation protocol. J Arthroplasty 2010;25:1015–21 [21 e1-2]. [33] Goldman RT, Scuderi GR, Insall JN. 2-stage reimplantation for infected total knee replacement. Clin Orthop Relat Res 1996;118–24. [34] Segawa H, Tsukayama DT, Kyle RF, Becker DA, Gustilo RB. Infection after total knee arthroplasty. A retrospective study of the treatment of eighty-one infections. J Bone Joint Surg Am 1999;81:1434–45. [35] Mortazavi SM, Schwartzenberger J, Austin MS, Purtill JJ, Parvizi J. Revision total knee arthroplasty infection: incidence and predictors. Clin Orthop Relat Res 2010;468: 2052–9. [36] Hoad-Reddick DA, Evans CR, Norman P, Stockley I. Is there a role for extended antibiotic therapy in a two-stage revision of the infected knee arthroplasty? J Bone Joint Surg Br 2005;87:171–4. [37] Kusuma SK, Ward J, Jacofsky M, Sporer SM, Della Valle CJ. What is the role of serological testing between stages of two-stage reconstruction of the infected prosthetic knee? Clin Orthop Relat Res 2011;469:1002–8. [38] Ghanem E, Azzam K, Seeley M, Joshi A, Parvizi J. Staged revision for knee arthroplasty infection: what is the role of serologic tests before reimplantation? Clin Orthop Relat Res 2009;467:1699–705. [39] Azzam KA, Seeley M, Ghanem E, Austin MS, Purtill JJ, Parvizi J. Irrigation and debridement in the management of prosthetic joint infection: traditional indications revisited. J Arthroplasty 2010;25:1022–7. [40] Haleem AA, Berry DJ, Hanssen AD. Mid-term to long-term followup of two-stage reimplantation for infected total knee arthroplasty. Clin Orthop Relat Res 2004; 35–9. [41] Jacovides CL, Kreft R, Adeli B, Hozack B, Ehrlich GD, Parvizi J. Successful identification of pathogens by polymerase chain reaction (pcr)-based electron spray ionization time-of-flight mass spectrometry (ESI-TOF-MS) in culture-negative periprosthetic joint infection. J Bone Joint Surg Am 2012;94:2247–54. [42] Parvizi J, Jacovides C, Antoci V, Ghanem E. Diagnosis of periprosthetic joint infection: the utility of a simple yet unappreciated enzyme. J Bone Joint Surg Am 2011;93: 2242–8. [43] Bedair H, Ting N, Jacovides C, Saxena A, Moric M, Parvizi J, et al. The Mark Coventry Award: diagnosis of early postoperative TKA infection using synovial fluid analysis. Clin Orthop Relat Res 2011;469:34–40. [44] Lonner JH, Siliski JM, Della Valle C, DiCesare P, Lotke PA. Role of knee aspiration after resection of the infected total knee arthroplasty. Am J Orthop (Belle Mead NJ) 2001; 30:305–9. [45] Moyad TF, Thornhill T, Estok D. Evaluation and management of the infected total hip and knee. Orthopedics 2008;31:581–8 [quiz 89–90]. [46] Masters JP, Smith NA, Foguet P, Reed M, Parsons H, Sprowson AP. A systematic review of the evidence for single stage and two stage revision of infected knee replacement. BMC Musculoskelet Disord 2013;14:222. [47] Gulhane S, Vanhegan IS, Haddad FS. Single stage revision: regaining momentum. J Bone Joint Surg Br 2012;94:120–2. [48] Gehrke T, Zahar A, Kendoff D. One-stage exchange: it all began here. Bone Joint J 2013;95-B:77–83. [49] Klinger HM, Spahn G, Schultz W, Baums MH. Arthrodesis of the knee after failed infected total knee arthroplasty. Knee Surg Sports Traumatol Arthrosc 2006;14: 447–53.