- Email: [email protected]
Clinical effectiveness of antibiotic-impregnated cement spacers for the treatment of infected implants of the hip joint
J Orthop Sci (2003) 8:823–828 DOI 10.1007/s00776-003-0722-y
Clinical effectiveness of antibiotic-impregnated cement spacers for the treatment of infected implants of the hip joint Kengo Yamamoto, Naoki Miyagawa, Toshinori Masaoka, Yoichi Katori, Takaaki Shishido, and Atsuhiro Imakiire Department of Orthopedic Surgery, Tokyo Medical University, 6-7-1 Nishi-shinjuku, Shinjuku-ku, Tokyo 160-0023, Japan
Abstract It is difficult to treat an infected implant of the hip joints, as it requires long-term treatment and eventually may lead to amputation or arthrodesis, involving immeasurable physical and psychological suffering for the patient. We utilized antibiotic-impregnated cement spacers for 17 infections after total hip arthroplasty and bipolar arthroplasty with good clinical results. We thoroughly removed any foreign material and formed an antibiotic-impregnated cement spacer into a shape similar to that of the implants. This enabled highconcentration antibiotics to act on the infected sites. It also can prevent leg-length discrepancy and atrophy of bones or muscles. Although cement spacers have been reported to have problems regarding shape and strength, we achieved good results with cement spacer molds in the present study. All revision surgeries were performed using a two-stage procedure. No infection has recurred at a mean follow-up of 3 years 2 months. Key words Total hip arthroplasty · Infected implants · Two-stage procedure · Cement spacer
Introduction Infection after total hip arthroplasty (THA) is one of the most severe complications in orthopedics. It not only burdens the patient with physical and mental hardships, it can pose a financial burden on both the patient and the institution.10,16 Through the introduction of aseptic operating rooms, the use of antibiotics as a preventive measure, and better understanding of the contraindications for THA, the infection rate after THA has been reduced to less than 1% at most institutions that specialize in joint replacement surgery.18 However, the absolute number of infected cases has not
Offprint requests to: K. Yamamoto Received: June 23, 2003 / Accepted: August 25, 2003
decreased owing to the increasing number of THAs performed every year. Eradication of the pathogens must be confirmed to prevent recurrent infections. Previously, closed continuous irrigation was commonly performed, whereas today the use of antibioticimpregnated bone cement has gained acceptance for treating infections. Antibiotic-impregnated cement beads are a common choice, but the use of an antibioticimpregnated cement spacer shaped as an artificial femoral prosthesis has been reported in the literature to give good clinical results.5 Another point in the debate on the treatment of infections is whether to perform a single-stage1,14 or two-stage3,5 procedure. We report the clinical results of antibiotic-impregnated cement spacers using a twostage procedure for treating infection after THA.
Patients and methods Between January 1998 and December 2002 a total of 17 hips in 17 patients (6 men, 11 women) with a total hip or bipolar prosthesis became infected. Consequently, these patients were treated with an antibioticimpregnated cement spacer at our institution. The mean age at the time of diagnosis of prosthetic joint infection at our hospital was 61.8 years (range 44–76 years). The diagnoses for the initial operations were femoral neck fracture (8 cases), avascular necrosis of the femoral head (3 cases), osteoarthritis (3 cases), fracture-dislocations of the hip (2 cases), and rheumatoid arthritis (1 case). Of the 17 patients, 10 had complications that, potentially, could have influenced the surgical site infection (Table 1). Of the 17 primary arthroplasties, 5 were performed at our institution and the other 12 elsewhere. The pathogens were coagulase-negative staphylococcus (CNS) (4 cases), Staphylococcus epidermidis (4 cases), methicillin-resistant Staphylococcus aureus (MRSA) (3 cases), Pseudomonas aeruginosa (1 case),
44 75 76 44 63 49 70 63 72 62 62 54 66 65 70 65 75
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
M M F M M F F F M F M F F F F F F
Sex
DF FNF FNF DF FNF ANF OA RA FNF FNF FNF FNF FNF ANF ANF OA OA
Diagnosis Diabetes mellitus Healthy Cardiac disease Healthy Renal insufficiency Healthy Cardiac disease RA Healthy Diabetes mellitus Cardiac disease Healthy Healthy Healthy Cardiac disease Diabetes mellitus Diabetes mellitus
Medical status Revision THA BHA BHA Æ revision THA Revision THA BHA Æ revision THA Revision THA THA THA BHA BHA BHA BHA BHA THA THA THA THA
Type of implant 35 36 46 57 47 23 48 45 58 8 73 63 37 18 32 7 9
Duration of infection (from onset to spacer) (days) 158 133 146 245 168 278 86 95 108 120 95 85 96 105 93 91 114
Time on spacer (days) S. epidermidis MRSA CNS MRSA MRSA CNS None detected None detected P. aeruginosa CNS CNS None detected S. epidermidis E. coli None detected S. epidermidis S. epidermidis
Type of bacteria 41 66 42 89 61 44 100 120 58 87 29 37 80 89 26 103 107
ESR (mm)
1.4 4.7 4.3 6.5 1.4 1.9 25.0 6.7 5.7 12.5 1.3 1.8 1.8 3.9 1.1 14.9 6.1
CRP (mg/dl)
ANF, avascular necrosis of the femoral head; OA, osteoarthritis of the hip; FNF, femoral neck fracture; DF, dislocation fracture of the hip; RA, rheumatoid arthritis; S. epidermidis, Staphylococcus epidermidis; MRSA, methicillin-resistant Staphylococcus aureus; CNS, coagulase-negative staphylococcus; P. aeruginosa, Pseudomonas aeruginosa; E. coli, Escherichia coli; THA, total hip arthroplasty; BHA, bipolar hip arthroplasty; ESR, erythrocyte sedimentation rate; CRP, C-reactive protein
Age
Case
Table 1. Characteristics of treated patients
824 K. Yamamoto et al.: Cement spacer for infected implants
K. Yamamoto et al.: Cement spacer for infected implants
a
825
b
Escherichia coli (1 case), and unknown (4 cases). The period of time from the primary prosthesis implantation to the date of diagnosis of the infection averaged 2 years 4 months (range 12 days to 9 years 1 month). The mean period between the initial diagnosis of the infection and removal of the implant was 37.8 days (range 7–73 days) (Table 1). According to Paprosky’s bone loss evaluation,12 three had type 1, four had type 2a, and three had type 2b on the acetabular side; and seven had type 1 and four had type 2a on the femoral side. After prosthesis removal, complete capsulectomy and débridement were performed to remove infected tissues and residual cement. The capsule, granulation tissues from both the acetabulum and the femoral canal, necrotized bone tissues, bone cement, and screws were cultured to ensure that all pathogens were recognized. After performing careful, aggressive irrigation with saline, we implanted an antibiotic-impregnated polymethylmethacrylate (PMMA) spacer in the femoral canal. Stable reduction was obtained in all cases. From January 1998 to April 2001, we made handmolded cement spacers based on the method of Ivarson et al.5 by separately duplicating the shape of the retrieved femoral head and the femoral stem component with bone cement (Surgical Simplex-P; Stryker, Allendale, NJ, USA) (group A). Since May 2001, Cement Spacer Molds (Biomet, Warsaw, IN, USA) (Fig. 1) have been available to our institution, and nine were utilized to mold the bone cement (group B). For both methods, two bent 2.0 mm diameter Kirschner wires were used as the core of the spacer (Fig. 2). The purpose of the Kirschner wires was to facilitate removal of
Fig. 1. a Cement spacer molds. b Antibiotic-impregnated cement spacer molded by cement spacer molds
Fig. 2. Radiography of the inserted cement spacer
cement in one piece in case the bone cement is broken. The 14 non-MRSA-infected patients were treated with 1 g gentamicin sulfate, and the 3 MRSA-infected patients were treated with a combination of 1 g gen-
826
tamicin sulfate and 2 g vancomycin in 80 g of PMMA. Cement spacer molds are made from medical grade silicon, and three sizes are available (head size/diameter of the stem/length are 43/9/125 mm, 51/9/125 mm, and 57/13/145 mm, respectively). Cement spacer molds were used to make cement spacers in the following fashion. The nozzle of an Exter Cement Gun (Stnyker) loaded with cement was inserted into the filling port of the cement spacer mold, and the mold was filled with cement while it still had low viscosity. After the cement in the cement spacer mold completely hardened, the mold was peeled away from the cement spacer. Active and passive exercises and muscle training were started 2 days after operation to increase the range of motion. A wheelchair was allowed 1 week after operation. Walking with crutches with partial weightbearing (⬍20 kg) started during the second week after operation. The inflammation markers in the blood [leukocyte count, erythrocyte sedimentation rate (ESR), Creactive protein (CRP)] were examined regularly after the surgery.15 Systemic intravenous antibiotics were given until the markers were normalized for three consecutive weeks or more and the weekly joint fluid cultures were negative for three consecutive cultures or more.4 All infections were eradicated. Fifteen patients underwent revision surgery, and two underwent Girdlestone resection arthroplasty because of their health conditions. Before each revision surgery, joint fluid obtained from the patient was cultured and confirmed to contain no pathogens. During revision surgery, tissue samples from multiple sites on the acetabulum and the femoral canal were also cultured. Clinical evaluations were made on the basis of the Harris hip score. An anteroposterior radiograph of the pelvis was used for radiographic examination. Statistical analysis was performed using a Mann-Whitney U-test and t-test. A risk rate of less than 5% was regarded as indicating a significant difference.
Results Pain in 12 of 17 patients disappeared after the infected prosthesis was removed and the spacer implanted. The mean interval between the first and second surgeries in these two-stage procedures was 129.9 days (range 91–278 days). There was one fracture of the spacer in group A (2 months after implantation), but there were no fractures in group B. There was one subluxation (dislocation) of the spacer in group A (1 month after the implantation), and a closed reduction was performed; there were no dislocations in group B (Table 2).
K. Yamamoto et al.: Cement spacer for infected implants
Immediately before the reimplantation, the mean flexion angles for groups A and B were 60.0° ⫾ 7.6° and 75.6° ⫾ 8.5°, respectively. The mean abduction angles for groups A and B were 16.3° ⫾ 5.2° and 21.1° ⫾ 6.0°, respectively. Consequently, our analysis indicated that there was a statistically significant difference in the flexion angle between the two groups (P ⬍ 0.005), but there was no statistically significant difference in the abduction angle between the two groups (P ⫽ 0.1102). The patients have been followed up, on average, for 38 months (range 14–62 months). There has been no recurrent infection in any of the patients. At the final clinical evaluation, the Harris hip scores of the two patients who had had a Girdlestone resection arthroplasty were low (43–60, mean 51.5), whereas the Harris hip scores of 14 patients who had had a revision THA were comparable to those measured prior to the onset of infection (70–94, mean 89). The radiographs at the final evaluation showed no radiolucent lines, focal osteolysis, or migration of the components.
Discussion It is debatable whether a single-stage or two-stage exchange procedure should be chosen to treat infection after hip arthroplasty. Garvin and Hanssen3 reported that the mean success rates for a single-stage and two-stage procedure using antibiotic-impregnated cement were 82% and 91%, respectively. Jackson and Schmalzreid also reported that the indication for singlestage revision is limited.6 Therefore, the two-stage procedure is considered to be more effective for preventing recurrent infection. However, there are several disadvantages to the two-stage procedure that must be considered; the patients must undergo two or more operations, and their activity level is reduced for a longer period of time. For controlling infection, it is reasonable to combine systemic intravenous antibiotics with local administration. Continuous irrigation has become less popular as a method of local administration owing to the difficulties associated with its management and limitation of the patient’s freedom. Instead, antibiotic-impregnated cement beads,11 spacers,5 and prostheses coated with antibiotic-impregnated cement2 have become the local administration methods of choice. Recently the effectiveness of antibiotic-impregnated cement spacers that release highly concentrated antibiotics at the site has been reported. Regarding the recurrent infection rate with the cement spacer, Koo et al. reported infection in only 1 of 22 patients7 and Magnan et al. in only 2 of 10 patients.9 Unfortunately there are some issues with the spacer: The antibiotics chosen must be heat-resistant, and the
Dislocation
Fracture of spacer
62 60 57 59 55 53 49 47 46 24 5 21 20 18 17 15 14 Mallory-Head long stem Mallory-Head long stem Mallory-Head long stem Mallory-Head long stem Not reimplanted Mallory-Head calcar replacement Mallory-Head long stem Mallory-Head long stem Mallory-Head calcar replacement S-ROM Mallory-Head long stem Mallory-Head long stem Mallory-Head calcar replacement Mallory-Head long stem Not reimplanted Mallory-Head long stem Mallory-Head calcar replacement 6 4 12 4 6 3 4 3 5 5 2 3 4 4 7 4 3 10 10 15 15 15 20 20 25 15 15 15 20 20 20 25 30 30 50 55 55 60 60 60 66 75 60 70 70 75 75 80 80 80 90 0.9 0.8 0.5 0.7 ⬍0.3 0.5 1.2 1.4 1.1 ⬍0.3 0.6 ⬍0.3 ⬍0.3 1.1 0.7 ⬍0.3 ⬍0.3 21 24 12 5 7 24 12 28 15 9 15 11 7 21 23 22 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Postoperative antibiotic treatment (months) Abduction (°) Flexion (°) CRP (mg/dl) ESR (mm) Case
Table 2. Infection treatment after implantation of the spacer
Prosthesis at reimplantation
Postoperative follow-up (months)
Complications and notes (fracture or dislocation)
K. Yamamoto et al.: Cement spacer for infected implants
827
releasing period of the antibiotics depends on the type of antibiotic used. However, the spacer maintains the appropriate space after the prosthesis is removed, maintaining the proper tension of the muscle around the site until the second surgery can be performed. This in turn enables earlier rehabilitation, prevents atrophy and muscle weakening by walking with partial weightbearing, and avoids shortening of the muscles of the affected leg. All these factors combine to reduce the difficulty of the revision THA. Another area to analyze is the amount of antibiotics secreted into the joint space. As the total surface area of the antibiotic-impregnated cement spacer is smaller than that of the antibiotic-impregnated cement beads, there may be less antibiotic released. However, Park et al.13 and Tonegawa et al.17 evaluated the disparity of the release rates and release patterns between gentamicin sulfate (GM) beads and a GM spacer; they reported that the GM spacer had a higher release rate despite the smaller surface area, and after the first 24 h there was no clear difference in the amount of release from the two cements. They concluded that the GM spacer is as effective as GM beads. Another issue to consider with the cement spacer is its mechanical strength. There have been some reports of spacers fracturing with full weight-bearing during rehabilitation.8 A screw or plate was then inserted in the spacer to strengthen it,8 and recently new, stronger spacers have been reported. Magnan et al.9 used a Spacer-G (Tecres S.p.A.; Sommacampagna, Verona, Italy), which has a central load-bearing hollow cylindrical rod made of stainless steel with a maximum diameter of 10 mm. This cylinder was then entirely covered with antibiotic-impregnated bone cement. They reported that 8 of 10 patients could walk with crutches for partial weight-bearing. The cement spacer made by cement spacer molds is considered to be not as strong as Spacer-G. However, all our patients treated with cement spacer molds were permitted to walk using crutches and partial weight-bearing. No cement spacers made using cement spacer molds have experienced fracture of the cement. Cement spacer molds facilitate production of an appropriately shaped femoral head, which benefits the hip joint with smooth articulation. As a result, the flexion angle of group B in our study averaged 75.6°, which was more than that of group A. Unfortunately, there has been a report of excessive migration of the femoral spacer due to the length of the neck. The main drawback of the cement spacer molds is the limited size variation. Although three head sizes are currently available (43, 51, and 57 mm), only the 43-mm head size is used for most Japanese patients. A wider range of head sizes would help prevent dislocation and migration of the spacers and obtain smoother hip move-
828
ment. Moreover, to insert the cement spacer without damaging the femoral bone, it is important to select a cement spacer mold with an appropriate stem diameter, but currently only those of 9 and 13 mm are available. We believe that continued consideration and study are necessary to find the appropriate sizes and methods essential to ensure successful implantation of an antibiotic-loaded spacer for two-stage treatment of infection after THA. Acknowledgments. The authors are indebted to Professor J. Patrick Barron and Assistant Professor Raoul Breugelmans of the International Medical Communications Center of Tokyo Medical University for their review of this manuscript.
References 1. Callaghan JJ, Katz RP, Johnston RC. One-stage revision surgery of the infected hip: a minimum 10-year follow-up study. Clin Orthop 1999;369:139–43. 2. Duncan CP, Masri BA. The role of antibiotic-loaded cement in the treatment of an infection after a hip replacement. J Bone Joint Surg Am 1994;76:1742–51. 3. Garvin KL, Hanssen AD. Infection after total hip arthroplasty: past, present, and future. J Bone Joint Surg Am 1995;77:1576– 88. 4. Hirose S, Sawai K, Hashimoto S, et al. Revision surgery for infected hip prosthesis: our concepts on antibiotic-loaded PMMA beads implantation. J Joint Surg 2002;21:349–55 (in Japanese). 5. Ivarsson I, Wahlstrom O, Djerf K, et al. Revision of infected hip replacement: two-stage procedure with a temporary gentamicin spacer. Acta Orthop Scand 1994;65:7–8. 6. Jackson WO, Schmalzried TP. Limited role of direct exchange arthroplasty in the treatment of infected total hip replacements. Clin Orthop 2000;381:101–5.
K. Yamamoto et al.: Cement spacer for infected implants 7. Koo K-H, Yang J-W, Cho S-H, et al. Impregnation of vancomycin, gentamicin and cefotaxime in a cement spacer for two-stage cementless reconstruction in infected total hip arthroplasty. J Arthroplasty 2001;16:882–92. 8. Leunig M, Chosa E, Speck M, et al. A cement spacer for two-stage revision of infected implants of the hip joint. Int Orthop 1998; 22:209–14. 9. Magnan B, Regis D, Biscaglia R, et al. Preformed acrylic bone cement spacer loaded with antibiotics: use of two-stage procedure in 10 patients because of infected hips after total replacement. Acta Orthop Scand 2001;72:591–4. 10. Masterson EL, Masri BA, Duncan CP. Treatment of infection at the site of total hip replacement. J Bone Joint Surg Am 1997;79: 1740–9. 11. Nelson CL, Evans RP, Blaha JD, et al. A comparison of gentamicin-impregnated polymethylmethacrylate bead implantation to conventional pareteral antibiotic therapy in infected total hip and knee arthroplasty. Clin Orthop 1993;295:96–101. 12. Paprosky WG, Brandford MS, Younger TI. Classificazione delle insufficienze di tessuto osseo nella protesi mobilizzata. Chir Organi Mov 1994;79:285–91 (in Italian). 13. Park J-S, Higuchi F, Mashima T, et al.Treatment of infected hip replacement with antibiotics cement spacer. Nihon Kotsu Kansetsu Kansensho Kenkyukai Zasshi 1996;10:39–41 ( in Japanese). 14. Raut VV, Siney PD, Wroblewski BM. One-stage revision of total hip arthroplasty for deep infection: long-term followup. Clin Orthop 1995;321:202–7. 15. Sanzen L, Carlsson AS. The diagnostic value of C-reactive protein in infected total hip arthroplasties. J Bone Joint Surg Br 1989; 71:638–41. 16. Sculco TP. The economic impact of infected total joint arthroplasty. American Academy of Orthopaedic Surgeons 1993;42: 349–51. 17. Tonegawa M, Hashimoto A, Sekiguchi M, et al. Release of gentamicin from acrylic bone cement spacer. J Musculoskeletal Syst 1998;11:1019–24 (in Japanese). 18. Wymenga AB, van Horn JR, Theeuwes A, et al. Perioperative factors associated with septic arthritis after arthroplasty: prospective multicenter study of 362 knee and 2651 hip operations. Acta Orthop Scand 1992;6:665–71.
Clinical effectiveness of antibiotic-impregnated cement spacers for the treatment of infected implants of the hip joint Kengo Yamamoto, Naoki Miyagawa, Toshinori Masaoka, Yoichi Katori, Takaaki Shishido, and Atsuhiro Imakiire Department of Orthopedic Surgery, Tokyo Medical University, 6-7-1 Nishi-shinjuku, Shinjuku-ku, Tokyo 160-0023, Japan
Abstract It is difficult to treat an infected implant of the hip joints, as it requires long-term treatment and eventually may lead to amputation or arthrodesis, involving immeasurable physical and psychological suffering for the patient. We utilized antibiotic-impregnated cement spacers for 17 infections after total hip arthroplasty and bipolar arthroplasty with good clinical results. We thoroughly removed any foreign material and formed an antibiotic-impregnated cement spacer into a shape similar to that of the implants. This enabled highconcentration antibiotics to act on the infected sites. It also can prevent leg-length discrepancy and atrophy of bones or muscles. Although cement spacers have been reported to have problems regarding shape and strength, we achieved good results with cement spacer molds in the present study. All revision surgeries were performed using a two-stage procedure. No infection has recurred at a mean follow-up of 3 years 2 months. Key words Total hip arthroplasty · Infected implants · Two-stage procedure · Cement spacer
Introduction Infection after total hip arthroplasty (THA) is one of the most severe complications in orthopedics. It not only burdens the patient with physical and mental hardships, it can pose a financial burden on both the patient and the institution.10,16 Through the introduction of aseptic operating rooms, the use of antibiotics as a preventive measure, and better understanding of the contraindications for THA, the infection rate after THA has been reduced to less than 1% at most institutions that specialize in joint replacement surgery.18 However, the absolute number of infected cases has not
Offprint requests to: K. Yamamoto Received: June 23, 2003 / Accepted: August 25, 2003
decreased owing to the increasing number of THAs performed every year. Eradication of the pathogens must be confirmed to prevent recurrent infections. Previously, closed continuous irrigation was commonly performed, whereas today the use of antibioticimpregnated bone cement has gained acceptance for treating infections. Antibiotic-impregnated cement beads are a common choice, but the use of an antibioticimpregnated cement spacer shaped as an artificial femoral prosthesis has been reported in the literature to give good clinical results.5 Another point in the debate on the treatment of infections is whether to perform a single-stage1,14 or two-stage3,5 procedure. We report the clinical results of antibiotic-impregnated cement spacers using a twostage procedure for treating infection after THA.
Patients and methods Between January 1998 and December 2002 a total of 17 hips in 17 patients (6 men, 11 women) with a total hip or bipolar prosthesis became infected. Consequently, these patients were treated with an antibioticimpregnated cement spacer at our institution. The mean age at the time of diagnosis of prosthetic joint infection at our hospital was 61.8 years (range 44–76 years). The diagnoses for the initial operations were femoral neck fracture (8 cases), avascular necrosis of the femoral head (3 cases), osteoarthritis (3 cases), fracture-dislocations of the hip (2 cases), and rheumatoid arthritis (1 case). Of the 17 patients, 10 had complications that, potentially, could have influenced the surgical site infection (Table 1). Of the 17 primary arthroplasties, 5 were performed at our institution and the other 12 elsewhere. The pathogens were coagulase-negative staphylococcus (CNS) (4 cases), Staphylococcus epidermidis (4 cases), methicillin-resistant Staphylococcus aureus (MRSA) (3 cases), Pseudomonas aeruginosa (1 case),
44 75 76 44 63 49 70 63 72 62 62 54 66 65 70 65 75
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
M M F M M F F F M F M F F F F F F
Sex
DF FNF FNF DF FNF ANF OA RA FNF FNF FNF FNF FNF ANF ANF OA OA
Diagnosis Diabetes mellitus Healthy Cardiac disease Healthy Renal insufficiency Healthy Cardiac disease RA Healthy Diabetes mellitus Cardiac disease Healthy Healthy Healthy Cardiac disease Diabetes mellitus Diabetes mellitus
Medical status Revision THA BHA BHA Æ revision THA Revision THA BHA Æ revision THA Revision THA THA THA BHA BHA BHA BHA BHA THA THA THA THA
Type of implant 35 36 46 57 47 23 48 45 58 8 73 63 37 18 32 7 9
Duration of infection (from onset to spacer) (days) 158 133 146 245 168 278 86 95 108 120 95 85 96 105 93 91 114
Time on spacer (days) S. epidermidis MRSA CNS MRSA MRSA CNS None detected None detected P. aeruginosa CNS CNS None detected S. epidermidis E. coli None detected S. epidermidis S. epidermidis
Type of bacteria 41 66 42 89 61 44 100 120 58 87 29 37 80 89 26 103 107
ESR (mm)
1.4 4.7 4.3 6.5 1.4 1.9 25.0 6.7 5.7 12.5 1.3 1.8 1.8 3.9 1.1 14.9 6.1
CRP (mg/dl)
ANF, avascular necrosis of the femoral head; OA, osteoarthritis of the hip; FNF, femoral neck fracture; DF, dislocation fracture of the hip; RA, rheumatoid arthritis; S. epidermidis, Staphylococcus epidermidis; MRSA, methicillin-resistant Staphylococcus aureus; CNS, coagulase-negative staphylococcus; P. aeruginosa, Pseudomonas aeruginosa; E. coli, Escherichia coli; THA, total hip arthroplasty; BHA, bipolar hip arthroplasty; ESR, erythrocyte sedimentation rate; CRP, C-reactive protein
Age
Case
Table 1. Characteristics of treated patients
824 K. Yamamoto et al.: Cement spacer for infected implants
K. Yamamoto et al.: Cement spacer for infected implants
a
825
b
Escherichia coli (1 case), and unknown (4 cases). The period of time from the primary prosthesis implantation to the date of diagnosis of the infection averaged 2 years 4 months (range 12 days to 9 years 1 month). The mean period between the initial diagnosis of the infection and removal of the implant was 37.8 days (range 7–73 days) (Table 1). According to Paprosky’s bone loss evaluation,12 three had type 1, four had type 2a, and three had type 2b on the acetabular side; and seven had type 1 and four had type 2a on the femoral side. After prosthesis removal, complete capsulectomy and débridement were performed to remove infected tissues and residual cement. The capsule, granulation tissues from both the acetabulum and the femoral canal, necrotized bone tissues, bone cement, and screws were cultured to ensure that all pathogens were recognized. After performing careful, aggressive irrigation with saline, we implanted an antibiotic-impregnated polymethylmethacrylate (PMMA) spacer in the femoral canal. Stable reduction was obtained in all cases. From January 1998 to April 2001, we made handmolded cement spacers based on the method of Ivarson et al.5 by separately duplicating the shape of the retrieved femoral head and the femoral stem component with bone cement (Surgical Simplex-P; Stryker, Allendale, NJ, USA) (group A). Since May 2001, Cement Spacer Molds (Biomet, Warsaw, IN, USA) (Fig. 1) have been available to our institution, and nine were utilized to mold the bone cement (group B). For both methods, two bent 2.0 mm diameter Kirschner wires were used as the core of the spacer (Fig. 2). The purpose of the Kirschner wires was to facilitate removal of
Fig. 1. a Cement spacer molds. b Antibiotic-impregnated cement spacer molded by cement spacer molds
Fig. 2. Radiography of the inserted cement spacer
cement in one piece in case the bone cement is broken. The 14 non-MRSA-infected patients were treated with 1 g gentamicin sulfate, and the 3 MRSA-infected patients were treated with a combination of 1 g gen-
826
tamicin sulfate and 2 g vancomycin in 80 g of PMMA. Cement spacer molds are made from medical grade silicon, and three sizes are available (head size/diameter of the stem/length are 43/9/125 mm, 51/9/125 mm, and 57/13/145 mm, respectively). Cement spacer molds were used to make cement spacers in the following fashion. The nozzle of an Exter Cement Gun (Stnyker) loaded with cement was inserted into the filling port of the cement spacer mold, and the mold was filled with cement while it still had low viscosity. After the cement in the cement spacer mold completely hardened, the mold was peeled away from the cement spacer. Active and passive exercises and muscle training were started 2 days after operation to increase the range of motion. A wheelchair was allowed 1 week after operation. Walking with crutches with partial weightbearing (⬍20 kg) started during the second week after operation. The inflammation markers in the blood [leukocyte count, erythrocyte sedimentation rate (ESR), Creactive protein (CRP)] were examined regularly after the surgery.15 Systemic intravenous antibiotics were given until the markers were normalized for three consecutive weeks or more and the weekly joint fluid cultures were negative for three consecutive cultures or more.4 All infections were eradicated. Fifteen patients underwent revision surgery, and two underwent Girdlestone resection arthroplasty because of their health conditions. Before each revision surgery, joint fluid obtained from the patient was cultured and confirmed to contain no pathogens. During revision surgery, tissue samples from multiple sites on the acetabulum and the femoral canal were also cultured. Clinical evaluations were made on the basis of the Harris hip score. An anteroposterior radiograph of the pelvis was used for radiographic examination. Statistical analysis was performed using a Mann-Whitney U-test and t-test. A risk rate of less than 5% was regarded as indicating a significant difference.
Results Pain in 12 of 17 patients disappeared after the infected prosthesis was removed and the spacer implanted. The mean interval between the first and second surgeries in these two-stage procedures was 129.9 days (range 91–278 days). There was one fracture of the spacer in group A (2 months after implantation), but there were no fractures in group B. There was one subluxation (dislocation) of the spacer in group A (1 month after the implantation), and a closed reduction was performed; there were no dislocations in group B (Table 2).
K. Yamamoto et al.: Cement spacer for infected implants
Immediately before the reimplantation, the mean flexion angles for groups A and B were 60.0° ⫾ 7.6° and 75.6° ⫾ 8.5°, respectively. The mean abduction angles for groups A and B were 16.3° ⫾ 5.2° and 21.1° ⫾ 6.0°, respectively. Consequently, our analysis indicated that there was a statistically significant difference in the flexion angle between the two groups (P ⬍ 0.005), but there was no statistically significant difference in the abduction angle between the two groups (P ⫽ 0.1102). The patients have been followed up, on average, for 38 months (range 14–62 months). There has been no recurrent infection in any of the patients. At the final clinical evaluation, the Harris hip scores of the two patients who had had a Girdlestone resection arthroplasty were low (43–60, mean 51.5), whereas the Harris hip scores of 14 patients who had had a revision THA were comparable to those measured prior to the onset of infection (70–94, mean 89). The radiographs at the final evaluation showed no radiolucent lines, focal osteolysis, or migration of the components.
Discussion It is debatable whether a single-stage or two-stage exchange procedure should be chosen to treat infection after hip arthroplasty. Garvin and Hanssen3 reported that the mean success rates for a single-stage and two-stage procedure using antibiotic-impregnated cement were 82% and 91%, respectively. Jackson and Schmalzreid also reported that the indication for singlestage revision is limited.6 Therefore, the two-stage procedure is considered to be more effective for preventing recurrent infection. However, there are several disadvantages to the two-stage procedure that must be considered; the patients must undergo two or more operations, and their activity level is reduced for a longer period of time. For controlling infection, it is reasonable to combine systemic intravenous antibiotics with local administration. Continuous irrigation has become less popular as a method of local administration owing to the difficulties associated with its management and limitation of the patient’s freedom. Instead, antibiotic-impregnated cement beads,11 spacers,5 and prostheses coated with antibiotic-impregnated cement2 have become the local administration methods of choice. Recently the effectiveness of antibiotic-impregnated cement spacers that release highly concentrated antibiotics at the site has been reported. Regarding the recurrent infection rate with the cement spacer, Koo et al. reported infection in only 1 of 22 patients7 and Magnan et al. in only 2 of 10 patients.9 Unfortunately there are some issues with the spacer: The antibiotics chosen must be heat-resistant, and the
Dislocation
Fracture of spacer
62 60 57 59 55 53 49 47 46 24 5 21 20 18 17 15 14 Mallory-Head long stem Mallory-Head long stem Mallory-Head long stem Mallory-Head long stem Not reimplanted Mallory-Head calcar replacement Mallory-Head long stem Mallory-Head long stem Mallory-Head calcar replacement S-ROM Mallory-Head long stem Mallory-Head long stem Mallory-Head calcar replacement Mallory-Head long stem Not reimplanted Mallory-Head long stem Mallory-Head calcar replacement 6 4 12 4 6 3 4 3 5 5 2 3 4 4 7 4 3 10 10 15 15 15 20 20 25 15 15 15 20 20 20 25 30 30 50 55 55 60 60 60 66 75 60 70 70 75 75 80 80 80 90 0.9 0.8 0.5 0.7 ⬍0.3 0.5 1.2 1.4 1.1 ⬍0.3 0.6 ⬍0.3 ⬍0.3 1.1 0.7 ⬍0.3 ⬍0.3 21 24 12 5 7 24 12 28 15 9 15 11 7 21 23 22 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Postoperative antibiotic treatment (months) Abduction (°) Flexion (°) CRP (mg/dl) ESR (mm) Case
Table 2. Infection treatment after implantation of the spacer
Prosthesis at reimplantation
Postoperative follow-up (months)
Complications and notes (fracture or dislocation)
K. Yamamoto et al.: Cement spacer for infected implants
827
releasing period of the antibiotics depends on the type of antibiotic used. However, the spacer maintains the appropriate space after the prosthesis is removed, maintaining the proper tension of the muscle around the site until the second surgery can be performed. This in turn enables earlier rehabilitation, prevents atrophy and muscle weakening by walking with partial weightbearing, and avoids shortening of the muscles of the affected leg. All these factors combine to reduce the difficulty of the revision THA. Another area to analyze is the amount of antibiotics secreted into the joint space. As the total surface area of the antibiotic-impregnated cement spacer is smaller than that of the antibiotic-impregnated cement beads, there may be less antibiotic released. However, Park et al.13 and Tonegawa et al.17 evaluated the disparity of the release rates and release patterns between gentamicin sulfate (GM) beads and a GM spacer; they reported that the GM spacer had a higher release rate despite the smaller surface area, and after the first 24 h there was no clear difference in the amount of release from the two cements. They concluded that the GM spacer is as effective as GM beads. Another issue to consider with the cement spacer is its mechanical strength. There have been some reports of spacers fracturing with full weight-bearing during rehabilitation.8 A screw or plate was then inserted in the spacer to strengthen it,8 and recently new, stronger spacers have been reported. Magnan et al.9 used a Spacer-G (Tecres S.p.A.; Sommacampagna, Verona, Italy), which has a central load-bearing hollow cylindrical rod made of stainless steel with a maximum diameter of 10 mm. This cylinder was then entirely covered with antibiotic-impregnated bone cement. They reported that 8 of 10 patients could walk with crutches for partial weight-bearing. The cement spacer made by cement spacer molds is considered to be not as strong as Spacer-G. However, all our patients treated with cement spacer molds were permitted to walk using crutches and partial weight-bearing. No cement spacers made using cement spacer molds have experienced fracture of the cement. Cement spacer molds facilitate production of an appropriately shaped femoral head, which benefits the hip joint with smooth articulation. As a result, the flexion angle of group B in our study averaged 75.6°, which was more than that of group A. Unfortunately, there has been a report of excessive migration of the femoral spacer due to the length of the neck. The main drawback of the cement spacer molds is the limited size variation. Although three head sizes are currently available (43, 51, and 57 mm), only the 43-mm head size is used for most Japanese patients. A wider range of head sizes would help prevent dislocation and migration of the spacers and obtain smoother hip move-
828
ment. Moreover, to insert the cement spacer without damaging the femoral bone, it is important to select a cement spacer mold with an appropriate stem diameter, but currently only those of 9 and 13 mm are available. We believe that continued consideration and study are necessary to find the appropriate sizes and methods essential to ensure successful implantation of an antibiotic-loaded spacer for two-stage treatment of infection after THA. Acknowledgments. The authors are indebted to Professor J. Patrick Barron and Assistant Professor Raoul Breugelmans of the International Medical Communications Center of Tokyo Medical University for their review of this manuscript.
References 1. Callaghan JJ, Katz RP, Johnston RC. One-stage revision surgery of the infected hip: a minimum 10-year follow-up study. Clin Orthop 1999;369:139–43. 2. Duncan CP, Masri BA. The role of antibiotic-loaded cement in the treatment of an infection after a hip replacement. J Bone Joint Surg Am 1994;76:1742–51. 3. Garvin KL, Hanssen AD. Infection after total hip arthroplasty: past, present, and future. J Bone Joint Surg Am 1995;77:1576– 88. 4. Hirose S, Sawai K, Hashimoto S, et al. Revision surgery for infected hip prosthesis: our concepts on antibiotic-loaded PMMA beads implantation. J Joint Surg 2002;21:349–55 (in Japanese). 5. Ivarsson I, Wahlstrom O, Djerf K, et al. Revision of infected hip replacement: two-stage procedure with a temporary gentamicin spacer. Acta Orthop Scand 1994;65:7–8. 6. Jackson WO, Schmalzried TP. Limited role of direct exchange arthroplasty in the treatment of infected total hip replacements. Clin Orthop 2000;381:101–5.
K. Yamamoto et al.: Cement spacer for infected implants 7. Koo K-H, Yang J-W, Cho S-H, et al. Impregnation of vancomycin, gentamicin and cefotaxime in a cement spacer for two-stage cementless reconstruction in infected total hip arthroplasty. J Arthroplasty 2001;16:882–92. 8. Leunig M, Chosa E, Speck M, et al. A cement spacer for two-stage revision of infected implants of the hip joint. Int Orthop 1998; 22:209–14. 9. Magnan B, Regis D, Biscaglia R, et al. Preformed acrylic bone cement spacer loaded with antibiotics: use of two-stage procedure in 10 patients because of infected hips after total replacement. Acta Orthop Scand 2001;72:591–4. 10. Masterson EL, Masri BA, Duncan CP. Treatment of infection at the site of total hip replacement. J Bone Joint Surg Am 1997;79: 1740–9. 11. Nelson CL, Evans RP, Blaha JD, et al. A comparison of gentamicin-impregnated polymethylmethacrylate bead implantation to conventional pareteral antibiotic therapy in infected total hip and knee arthroplasty. Clin Orthop 1993;295:96–101. 12. Paprosky WG, Brandford MS, Younger TI. Classificazione delle insufficienze di tessuto osseo nella protesi mobilizzata. Chir Organi Mov 1994;79:285–91 (in Italian). 13. Park J-S, Higuchi F, Mashima T, et al.Treatment of infected hip replacement with antibiotics cement spacer. Nihon Kotsu Kansetsu Kansensho Kenkyukai Zasshi 1996;10:39–41 ( in Japanese). 14. Raut VV, Siney PD, Wroblewski BM. One-stage revision of total hip arthroplasty for deep infection: long-term followup. Clin Orthop 1995;321:202–7. 15. Sanzen L, Carlsson AS. The diagnostic value of C-reactive protein in infected total hip arthroplasties. J Bone Joint Surg Br 1989; 71:638–41. 16. Sculco TP. The economic impact of infected total joint arthroplasty. American Academy of Orthopaedic Surgeons 1993;42: 349–51. 17. Tonegawa M, Hashimoto A, Sekiguchi M, et al. Release of gentamicin from acrylic bone cement spacer. J Musculoskeletal Syst 1998;11:1019–24 (in Japanese). 18. Wymenga AB, van Horn JR, Theeuwes A, et al. Perioperative factors associated with septic arthritis after arthroplasty: prospective multicenter study of 362 knee and 2651 hip operations. Acta Orthop Scand 1992;6:665–71.