Eur J Orthop Surg Traumatol (2014) 24:1279–1283 DOI 10.1007/s00590-013-1380-1
ORIGINAL ARTICLE
Two-stage revision for infected total knee arthroplasty: our experience with interval prosthesis N. Prasad • V. Paringe • R. Kotwal A. Ghandour • R. Morgan Jones
•
Received: 23 August 2013 / Accepted: 22 November 2013 / Published online: 5 December 2013 Ó Springer-Verlag France 2013
Abstract Introduction To review our practice of performing twostage revision for infected total knee arthroplasty using articulating interval prosthesis and to compare the incidence of the recurrence of infection and re-operation rate in patients undergoing two-stage revision as planned with the group of patients who choose not to proceed to the second stage. Method This study is a retrospective review of 60 consecutive patients undergoing a two-stage revision for infected total knee arthroplasty using articulating interval prosthesis. All cases managed by a single surgeon using a uniform peri-operative protocol, and short-course parenteral antibiotic therapy. Result Thirty-four patients (57 %) (Group 1) underwent the two-stage revision as planned. However, twenty-six patients (43 %) (Group 2) opted not to have a second-stage procedure as the first-stage and interval prosthesis had eradicated the infection, resolved the pain and achieved good functional outcome. There were five cases of recurrent infection in the 60 patients (8 %) at a mean follow-up 5 years. In those completing the two-stage revision, two patients had recurrent infection. Of the patients who retained the interval prosthesis, there were three recurrent infections. There was no statistically significant difference between the groups in terms of recurrence of infection or re-revision.
N. Prasad V. Paringe R. Kotwal A. Ghandour R. M. Jones Cardiff and Vale NHS Trust, Cardiff, UK N. Prasad (&) 5 Lovage Close, Pontprennau, Cardiff CF23 8SB, UK e-mail: [email protected]
Conclusion Two-stage revision with interval prostheses represents a safe and reliable method of treating infected knee prosthesis; however, there may be a role for one-stage revision in selected cases. Keywords Knee revision Infection Two-stage revision Interval prosthesis
Introduction Management of deep infection following joint replacement surgery in orthopaedics remains challenging. The incidence of infection following primary total knee replacement is between 1 and 2 % with an increased rate in rheumatoid patients [1–3]. This leads to significant morbidity to the patient. Two-stage revision surgery proposed by Insall [4] is considered as the ‘gold standard’ for the treatment of infected total knee replacement and can be performed by different techniques, using static or articulating spacers during the first stage. The interim use of an antibioticimpregnated static cement spacer has been reported to successfully eradicate infection in more than 90 % of infected total knee arthroplasty (TKA) [5–8] Two-stage revision using an articulating spacer have also shown good clinical outcome with successful eradication of infection [9, 10]. The senior author has been performing two-stage revision procedures for infected total knee replacement using interval prosthesis as articulating spacer, and over years it has come to his attention that good number of patients do not proceed for second stage as the infection is cured following first stage, and they get a reasonable function with articulating spacer. The aim of our study was (1) to review
123
1280
our practice of performing two-stage revision for infected total knee arthroplasty using an articulating interval prosthesis and short-course parenteral antibiotic therapy by a single surgeon in a tertiary orthopaedic unit, (2) to compare the incidence of the recurrence of infection in patients undergoing two-stage revision as planned with the group of patients who choose not to proceed to the second stage and (3) to find the incidence of re-operation rate for all causes in both the groups.
Patients and methods We identified a consecutive series of 70 patients operated between 2001 and 2008 for infected total knee replacement in our hospital. We carried out a clinical and radiological review on these patient’s case notes. The inclusion criteria for the study were patients with chronically infected primary total knee replacement diagnosed clinically and radiologically with or without proven bacteriology from previous knee aspirate or arthroscopic knee biopsy. We also included patients who had exchange of liner for acute infection following primary total knee replacement subsequently presented as chronic infection. We excluded patients who has already had two-stage revision surgery for infection elsewhere and was referred to our unit, as there was further recurrence of infection. We also excluded patients who had revision surgery for aseptic loosening and subsequently referred to our unit as they became infected. A total 60 patients were considered for the study. There were 32 males and 28 female patients. The mean age at the time of first-stage surgery was 66 years (45–83 years). Four patients were dead at the time of final review, so the last available data were used for analysis. We divided the patients into two groups; patients who had planned two-stage revision (Group 1) and the second group was made up of patients who declined second stage and decided to live on with the interval prosthesis (Group 2). We looked at the subsequent recurrence of infection and also re-revision surgery for infection or any other cause. Statistical analysis was performed using v2 test to compare the recurrence of infection and re-revision surgery between the groups.
Eur J Orthop Surg Traumatol (2014) 24:1279–1283
Multiple tissue samples (minimum of 5) are sent for microbiological analysis. Sequential and repeated intramedullary reaming, lavage and curettage were used to achieve satisfactory debridement prior to insertion of the interval prosthesis. We consider the debridement at this stage as the most important step as this facilitates removal of the biofilm. Primary condylar femoral components, both posterior stabilised (PS) and cruciate retaining (CR) and the rotating platform polyethylene tibial insert without the metal tray were used to construct the interval prosthesis. The rotating platform polyethylene insert is chosen as the ‘keel’ and this allowed greater stability. Occasionally, in cases of significant bone loss, additional support for the interval prosthesis was required using a femoral stem and tibial tray and stem (all components from DePuy/J&J). In all cases, maximal ligament balance was achieved using a combination of femoral and tibial components and the cement mantle. The components were inserted using antibiotic-loaded cement (Palacos R; Zimmer, UK). A double mix of Palacos cement with gentamicin plus 1 gm of vancomycin per mix was used in all cases. Cement was not maximally pressurised and used late in the cure phase, to facilitate easy removal during the second-stage procedure. The cement mantle filled all ‘dead space’ between implants and bone interface (Fig. 1a, b). Patients were given intravenous teicoplanin for 5 days and then converted to dual oral antibiotics, depending upon the culture and sensitivities after discussing with the microbiologist, for 6 weeks. If no organisms were identified from the multiple intraoperative tissue samples, oral antibiotics were given based on previous positive results from the hospital of referral. Once established on oral antibiotics, the patients were discharged home. Patients were allowed to fully weight bear, with unrestricted range of movement, following first-stage surgery. No knee bracing was used post-operatively. The patients were monitored clinically and biochemically with inflammatory markers, CRP and ESR checked regularly. The second stage is planned at 6 months whenever possible. A 6-month interval is preferred to allow the soft tissue envelope to reach a quiescent, healed state and maximise mobility and function of the knee and patient. A prolonged interval also allows greater confidence that infection has been eradicated.
Operative technique The surgical technique followed was standardised and identical for all the patients. The same senior surgeon performed all surgeries. A tourniquet was used in all the cases. During the first stage, a thorough debridement is performed after removal of loose infected components and bone cement. Lautenbach compartmental debridement techniques [11] were used to debride the femur and tibia.
123
Results In our series of 60 patients, 34 patients (57 %) had a planned second-stage procedure a mean 6 months following the first stage. The remaining 26 patients (43 %) opted not to have the second-stage procedure, as they were infection free, functionally satisfied and pain free with the
Eur J Orthop Surg Traumatol (2014) 24:1279–1283
1281
Fig. 1 Post-operative radiographs showing the interval prosthesis in situ (AP and lateral). Note the antibioticloaded cement mantle fills all the dead space between prosthesis and bone interface
Table 1 List of organisms isolated from biopsy sample during firststage revision knee arthroplasty Organisms
Numbers
Coagulase-negative Staph aureus
29
Staph aureus
7
MRSA
5
Alpha-haemolytic streptococci
4
Serratia Coliforms
2 3
Beta-haemolytic streptococci
2
Pseudomonas
2
Gram-negative cocci
2
Gram-positive Bacillus
1
Enterococci
1
Candida
1
Microaerophilic streptococci
1
No growth
6
interval prosthesis. The predominant organism in our series was coagulase-negative Staph aureus (Table 1). Of the 34 patients who had two-stage procedure, there were 18 males and 16 females. The mean age at the time of first stage was 69 years (Table 2). The mean duration between first-stage interval prosthesis and definitive second-stage procedure was 26 weeks (range 12–48 weeks). The mean follow-up after second-stage procedure was 60 months (range 24–96 months). The range of motion was more than 90° in 30 out of 34 patients. The mean flexion
Table 2 Patient demographics with follow-up details in both the groups Type of surgical procedure
Male: female
Age mean (range) (years)
Follow-up mean (range) (months)
Group 1
Two stage (n = 34)
18:16
69 (53–81)
60 (24–96)
Group 2
First stage only (n = 26)
18:8
63 (55–82)
48 (24–64)
deformity at the time of final follow was 4° (0–10), and the mean flexion was 98° (70–115). All of the knees were clinically stable. There was recurrence of infection in two patients requiring further surgery. In the group of patients who declined second stage (n = 26), 4 patients had deceased at the time of this study, without recurrent infection. The mean age at the time of surgery was 61 years. The mean follow-up was 48 months (range 24–64 months; Table 2). The shorter follow-up reflecting patients later in the series when confidence in retaining the interval prosthesis were greater. The range of motion was more than 90° in 19 out of 21 patients. The mean flexion deformity was 5° (range 0–10), and the mean flexion was 95° (range 80–115). In this group, five patients required re-operation. Two patients had revision surgery at 14 and 18 months due to instability, and three patients had late recurrence of infection which required revision surgery at 17, 22 and 26 months following insertion of the interval prosthesis.
123
1282
Eur J Orthop Surg Traumatol (2014) 24:1279–1283
Table 3 Recurrence infection and re-revision rate in two groups
Recurrence of infection number (%)
Further revision procedure for infection number (%)
Further revision surgery for reasons other than infection number (%)
Total re-revision procedure number (%)
2
2
0
2
First stage only (n = 26)
3
3
2
5
p value (Chi test)
0.43
0.43
0.07
0.11
Two stage (n = 34)
Of the total of five recurrent infections, four underwent a further revision with eradication of infection. The last patient underwent an arthrodesis due to a compromised soft tissue envelope, but remains infected. The statistical analysis did not show any difference in the incidence of recurrence of infection or re-revision between the two groups. There was no re-revision in the two-stage group for non-infective cause, and on comparison the one-stage group had two re-revision for instability; however, statistical analysis did not show any significant difference although the p value was very close to 0.05 (p = 0.07; Table 3).
Discussion A two-stage revision total knee arthroplasty using a static spacer can lead to problems when it comes to re-implantation in the second stage in the form of stiffness and loss of bone stock [14] This led to the evolution of use of articulating spacers, and its usage has been reported [9, 10, 12, 13]. Hofmann et al. [12] described technique in which the femoral component is removed, autoclaved and reinserted along with a new polyethylene tibia. In his series of 55 patients, there was 12 % recurrence of infection at 73 months [13]. They allow only partial weight-bearing mobilisation after stage 1. Two-stage revision using a PROSTALAC prosthesis as an articulating interval prosthesis has shown an overall recurrence rate of 9 % in a series of 45 patients with a mean follow-up of 48 months [9]. PROSTALAC spacers are expensive and are discarded after 6 weeks. In our series, the overall rate of recurrent infection following revision and insertion of an interval prosthesis is 8 % (five recurrences, two in the two-stage group and three in the retained interval prosthesis group). The advantages of using interval prosthesis as an articulating spacer are many. The interval prosthesis provides stability and balance of the ligaments and soft tissue envelope, allowing free movement and full weight bearing. Improved patient mobility allows for a planned second
123
stage at a time when the patient has physiologically recovered from surgery and the limb has become quiescent with an improved soft tissue envelope. This gives the surgeon and the patient flexibility in the timing of the second stage. The subsequent revision surgery is technically easier as mobility is improved and fibrotic scarring less. We also note that the space-filling cement mantle provides a ‘macro-lock’ into the bone interface, thus providing bone-cement stability and preventing progressive bone loss. The cement mantle plus choosing an appropriately sized femoral and tibial implant allows a balanced construct with appropriately tensioned collateral ligaments. This avoids progressive soft tissue damage resulting from knee instability and allows the soft tissue envelope to recover following eradication of infection and debridement. In our series, 43 % of patient did not require the second stage, supporting the possible role for one-stage revision of infected total knee replacement. In our series, there was no difference in the incidence of recurrence of infection between two-stage group and one-stage group. We had two cases of re-revision in one-stage (interval prosthesis) group for instability, and this is expected as the interval prosthesis was used as temporary step before definitive surgery. A one-stage revision using a definitive prosthesis could theoretically avoid this. One-stage revision is well accepted in total hip replacement [15–17]. The advantages of performing a one-stage revision are clinical (single operation, shorter hospitalisation, reduced medication), functional (improved movement, shorter rehabilitation, reduced disability) and significant economic savings. We have started performing one-stage revision in selected cases drawing experience from our two-stage revision study. The preliminary result of one-stage revision knee arthroplasty was presented in the British Association for Surgery of the Knee Spring Meeting [18]. Recent literature search yields encouraging results for one-stage revision knee arthroplasty for infection [19, 20] although more research is required. The recently concluded international consensus on peri-prosthetic joint infection supports our views on single-stage excision arthroplasty for infection. There was
Eur J Orthop Surg Traumatol (2014) 24:1279–1283
strong agreement that a single-stage revision arthroplasty for infection may be performed in circumstances where sensitive antibiotics where available and patient is not in sepsis. Lack of identification of organism, presence of sinus tract and severe soft tissue damage requiring flap coverage are considered as contraindication for single-stage revision arthroplasty in peri-prosthetic joint infection [21]. The infecting microorganism secretes a matrix of extracellular polymeric substance, and this along with the bacteria forms what is known as ‘biofilms’, which are inherently resistant to antimicrobials, and forms the root cause for chronic infection[22]. The initial debridement allows the surgeons to remove these ‘biofilms’. We believe that initial debridement performed during the removal of infected prosthesis is the most important step, and it is surgeons choice that whether you perform this in two stage or one stage. We acknowledge some of the limitations of our study. Firstly, this is a retrospective non-randomised study. Secondly, 43 % of patients opted not to have the second stage, and this has been purely patient’s choice. This in fact reiterates the argument that interval prosthesis provides good function and subsequently would help to achieve good results after a two-stage procedure. Thirdly, we did not include the results of functional scoring at final follow-up, as our study aim was to look at the recurrence rate of infection.
Conclusion Two-stage revision with an interval prosthesis using an articulating spacer gives low recurrence of infection with good knee range of motion and allows the patient to mobilise full weight bearing during and after the surgery. We would also recommend that such cases should be referred to surgeons with appropriate training and expertise to minimise recurrence and avoid the poor outcomes associated with multiple revisions. The senior author suggests that recurrent infected knee revisions should be referred to a regional centre of expertise to give patients the ‘best chance’ of successful eradication of infection and return of function. Onestage revision in selected case may be way forward; however, further research is needed on this. Conflict of interest
None.
References
1283 2. Bengston S, Knutson K, Lidgren L (1989) Treatment of infected knee arthroplasty. Clin Orthop Relat Res 245:173–178 3. Blom AW, Brown J, Taylor AH, Pattison G, Whitehouse S, Bannister GC (2004) Infection after total knee arthroplasty. J Bone Joint Surg Br 86(B):688–691 4. Insall JN, Thompson FM, Brause BD (1983) Two-stage reimplantation for the salvage of infected total knee arthroplasty. J Bone Joint Surg Am 65(8):1087–1098 5. Wilde AH, Ruth JT (1988) Two-stage reimplantation in infected total knee arthroplasty. Clin Orthop 236:23 6. Goldman RT, Scuderi GR, Insall JN (1996) Two-Stage reimplantation for infected total knee replacement. Clin Orthop 331:118 7. Simmons TD, Stern SH (1996) Diagnosis and management of the infected total knee arthroplasty. Am J Knee Surg 2:99 8. Windsor RE, Insall JN, Urs WK et al (1990) Two-stage reimplantation for the salvage of total knee arthroplasty complicated by infection. Further follow-up and refinement of indications. J Bone Joint Surg Am 72:272 9. Haddad FS, Masri BA, Campbell D, McGraw RW, Beauchamp CP, Duncan CP (2000) The PROSTALAC functional spacer in two-stage revision for infected knee replacements. Prosthesis of antibiotic-loaded acrylic cement. J Bone Joint Surg Br 82: 807–812 10. Trezies A, Parish E, Dixon P, Cross M (2006) The use of an articulating spacer in the management of infected total knee arthroplasties. J Arthroplasty 21(5):702–704 11. Weber FA, Lautenbach EEG (1986) Revision of infected total hip arthroplasty. Clin Orthop 211:108–115 12. Hofmann AA, Kane KR, Tkach TK et al (1995) Treatment of infected total knee arthroplasty using an articulating spacer. Clin Orthop 321:45 13. Hofmann AA, Goldberg T, Tanner A et al (2005) Treatment of infected total knee arthroplasty using an articulating spacer: 2- to 12-year experience. Clin Orthop 430:125 14. Fehring TK, Odum S, Calton TF et al (2000) Articulating versus static spacers in revision total knee arthroplasty for sepsis. Clin Orthop 380:9 15. Wroblewski BM (1986) One-stage revision of infected cemented total hip arthroplasty. Clin Orthop Relat Res 211:103–107 16. Callaghan JJ, Katz RP, Johnston RC (1999) One-stage revision surgery of the infected hip. A minimum 10-year followup study. Clin Orthop Relat Res 369:139–143 17. Rudelli S, Uip D, Honda E, Lima AL (2008) One-stage revision of infected total hip arthroplasty with bone graft. J Arthroplasty 23(8):1165–1177 18. Kotwal R, Prasad N, Morgan-Jones R (2009) One stage revision for infected total knee arthroplasty. BASK meeting Edinburgh 19. Parkinson RW, Kay PR, Rawal A (2011) A case for one-stage revision in infected total knee arthroplasty? Knee 18(1):1–4 20. Bauer T, Piriou P, Lhotellier L, Leclerc P, Mamoudy P, LortatJacob A (2006) Results of reimplantation for infected total knee arthroplasty: 107 cases. Rev Chir Orthop Reparatrice Appar Mot 92(7):692–700 21. Parvizi J, Gehrke T, Chen AF (2013) Proceedings of the international consensus on peri-prosthetic joint infection. Bone Joint J 95-B:1450–1452 22. Costerton JW (1999) Bacterial biofilms: a common cause of persistent infections. Science 284:1318–1322
1. Rand JA, Bryan RS, Morrey BF, Westholm F (1986) Management of infected total knee arthroplasty. Clin Orthop Relat Res 205:75–85
123
ORIGINAL ARTICLE
Two-stage revision for infected total knee arthroplasty: our experience with interval prosthesis N. Prasad • V. Paringe • R. Kotwal A. Ghandour • R. Morgan Jones
•
Received: 23 August 2013 / Accepted: 22 November 2013 / Published online: 5 December 2013 Ó Springer-Verlag France 2013
Abstract Introduction To review our practice of performing twostage revision for infected total knee arthroplasty using articulating interval prosthesis and to compare the incidence of the recurrence of infection and re-operation rate in patients undergoing two-stage revision as planned with the group of patients who choose not to proceed to the second stage. Method This study is a retrospective review of 60 consecutive patients undergoing a two-stage revision for infected total knee arthroplasty using articulating interval prosthesis. All cases managed by a single surgeon using a uniform peri-operative protocol, and short-course parenteral antibiotic therapy. Result Thirty-four patients (57 %) (Group 1) underwent the two-stage revision as planned. However, twenty-six patients (43 %) (Group 2) opted not to have a second-stage procedure as the first-stage and interval prosthesis had eradicated the infection, resolved the pain and achieved good functional outcome. There were five cases of recurrent infection in the 60 patients (8 %) at a mean follow-up 5 years. In those completing the two-stage revision, two patients had recurrent infection. Of the patients who retained the interval prosthesis, there were three recurrent infections. There was no statistically significant difference between the groups in terms of recurrence of infection or re-revision.
N. Prasad V. Paringe R. Kotwal A. Ghandour R. M. Jones Cardiff and Vale NHS Trust, Cardiff, UK N. Prasad (&) 5 Lovage Close, Pontprennau, Cardiff CF23 8SB, UK e-mail: [email protected]
Conclusion Two-stage revision with interval prostheses represents a safe and reliable method of treating infected knee prosthesis; however, there may be a role for one-stage revision in selected cases. Keywords Knee revision Infection Two-stage revision Interval prosthesis
Introduction Management of deep infection following joint replacement surgery in orthopaedics remains challenging. The incidence of infection following primary total knee replacement is between 1 and 2 % with an increased rate in rheumatoid patients [1–3]. This leads to significant morbidity to the patient. Two-stage revision surgery proposed by Insall [4] is considered as the ‘gold standard’ for the treatment of infected total knee replacement and can be performed by different techniques, using static or articulating spacers during the first stage. The interim use of an antibioticimpregnated static cement spacer has been reported to successfully eradicate infection in more than 90 % of infected total knee arthroplasty (TKA) [5–8] Two-stage revision using an articulating spacer have also shown good clinical outcome with successful eradication of infection [9, 10]. The senior author has been performing two-stage revision procedures for infected total knee replacement using interval prosthesis as articulating spacer, and over years it has come to his attention that good number of patients do not proceed for second stage as the infection is cured following first stage, and they get a reasonable function with articulating spacer. The aim of our study was (1) to review
123
1280
our practice of performing two-stage revision for infected total knee arthroplasty using an articulating interval prosthesis and short-course parenteral antibiotic therapy by a single surgeon in a tertiary orthopaedic unit, (2) to compare the incidence of the recurrence of infection in patients undergoing two-stage revision as planned with the group of patients who choose not to proceed to the second stage and (3) to find the incidence of re-operation rate for all causes in both the groups.
Patients and methods We identified a consecutive series of 70 patients operated between 2001 and 2008 for infected total knee replacement in our hospital. We carried out a clinical and radiological review on these patient’s case notes. The inclusion criteria for the study were patients with chronically infected primary total knee replacement diagnosed clinically and radiologically with or without proven bacteriology from previous knee aspirate or arthroscopic knee biopsy. We also included patients who had exchange of liner for acute infection following primary total knee replacement subsequently presented as chronic infection. We excluded patients who has already had two-stage revision surgery for infection elsewhere and was referred to our unit, as there was further recurrence of infection. We also excluded patients who had revision surgery for aseptic loosening and subsequently referred to our unit as they became infected. A total 60 patients were considered for the study. There were 32 males and 28 female patients. The mean age at the time of first-stage surgery was 66 years (45–83 years). Four patients were dead at the time of final review, so the last available data were used for analysis. We divided the patients into two groups; patients who had planned two-stage revision (Group 1) and the second group was made up of patients who declined second stage and decided to live on with the interval prosthesis (Group 2). We looked at the subsequent recurrence of infection and also re-revision surgery for infection or any other cause. Statistical analysis was performed using v2 test to compare the recurrence of infection and re-revision surgery between the groups.
Eur J Orthop Surg Traumatol (2014) 24:1279–1283
Multiple tissue samples (minimum of 5) are sent for microbiological analysis. Sequential and repeated intramedullary reaming, lavage and curettage were used to achieve satisfactory debridement prior to insertion of the interval prosthesis. We consider the debridement at this stage as the most important step as this facilitates removal of the biofilm. Primary condylar femoral components, both posterior stabilised (PS) and cruciate retaining (CR) and the rotating platform polyethylene tibial insert without the metal tray were used to construct the interval prosthesis. The rotating platform polyethylene insert is chosen as the ‘keel’ and this allowed greater stability. Occasionally, in cases of significant bone loss, additional support for the interval prosthesis was required using a femoral stem and tibial tray and stem (all components from DePuy/J&J). In all cases, maximal ligament balance was achieved using a combination of femoral and tibial components and the cement mantle. The components were inserted using antibiotic-loaded cement (Palacos R; Zimmer, UK). A double mix of Palacos cement with gentamicin plus 1 gm of vancomycin per mix was used in all cases. Cement was not maximally pressurised and used late in the cure phase, to facilitate easy removal during the second-stage procedure. The cement mantle filled all ‘dead space’ between implants and bone interface (Fig. 1a, b). Patients were given intravenous teicoplanin for 5 days and then converted to dual oral antibiotics, depending upon the culture and sensitivities after discussing with the microbiologist, for 6 weeks. If no organisms were identified from the multiple intraoperative tissue samples, oral antibiotics were given based on previous positive results from the hospital of referral. Once established on oral antibiotics, the patients were discharged home. Patients were allowed to fully weight bear, with unrestricted range of movement, following first-stage surgery. No knee bracing was used post-operatively. The patients were monitored clinically and biochemically with inflammatory markers, CRP and ESR checked regularly. The second stage is planned at 6 months whenever possible. A 6-month interval is preferred to allow the soft tissue envelope to reach a quiescent, healed state and maximise mobility and function of the knee and patient. A prolonged interval also allows greater confidence that infection has been eradicated.
Operative technique The surgical technique followed was standardised and identical for all the patients. The same senior surgeon performed all surgeries. A tourniquet was used in all the cases. During the first stage, a thorough debridement is performed after removal of loose infected components and bone cement. Lautenbach compartmental debridement techniques [11] were used to debride the femur and tibia.
123
Results In our series of 60 patients, 34 patients (57 %) had a planned second-stage procedure a mean 6 months following the first stage. The remaining 26 patients (43 %) opted not to have the second-stage procedure, as they were infection free, functionally satisfied and pain free with the
Eur J Orthop Surg Traumatol (2014) 24:1279–1283
1281
Fig. 1 Post-operative radiographs showing the interval prosthesis in situ (AP and lateral). Note the antibioticloaded cement mantle fills all the dead space between prosthesis and bone interface
Table 1 List of organisms isolated from biopsy sample during firststage revision knee arthroplasty Organisms
Numbers
Coagulase-negative Staph aureus
29
Staph aureus
7
MRSA
5
Alpha-haemolytic streptococci
4
Serratia Coliforms
2 3
Beta-haemolytic streptococci
2
Pseudomonas
2
Gram-negative cocci
2
Gram-positive Bacillus
1
Enterococci
1
Candida
1
Microaerophilic streptococci
1
No growth
6
interval prosthesis. The predominant organism in our series was coagulase-negative Staph aureus (Table 1). Of the 34 patients who had two-stage procedure, there were 18 males and 16 females. The mean age at the time of first stage was 69 years (Table 2). The mean duration between first-stage interval prosthesis and definitive second-stage procedure was 26 weeks (range 12–48 weeks). The mean follow-up after second-stage procedure was 60 months (range 24–96 months). The range of motion was more than 90° in 30 out of 34 patients. The mean flexion
Table 2 Patient demographics with follow-up details in both the groups Type of surgical procedure
Male: female
Age mean (range) (years)
Follow-up mean (range) (months)
Group 1
Two stage (n = 34)
18:16
69 (53–81)
60 (24–96)
Group 2
First stage only (n = 26)
18:8
63 (55–82)
48 (24–64)
deformity at the time of final follow was 4° (0–10), and the mean flexion was 98° (70–115). All of the knees were clinically stable. There was recurrence of infection in two patients requiring further surgery. In the group of patients who declined second stage (n = 26), 4 patients had deceased at the time of this study, without recurrent infection. The mean age at the time of surgery was 61 years. The mean follow-up was 48 months (range 24–64 months; Table 2). The shorter follow-up reflecting patients later in the series when confidence in retaining the interval prosthesis were greater. The range of motion was more than 90° in 19 out of 21 patients. The mean flexion deformity was 5° (range 0–10), and the mean flexion was 95° (range 80–115). In this group, five patients required re-operation. Two patients had revision surgery at 14 and 18 months due to instability, and three patients had late recurrence of infection which required revision surgery at 17, 22 and 26 months following insertion of the interval prosthesis.
123
1282
Eur J Orthop Surg Traumatol (2014) 24:1279–1283
Table 3 Recurrence infection and re-revision rate in two groups
Recurrence of infection number (%)
Further revision procedure for infection number (%)
Further revision surgery for reasons other than infection number (%)
Total re-revision procedure number (%)
2
2
0
2
First stage only (n = 26)
3
3
2
5
p value (Chi test)
0.43
0.43
0.07
0.11
Two stage (n = 34)
Of the total of five recurrent infections, four underwent a further revision with eradication of infection. The last patient underwent an arthrodesis due to a compromised soft tissue envelope, but remains infected. The statistical analysis did not show any difference in the incidence of recurrence of infection or re-revision between the two groups. There was no re-revision in the two-stage group for non-infective cause, and on comparison the one-stage group had two re-revision for instability; however, statistical analysis did not show any significant difference although the p value was very close to 0.05 (p = 0.07; Table 3).
Discussion A two-stage revision total knee arthroplasty using a static spacer can lead to problems when it comes to re-implantation in the second stage in the form of stiffness and loss of bone stock [14] This led to the evolution of use of articulating spacers, and its usage has been reported [9, 10, 12, 13]. Hofmann et al. [12] described technique in which the femoral component is removed, autoclaved and reinserted along with a new polyethylene tibia. In his series of 55 patients, there was 12 % recurrence of infection at 73 months [13]. They allow only partial weight-bearing mobilisation after stage 1. Two-stage revision using a PROSTALAC prosthesis as an articulating interval prosthesis has shown an overall recurrence rate of 9 % in a series of 45 patients with a mean follow-up of 48 months [9]. PROSTALAC spacers are expensive and are discarded after 6 weeks. In our series, the overall rate of recurrent infection following revision and insertion of an interval prosthesis is 8 % (five recurrences, two in the two-stage group and three in the retained interval prosthesis group). The advantages of using interval prosthesis as an articulating spacer are many. The interval prosthesis provides stability and balance of the ligaments and soft tissue envelope, allowing free movement and full weight bearing. Improved patient mobility allows for a planned second
123
stage at a time when the patient has physiologically recovered from surgery and the limb has become quiescent with an improved soft tissue envelope. This gives the surgeon and the patient flexibility in the timing of the second stage. The subsequent revision surgery is technically easier as mobility is improved and fibrotic scarring less. We also note that the space-filling cement mantle provides a ‘macro-lock’ into the bone interface, thus providing bone-cement stability and preventing progressive bone loss. The cement mantle plus choosing an appropriately sized femoral and tibial implant allows a balanced construct with appropriately tensioned collateral ligaments. This avoids progressive soft tissue damage resulting from knee instability and allows the soft tissue envelope to recover following eradication of infection and debridement. In our series, 43 % of patient did not require the second stage, supporting the possible role for one-stage revision of infected total knee replacement. In our series, there was no difference in the incidence of recurrence of infection between two-stage group and one-stage group. We had two cases of re-revision in one-stage (interval prosthesis) group for instability, and this is expected as the interval prosthesis was used as temporary step before definitive surgery. A one-stage revision using a definitive prosthesis could theoretically avoid this. One-stage revision is well accepted in total hip replacement [15–17]. The advantages of performing a one-stage revision are clinical (single operation, shorter hospitalisation, reduced medication), functional (improved movement, shorter rehabilitation, reduced disability) and significant economic savings. We have started performing one-stage revision in selected cases drawing experience from our two-stage revision study. The preliminary result of one-stage revision knee arthroplasty was presented in the British Association for Surgery of the Knee Spring Meeting [18]. Recent literature search yields encouraging results for one-stage revision knee arthroplasty for infection [19, 20] although more research is required. The recently concluded international consensus on peri-prosthetic joint infection supports our views on single-stage excision arthroplasty for infection. There was
Eur J Orthop Surg Traumatol (2014) 24:1279–1283
strong agreement that a single-stage revision arthroplasty for infection may be performed in circumstances where sensitive antibiotics where available and patient is not in sepsis. Lack of identification of organism, presence of sinus tract and severe soft tissue damage requiring flap coverage are considered as contraindication for single-stage revision arthroplasty in peri-prosthetic joint infection [21]. The infecting microorganism secretes a matrix of extracellular polymeric substance, and this along with the bacteria forms what is known as ‘biofilms’, which are inherently resistant to antimicrobials, and forms the root cause for chronic infection[22]. The initial debridement allows the surgeons to remove these ‘biofilms’. We believe that initial debridement performed during the removal of infected prosthesis is the most important step, and it is surgeons choice that whether you perform this in two stage or one stage. We acknowledge some of the limitations of our study. Firstly, this is a retrospective non-randomised study. Secondly, 43 % of patients opted not to have the second stage, and this has been purely patient’s choice. This in fact reiterates the argument that interval prosthesis provides good function and subsequently would help to achieve good results after a two-stage procedure. Thirdly, we did not include the results of functional scoring at final follow-up, as our study aim was to look at the recurrence rate of infection.
Conclusion Two-stage revision with an interval prosthesis using an articulating spacer gives low recurrence of infection with good knee range of motion and allows the patient to mobilise full weight bearing during and after the surgery. We would also recommend that such cases should be referred to surgeons with appropriate training and expertise to minimise recurrence and avoid the poor outcomes associated with multiple revisions. The senior author suggests that recurrent infected knee revisions should be referred to a regional centre of expertise to give patients the ‘best chance’ of successful eradication of infection and return of function. Onestage revision in selected case may be way forward; however, further research is needed on this. Conflict of interest
None.
References
1283 2. Bengston S, Knutson K, Lidgren L (1989) Treatment of infected knee arthroplasty. Clin Orthop Relat Res 245:173–178 3. Blom AW, Brown J, Taylor AH, Pattison G, Whitehouse S, Bannister GC (2004) Infection after total knee arthroplasty. J Bone Joint Surg Br 86(B):688–691 4. Insall JN, Thompson FM, Brause BD (1983) Two-stage reimplantation for the salvage of infected total knee arthroplasty. J Bone Joint Surg Am 65(8):1087–1098 5. Wilde AH, Ruth JT (1988) Two-stage reimplantation in infected total knee arthroplasty. Clin Orthop 236:23 6. Goldman RT, Scuderi GR, Insall JN (1996) Two-Stage reimplantation for infected total knee replacement. Clin Orthop 331:118 7. Simmons TD, Stern SH (1996) Diagnosis and management of the infected total knee arthroplasty. Am J Knee Surg 2:99 8. Windsor RE, Insall JN, Urs WK et al (1990) Two-stage reimplantation for the salvage of total knee arthroplasty complicated by infection. Further follow-up and refinement of indications. J Bone Joint Surg Am 72:272 9. Haddad FS, Masri BA, Campbell D, McGraw RW, Beauchamp CP, Duncan CP (2000) The PROSTALAC functional spacer in two-stage revision for infected knee replacements. Prosthesis of antibiotic-loaded acrylic cement. J Bone Joint Surg Br 82: 807–812 10. Trezies A, Parish E, Dixon P, Cross M (2006) The use of an articulating spacer in the management of infected total knee arthroplasties. J Arthroplasty 21(5):702–704 11. Weber FA, Lautenbach EEG (1986) Revision of infected total hip arthroplasty. Clin Orthop 211:108–115 12. Hofmann AA, Kane KR, Tkach TK et al (1995) Treatment of infected total knee arthroplasty using an articulating spacer. Clin Orthop 321:45 13. Hofmann AA, Goldberg T, Tanner A et al (2005) Treatment of infected total knee arthroplasty using an articulating spacer: 2- to 12-year experience. Clin Orthop 430:125 14. Fehring TK, Odum S, Calton TF et al (2000) Articulating versus static spacers in revision total knee arthroplasty for sepsis. Clin Orthop 380:9 15. Wroblewski BM (1986) One-stage revision of infected cemented total hip arthroplasty. Clin Orthop Relat Res 211:103–107 16. Callaghan JJ, Katz RP, Johnston RC (1999) One-stage revision surgery of the infected hip. A minimum 10-year followup study. Clin Orthop Relat Res 369:139–143 17. Rudelli S, Uip D, Honda E, Lima AL (2008) One-stage revision of infected total hip arthroplasty with bone graft. J Arthroplasty 23(8):1165–1177 18. Kotwal R, Prasad N, Morgan-Jones R (2009) One stage revision for infected total knee arthroplasty. BASK meeting Edinburgh 19. Parkinson RW, Kay PR, Rawal A (2011) A case for one-stage revision in infected total knee arthroplasty? Knee 18(1):1–4 20. Bauer T, Piriou P, Lhotellier L, Leclerc P, Mamoudy P, LortatJacob A (2006) Results of reimplantation for infected total knee arthroplasty: 107 cases. Rev Chir Orthop Reparatrice Appar Mot 92(7):692–700 21. Parvizi J, Gehrke T, Chen AF (2013) Proceedings of the international consensus on peri-prosthetic joint infection. Bone Joint J 95-B:1450–1452 22. Costerton JW (1999) Bacterial biofilms: a common cause of persistent infections. Science 284:1318–1322
1. Rand JA, Bryan RS, Morrey BF, Westholm F (1986) Management of infected total knee arthroplasty. Clin Orthop Relat Res 205:75–85
123
