Secondary Logo

A Technique for Intraoperative Construction of Antibiotic Spacers

Ha, Chul-Won

Clinical Orthopaedics and Related Research: April 2006 - Volume 445 - Issue - p 204-209
doi: 10.1097/01.blo.0000201161.52196.c5

A technique for intraoperatively creating an antibiotic spacer for two-stage treatment of infected total knee replacements is described. An intraoperative mold is made from the removed components and used to create antibiotic spacers with surface contours similar to those of the original total knee replacement. The spacers restore leg length and knee stability. This allows limited function during the interval before reimplantation of the new total knee replacement. It is a cost-effective and convenient technique for creating a suitably shaped and sized cement spacer for two-stage revision total knee replacement after infection. The clinical results of 12 consecutive patients using this technique with minimum of 2 years followup seem to be at least equal or better than results reported in previous studies.

Level of Evidence: Prognostic study, Level II (retrospective study). See the Guidelines for Authors for a complete description of levels of evidence.

From the Department of Orthopedic Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, KangNam Gu, IlWon Dong, Seoul, Korea.

Received: December 25, 2004

Revised: July 15, 2005; October 17, 2005

Accepted: November 7, 2005

The author certifies that he has no commercial associations (eg, consultancies, stock ownership, equity interest, patent/licensing arrangements, etc.) that might pose a conflict of interest in connection with the submitted article. The author certifies that his institution has approved or waived approval for the human protocol for this investigation and that all investigations were conducted in conformity with ethical principles of research.

Correspondence to: Chul-Won Ha, MD, Department of Orthopedic Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, KangNam Gu, IlWon Dong 50, Seoul 135-710, Korea. Phone: 82-2-3410-0275; Fax: 82-2-3410-0061; E-mail:

Infection remains one of the most devastating complications after total knee arthroplasty (TKA). Two-stage reimplantation using antibiotic cement spacers seems the mainstay in treating infected total knee replacements (TKRs).1,3,5,8,11

Numerous authors have reported intraoperative construction of articulating antibiotic cement spacers.5,6,9 Hofmann et al9 removed and resterilized the implants, then reimplanted them using antibiotic bone cement primarily for the prosthesis-bone interface. The relatively small amount of antibiotic-impregnated cement and the persistence of a large volume of the once-infected metal during the period of infection control seems to be a concern. Fehring et al5 reported intraoperative construction of articulating cement spacers and the clinical results of their technique. A premanufactured metallic mold was needed to intraoperatively construct articulating cement spacers by their technique. Goldstein et al6 reported a more convenient technique by creating articulating cement spacers intraoperatively by manual molding of doughy bone cement using the polyethylene insert trial as a partial mold. Their technique enables intraoperative construction of articulating cement spacers that have a similar contour of the TKR. However, manual molding of the cement spacers sometimes may be unsuccessful in reproducing the surface contour of the TKR, and depending on the surgeon's skill, may require more time and effort. If the shape of the manually molded spacer is distorted while balancing the flexion and extension gap, it sometimes may be troublesome to reshape the spacer and balance the gap again. This tends to predispose an unbalanced or less mobile knee.

Building on these approaches, I developed a technique for intraoperative construction of an antibiotic spacer for treating infected TKRs with surface contours similar to the original TKR. I I assessed whether the technique works as well as the previously reported technique by assessing range of motion (ROM) of the knee before and after insertion of the articulating spacer, improvement of knee function, any complications related to the technique, and recurrence of infection.

Back to Top | Article Outline


I retrospectively reviewed 12 consecutive patients (two men and 10 women) who had two-stage reimplantation TKAs with the described technique. All the chronic infected TKRs during the study period were treated with the same technique. Acutely infected TKRs that could be treated by débridement with retention of the implants were excluded. The average age of the patients was 65.7 years (range, 54-73 years). All patients were followed up for a minimum of 2 years (range, 2-3.5 years).

The knee was exposed along the previous surgical scar. The implants were removed, and thorough débridement of the infected tissue and previously used bone cement was done. The removed femoral component and polyethylene insert were washed and sterilized by autoclave. A custom mold was manufactured intraoperatively from the removed components using bone cement. The articular surface of the removed component was coated with lubricant (sterile mineral oil), and bone cement was applied to a moderately doughy articular surface. After the cement set, the mold was used to create a custom component with surface contours similar the original component (Fig 1). The cement spacer for the tibial side was constructed first. A bolus of bone cement was put on the upper end of tibia in the late doughy phase. The previously manufactured mold for the tibial side was coated with lubricant and applied to the upper end of the cement bolus with firm pressure. The excess cement that protruded outward beyond the margin of proximal tibia was removed to avoid soft tissue irritation postoperatively, especially around the patellar tendon and medial collateral ligament (Fig 2). The appropriate thickness of the cement spacer on the tibial side was created by controlling the pressure placed on the mold. The thickness of the tibial spacer was roughly reproduced considering the removed implant's thickness, which was between 10 to 15 mm in every case. The extension gap was stabilized by adjusting the distal thickness of the femoral spacer. The thickness of the removed tibial component and the position of femoral epicondyles, tip of the fibula, and patellar height were considered. The underlying contour of the cement spacer constructed by this technique fit well with the irregular contour of the underlying tibial bone. This allowed good stability of the cement spacer. Bonding of the cement spacer and the underlying bone was prevented by taking the cement construct out of the tibia a few times before complete setting of the bone cement. The articular surface contour of the spacer sometimes may be deformed during this procedure. The surface contour can be reshaped easily by applying the mold.

Fig 1A

Fig 1A

Fig 2A

Fig 2A

A similar process was used on the femoral side. The posterior condylar part of femoral cement spacer also was molded manually before complete hardening of the cement. The posterior offset of the femoral cement spacer was reduced (Fig 3) by manual molding to make the flexion gap loose, which facilitated postoperative flexion (Fig 4). Before the femoral cement set, at the late doughy phase, the knee was gently extended to check the extension gap. While doing this, the femoral spacer's shape often was deformed, which easily was restored by reapplying the custom mold, resulting in proper thickness and shape. A stable extension gap was established and resulted in good walking function during the interval before reimplantation of the new implants.

Fig 3A

Fig 3A

Fig 4A

Fig 4A

Continuous passive motion (CPM) was used. The patients were instructed to just maintain or mildly increase bending of the knee (usually up to 90°) and to do their usual daily activities, such as short-distance walking. An extension knee brace for walking was recommended. Followup laboratory tests included leukocyte count, erythrocyte sedimentation rate, and C-reactive protein.

Reimplantation of the new implant was performed at 6 to 8 weeks after the first-stage operation if the laboratory results were within or near the normal range without any clinical evidence of sustained infection for more than 2 weeks. Previously inserted articulating cement spacers were removed and thorough débridement was done. This was followed by reimplantation of a new implant. The patients were discharged from the hospital 1 to 2 weeks postoperatively with appropriate oral antibiotics if available. If susceptible oral antibiotics were not available, the patients stayed in hospital and received intravenous antibiotics for 4 to 8 weeks postoperatively, which had been the usual practice. The practice in the author's institution changed recently, but 1 week of antibiotics after reimplantation of an infected implant is still used. All patients were seen at an outpatient clinic every 3 months until 1 year postoperatively, and then yearly thereafter.

The following information was taken from the patient records: range of motion before the first-stage operation, before the second-stage reimplantation, and at the last postoperative followup; time between the first-stage and second-stage operations, Knee Society knee score and function score, complications related to the technique, recurrence of infection, infecting organisms, surgical time of the reimplantation, and additional operative procedures needed for adequate exposure for the reimplantation. The ROM and Knee Society scores were measured in a blinded manner by a research assistant responsible for collecting clinical data before and after joint replacement surgery. The infecting organisms were confirmed by preoperative or intraoperative microbiologic culture studies. Recurrence of infection was evaluated by radiologic and laboratory studies and careful physical examination at every postoperative visit.

These clinical parameters were compared with parameters in previous reports for treatment of infected TKRs using articulating cement spacers.

Back to Top | Article Outline


The ROM of the knee was maintained or increased in all patients during the interval before reimplantation. Knee flexion was 79° (range, 40°-130°) before the first-stage operation and 85° (range, 40°-130°) before the second-stage operation. Knee flexion improved to 104° (range, 70°-140°) 3 months after the reimplantation and 102° (range, 75°-140°) by final followup. Flexion contracture was 8° (range, 0°-30°) before the first-stage operation,

5.4° (range, 0°-20°) before reimplantation, and 2.1° (0°-15°) at final followup.

The function of the knee was improved. The Knee Society knee score improved from 30 preoperatively to 87 at final followup, and the Knee Society function score improved from 24 to 80. Some patients who were noncompliant were able to walk quite well without wearing a brace, but there were no associated problems.

There were no complications related to this technique. Complications related to surgical exposure, such as patellar tendon rupture or tibial tubercle avulsion, were not seen during the reimplantation of TKRs. No unexpected bone loss or collateral ligament problems occurred during the second-stage operations. One patient had patellar tendon avulsion 6 months after reimplantation caused by forceful buckling of the knee after slipping and falling. It does not seem to be related to the described technique because there was partial bone loss from a previous infection at the patellar tendon insertion. Although bone grafting was done during the reimplantation surgery, it seemed to predispose the tendon to the avulsion. The patient was treated successfully by repair and extensor mechanism augmentation with an Achilles allograft.

No patients have experienced recurrent infections, although infecting organisms that are hard to eradicate were identified in some cases (Table 1). Four patients had a Methicillin-resistant bacterial infection and one patient had a fungal infection.



The second-stage operation seemed to be facilitated with the technique. The average time between the first-stage and second-stage operations was 9 weeks (range, 4-16 weeks). The operation time (skin-to-skin) for the reimplantation was 132 minutes (range, 80-150 minutes). The additional procedure for an adequate exposure in the second-stage operation was the rectus snip procedure in eight of 12 patients. No additional procedures were needed in the remaining four patients.

Back to Top | Article Outline


To overcome the disadvantages of previously reported techniques, I developed a technique for intraoperative construction of articulating cement spacers that is cost effective and convenient. The clinical results of the treatment of infected TKRs using the technique were assessed to determine the efficacy of the new technique.

The major limitations of this study are that this is a retrospective study with a limited number of patients and a relatively short-term followup (minimum, 2 years). Although this study included only 12 patients, they were consecutive and treated successfully with the same techniques. The retrospective review of this limited number of patients seems to be sufficient to evaluate the new technique. The clinical results seem to be at least equal or better than results in previous studies. There have few reports describing intraoperative construction of articulating cement spacers, and some do not include clinical data.6,11 Fehring et al5 reported a 7% reinfection rate after using articulating cement spacers and ROM averaging 105°. Hofmann et al9 reported 92% good to excellent results and ROM after reimplantation of 5° to 106°. None of the current patients had recurrence of infection, and their ROM was 2° to 120° at final followup.

Nonarticulating, static cement spacers often result in quadriceps and soft tissue contractures, which may cause loss of knee motion and difficulties in surgical exposure at the second-stage reimplantation surgery. They also may increase the risk of complications like patellar tendon rupture or tibial tubercle avulsion during the reimplantation surgery. The need for extended procedures such as V-Y quadricepsplasty, patella turn-down procedure, or muscle flaps4,5,12 may be increased. Other possible disadvantages include bone loss from migration of the spacer, periarticular osteoporosis, loss of medial and lateral gutters, damage to the quadriceps tendon adhering to the anterior surface of the distal femur, and collateral ligament problems associated with difficult exposure.2,5,7,9,10

Articulating antibiotic cement spacers have been developed to overcome these possible problems related to the static spacers. Two-stage reimplantation TKAs using articulating cement spacers had a better clinical result with less surgical difficulty in the second-stage operation and less complications.5

Premanufactured functional spacers like the PROSTALAC® (DePuy Orthopaedics, Inc, Warsaw, IN) were introduced to overcome such problems.8 The application, however, still has some limitations probably related to limited versatility because of size of the temporary implants and antibiotics that are impregnated.

The technique I have described seems to overcome or minimize the disadvantages of intraoperative construction of articulating antibiotic cement spacers.5,6,9 Compared with the technique used by Hofmann et al,9 my technique does not need to put any metallic or polyethylene implant back in the once-infected joint, and a larger amount of antibiotic-impregnated cement can be put in, which seems to be advantageous for increased local antibiotic concentration and long-term release of antibiotics. The technique of Fehring et al5 is convenient for constructing a suitably shaped articulating spacer, but premanufactured metallic molds should be created preoperatively. With my technique, a custom mold can be created conveniently and intraoperatively using the removed implants. The technique described by Goldstein et al6 could be successful in intraoperative construction of articulating cement spacers that have a similar contour of the TKR. However, the manual molding technique sometimes may not be successful in reproducing the surface contour of the TKR depending on the surgeon's skill, and more time and effort may be needed to get well-shaped spacers than when using a premanufactured mold. Also, they used an aluminum foil covering for the underlying bone during the manual molding, which seems to hinder the bone cement from fitting to the irregularly contoured underlying bone.6 With my technique, a custom mold is manufactured intraoperatively (Fig 1), and construction of the mold by curing the cement can be done at the back table by an assistant while the operator is performing thorough débridement after removal of the infected prostheses. The actual additional time to cure the cement was not significant. Once constructed, the mold was used to create the articulating spacer that has surface contours similar to the TKR (Fig 2). A stem-like intramedullary extension (Fig 3) and good fit of the cement spacer with the underlying bone increased the stability of the spacer, which seemed to reduce pain during early postoperative mobilization of the knee.

The cement particles generated from this type of articulating cement spacer may be a concern. Although, from a literature review, there is no evidence that abrasion between the cement surfaces produces deleterious particle debris, decreased activity was recommended to avoid generation of particle debris. Thorough débridement was repeated during reimplantation surgery to minimize the remaining particle debris.

The choice of antibiotics to be impregnated into the articulating spacer is crucial for successful treatment. Tobramycin or vancomycin was used in most patients in this series according to the culture results. The doses used were 4.8 g tobramycin or 4 g vancomycin to each 40-g batch of cement. Mycostatin (BMS, New York, NY) was used for the patient with a fungal infection.

This technique creates an antibiotic spacer used for treating infected TKRs with surface contours similar to the original TKR. It allows the gap to be filled with antibiotic-loaded cement and restores length and stability to the knee, allowing limited function during the interval before reim-plantation of the new TKR. The technique is cost effective and convenient in creating a suitably sized and shaped cement spacer for two-stage revision TKR for infection.

Back to Top | Article Outline


1. Buechel F, Femino F, D'Alessio J. Primary exchange revision arthroplasty for infected total knee replacement: a long-term study. Am J Orthop. 2004;33:190-198.
2. Calton TF, Fehring TK, Griffin WL. Bone loss associated with the use of spacer blocks in infected total knee arthroplasty. Clin Orthop Relat Res. 1997;345:148-154.
3. Dixon P, Parish EN, Cross MJ. Arthroscopic debridement in the treatment of the infected total knee replacement. J Bone Joint Surg. 2004;86:39-42.
4. Emerson RH Jr, Muncie M, Tarbox TR, Higgins LL. Comparison of a static with a mobile spacer in total knee infection. Clin Orthop Relat Res. 2002;404:132-138.
5. Fehring TK, Odum S, Calton TF, Mason JB. Articulating versus static spacers in revision total knee arthroplasty for sepsis: the Ranawat Award. Clin Orthop Relat Res. 2000;380:9-16.
6. Goldstein WM, Kopplin M, Wall R, Berland K. Temporary articulating methylmethacrylate antibiotic spacer (TAMMAS): a new method of intraoperative manufacturing of a custom articulating spacer. J Bone Joint Surg. 2001;83(Suppl 2 Pt 2):92-97.
7. Gusso MI, Capone A, Civinini R, Scoccianti G. The spacer block technique in revision of total knee arthroplasty with septic loosening. Chir Organi Mov. 1995;80:21-27.
8. 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. 2000;82:807-812.
9. Hofmann AA, Kane KR, Tkach TK, Plaster RL, Camargo MP. Treatment of infected total knee arthroplasty using an articulating spacer. Clin Orthop Relat Res. 1995;321:45-54.
10. McMaster WC. Technique for intraoperative construction of PMMA spacer in total knee revision. Am J Orthop. 1995;24:178-180.
11. McPherson EJ, Lewonowski K, Dorr LD. Techniques in arthroplasty: use of an articulated PMMA spacer in the infected total knee arthroplasty. J Arthroplasty. 1995;10:87-89.
12. McPherson EJ, Patzakis MJ, Gross JE, Holtom PD, Song M, Dorr LD. Infected total knee arthroplasty: two-stage reimplantation with a gastrocnemius rotational flap. Clin Orthop Relat Res. 1997;341: 73-81.
© 2006 Lippincott Williams & Wilkins, Inc.